Field of the invention

The present invention relates to a method for obtaining a spatial pattern of an anatomical structure of a subject. The invention further relates to a system configured to carry out such method.

Background of the invention

Biometric and morphological analysis of biological structures are topics of great interest in biomechanical research and in a variety of clinical fields. The possibility of analysing the morphology and to extract biometric data, such as geometric and dimensional data, relating to biological structures of the human or animal body first appeared with the introduction of medical imaging techniques, which allowed to reveal internal structures hidden by the skin or by other organs.

The extraction of biometric data plays an important role for example in the fields of traumatology and orthopaedic, in appliances for preventing, detecting and monitoring postural disorders or bone deformities, or also in the morphological analysis of a subject’s biological structure, for example for research aims.

In recent decades especially spinal and postural functional assessment has progressively gained importance for clinical practice as a result of both financial and clinical issues. Epidemiological studies recognize that postural defects frequently involve concurrent diseases that may become manifest years after the onset of the first symptoms, and may further degenerate into more serious pathologic conditions even in the absence of previous trauma or genetic predisposition. For these reasons, it would be extremely useful an early detection of postural defects even when no advanced symptoms are observed. Postural diseases are also frequently caused by stability issues due to unbalance of the neuromuscular tone. Especially in this case, a continuous monitoring of results is important for a positive outcome of related treatments.

Morphologic and biometric analysis of biological structures is conventionally carried out on radiographic or tomographic images. However, these imaging techniques do not constitute an optimal choice for screening or routine monitoring exams, because frequent exposure to ionising radiations might be harmful for the patients.

Related art

Methods and systems for obtaining spatial patterns of biological structures that do not involve ionising radiations are known in the art.

In particular, among different alternatives, stereophotogrammetry is emerging as a robust non-contacting measurement technique that allows to estimate three-dimensional coordinates of points on an object/subject based on measurements made in a series of photographic images taken from different angles. EP 3 225 155 A1 discloses a system and method for detecting and monitoring postural diseases in a patient, based on photographic images of the patient carrying position reflectors in predefined positions.

The system comprises a computer, a digital image capturing device (such as a camera or a webcam) and a coaxial illumination system, position reflectors (also indicated as luminous tags) applied on the patient, a suitable software for elaboration and data acquisition, and a platform.

The method comprises the steps of positioning luminous tags at predefined locations of the body of the patient; connecting via a USB port the digital image capturing device (camera or webcam) to the computer; acquiring images or photograms; elaborating the images through the software; printing and/or copying data on a hardware support.

The position reflectors consist of simple reflecting bands attached, at predefined locations, onto the skin of the patient by means of an adhesive.

The document further describes that the software allows, through an automatic search of the position of the reflectors within the image of the patient, to reproduce in 3D the spinal curve of the patient and to analyse it for detecting possible defects. The software comprises an algorithm for video elaboration, for position search in the image, for fitting a diagram and for storing the data analysed.

US 2003/181830 A1 discloses a system and method for automatically analyzing the location of anatomical markers placed over skeletal landmarks to obtain biomechanical parameters, using these biomechanical data to automatically detect and/or quantify postural deviations from correct anatomical alignment, and further generating corrective exercise routines.

According to this document, position data for each of the marker positions is obtained by acquiring photographic images of the patient carrying the markers, positioned against a backdrop. The markers have high contrast with respect to the backdrop. The backdrop includes a plurality of scale and orientation reference marker points, so that the system is dimensionally calibrated and quantitative data can be extracted from the pictures.

Summary of the invention

The Applicant has observed that a critical factor to consider when stereophotogrammetric measurements are involved is the calibration procedure, that allows to extract dimensional data from the pictures subsequently acquired within the same system.

As explained above, US 2003/181830 A1 exploits the backdrop for calibration, by providing scale and reference markers applied thereon. The Applicant observes that, to obtain reliable results, such calibration method poses a number of constraints on the acquisition space, e.g. in terms of color of the backdrop (contrasting with respect to that of the markers), sufficient light in the room where measurement is carried out, precise vertical alignment of the backdrop with respect to the ground, etc., thereby negatively affecting the versatility of the overall system.

The Applicant has found that at least some of the issues mentioned above in connection with the calibration procedure may be overcome by making such procedure independent from any vertical element (such as a backdrop wall) in the acquisition space.

Accordingly, in a first aspect, the invention relates to a method for obtaining a spatial pattern of an anatomical structure of a subject, comprising the steps of: a) acquiring, from a digital image capturing device, at least one uncalibrated image of a calibration reference applied on a surface configured to receive the subject, said calibration reference defining at least a first distance along a first reference direction and a second distance along a second reference direction, wherein said first reference direction and said second reference direction are perpendicular to one another and substantially parallel to said surface, and wherein said second reference direction is substantially parallel to a focal axis of said digital image capturing device; b) defining an absolute calibrated reference system of coordinates based on the uncalibrated image of the calibration reference, comprising b1 ) rectifying a perspective distortion of the uncalibrated image, so as to obtain a perspectically rectified uncalibrated image, b2) defining, in said perspectically rectified uncalibrated image, a first coordinate parallel to said first reference direction and a second coordinate along a direction perpendicular to both said first and second reference directions, said first and second coordinates defining an image plane substantially perpendicular to said surface, and b3) determining a pixel-to-real-distance conversion factor that correlates a number of pixels in the perspectically rectified uncalibrated image to a real distance measured in said plane; and c) acquiring, from said digital image capturing device, at least one calibrated image of a plurality of markers positioned on a corresponding plurality of body landmarks of the anatomical structure of the subject; wherein said at least one calibrated image depicts a marker spatial arrangement in said image plane defined by the absolute calibrated reference system of coordinates.

Within the framework of the present description and in the following claims, the expression “marker positioned on a body landmark” in all its possible forms is meant to encompass both a marker physically applied or attached on the skin of a subject at a body landmark, and a marker drawn on the skin of a subject at such body landmark.

Within the framework of the present description and in the following claims, the expression “uncalibrated image” is used, in contrast with the expression “calibrated image”, to indicate an image which has not undergone a complete image calibration operation yet. Within the framework of the present description and in the following claims, by the expression “surface configured to receive the subject”, a surface substantially parallel to or coinciding with the ground is meant to be indicated, said surface extending at least to include a vertical projection of the subject positioned thereon when the at least one calibrated image is captured. The calibrated images are captured by means of a digital image capturing device, therefore avoiding both the exposure of the subject to the harmful ionising radiations of conventional radiographic examinations, as well as the consequent negative environmental impact due to the disposal of related radioactive waste materials.

Since step b) of defining the absolute calibrated reference system of coordinates is carried out based on a calibration reference applied on a surface configured to receive the subject (e.g. the ground surface or any surface parallel to the ground), the calibration procedure can be carried out with no particular restraints on the surrounding acquisition space.

Moreover, data obtained from the above method can advantageously be further used to evaluate postural issues, such as for example scoliosis, lordosis or other polymorphisms, related for example to algic diseases, such as cervical pain, lumbago, sciatica, knee pain, coxalgia, back pain, bursitis, or to degenerative diseases, such as arthrosis, meniscal diseases, ligament or tendon diseases, contractures, or yet post-surgical diseases due to knee, foot, hip or spinal surgery. Other possible advantageous applications of the method are for example the extraction of measurements of body structures, preferably skeletal structures, such as for example long bones (e.g. femur, tibia).

According to a second aspect, the present invention relates to a system for obtaining a spatial pattern of an anatomical structure of a subject, said system comprising:

- a calibration reference applied on a surface configured to receive the subject, said calibration reference defining at least a first distance along a first reference direction and a second distance along a second reference direction, wherein said first reference direction and said second reference direction are perpendicular to one another and substantially parallel to said surface, and wherein said second reference direction is substantially parallel to a focal axis of said digital image capturing device;

- a plurality of markers configured to be positioned on a corresponding plurality of body landmarks of the anatomical structure of the subject;

- a digital image capturing device configured to:

(i) capture at least one uncalibrated image of said calibration reference; and

(ii) capture at least one calibrated image of said plurality of markers positioned on said corresponding plurality of body landmarks, said at least one calibrated image depicting a marker spatial arrangement in an image plane defined by an absolute calibrated reference system of coordinates; and

- a processor (114) programmed to:

(i) acquire said at least one uncalibrated image of the calibration reference from the digital image capturing device;

(ii) define said absolute calibrated reference system of coordinates based on the uncalibrated image of the calibration reference, by: b1 ) rectifying a perspective distortion of the uncalibrated image, so as to obtain a perspectically rectified uncalibrated image, b2) defining, in said perspectically rectified uncalibrated image, a first coordinate parallel to said first reference direction and a second coordinate along a direction perpendicular to both said first and second reference directions, said first and second coordinates defining an image plane substantially perpendicular to said surface, b3) determining a pixel-to-real-distance conversion factor that correlates a number of pixels in the perspectically rectified uncalibrated image to a real distance measured in said image plane; and

(iii) acquire said at least one calibrated image from the digital image capturing device.

Advantageously, the system according to the present invention is particularly simple as it only requires a plurality of markers configured to be positioned on a corresponding plurality of body landmarks of the anatomical structure of the subject, at least one suitably programmed processor, and a digital image capturing device.

All the advantages and effects mentioned above in connection with the method according to the first aspect also apply to the system according to the second aspect of the invention. The present invention can have, in one or more of the above aspects thereof, one or more of the preferred features described hereinafter, which can be combined with one another as desired depending on the application requirements.

Within the framework of the present description and in the following claims, all numerical values indicating amounts, parameters, percentages and so on are always to be intended as preceded by the term “about”, if not otherwise stated. Moreover, all numerical value ranges include all possible combinations of the maximum and minimum numerical values and all possible intermediate ranges, besides those specifically indicated below. The endpoints of the disclosed range are included unless indicated otherwise.

The Applicant has found that in systems and methods for obtaining spatial patterns of anatomical structures based on stereophotogrammetric measurements, precise positioning of markers on the relevant body landmarks of the patient is a critical factor for obtaining a reliable reconstruction of the structure. Precise positioning strongly depends on the skills and experience of the technical staff, variables that can be controlled only to a certain extent.

The Applicant has noted that the prior art documents cited above disclose markers configured as simple flat elements capable of being recognized in photographic images by virtue of the reflective and/or contrasting color of their visible surface with respect to the background. The positional information carried by such markers is thus related to their visible surface in its entirety and does not allow to accurately determine, as it would be desirable, the actual position of the body landmarks, but rather, only an approximate position thereof which could be in any place within the perimeter of the visible surface of each marker. This limitation negatively affects the reliability and reproducibility of the positional information which may be achieved by the above systems.

Therefore, in order to address this further issue found in the prior art, in particularly preferred embodiments of the invention each marker of said plurality of markers comprises a contact face with the skin of the subject, on which said contact point is defined, and an exposed face having a distinctive feature depicted thereon.

Preferably, said distinctive feature is identifiable in said at least one calibrated image by means of an image recognition algorithm.

Preferably, said distinctive feature is arranged in a known geometric relationship with the contact point of the marker with the respective body landmark.

Within the framework of the present description and in the following claims, the expression “contact point of the marker with the body landmark” or similar expressions are meant to indicate a point located substantially at the geometrical centre, more in particular at the centre of symmetry, of a contact face of the marker with the skin of the subject. This contact point substantially corresponds, when the marker is applied with a sufficient precision on the subject, to the position of the relevant body landmark -also defined sometimes repere point- onto which the marker is meant to be applied.

In this way, the markers are “enhanced” by an informational content, carried by a recognizable distinctive feature, which is in turn related to the spatial arrangement of contact points of the markers with the respective body landmarks. Such informational content can in particular be further employed, by means of suitable geometric operations, to “map” the anatomical structure of interest, by extracting the specific position of the contact points of the markers with the related centre of body landmarks.

Markers including a distinctive feature as stated above advantageously allow to accurately determine the specific position of contact points of the markers with the related centre of body landmarks, thereby allowing to obtain a precise reconstruction of an anatomical structure of a patient.

Preferably, the method further comprises the step of: d) determining for each marker of said plurality of markers a position thereof within said absolute calibrated reference system.

More preferably, said step d) comprises determining for each marker of said plurality of markers a position of the contact point with the respective body landmark of the subject within said absolute calibrated reference system.

The positions of the contact points of markers so obtained, also overall indicated, sometimes, as “pattern of points”, can advantageously be used, by means of further elaboration, to reconstruct a shape of the anatomical structure and to extract parameters of interest of such structure.

In embodiments, said steps a)-d) are executed by a same processor.

In this way, the overall method can be carried out by a same processor belonging to a same computational system or device. Therefore, both the preliminary calibration operations and the determination of the position of markers can be carried out on-site, during the image capturing session, or the determination of the position of markers can also be carried out on-site soon after. In alternative embodiments, said steps a)-c) are executed by a first processor, and said step d) is executed by a second processor, said second processor optionally being a remote processor with respect to said first processor.

In this case, the determination of the position of the contact points of the markers can be carried out by a processor belonging to a different (or even remote) computational system with respect to the first processor on which the preliminary calibration has been executed during the image capturing session. For example, the first processor can be a “local” processor, located in proximity of the location in which image capturing is carried out.

The term “remote” is meant to indicate, in the present description and in the attached claims, a processor located physically distant in space from the first processor, such as in another room, building or even city or country.

Therefore, according to this embodiment of the invention the elaboration to determine the pattern of points can be deferred in time and carried out in different locations, and can be advantageously executed on a more performing computational device or system. In turn, the first “local” processor can be implemented by a simplified system requiring less computational power, such as for example a personal computer or a tablet.

In embodiments, the first processor is programmed to further:

(iv) determine for each marker of said plurality of markers a position of the contact point with the respective body landmark of the subject within said absolute calibrated reference system.

According to this embodiment, the system requires a single processor, that can be embodied e.g. by a same computational device located on-site, at the location where image capturing is carried out, or elsewhere and suitably connected to the digital image capturing device.

In alternative embodiments, after (iii) acquiring the at least one calibrated image from the digital image capturing device, the first processor is programmed to further:

(iii bis) transmit said at least one calibrated image of said plurality of markers to a second processor.

In such case, the system preferably further comprises:

- a second processor programmed to: (iii_ter) acquire said at least one calibrated image of said plurality of markers from said first processor; and

(iv) determine for each marker of said plurality of markers the position of the contact point with the respective body landmark of the subject within said absolute calibrated reference system.

The second processor is optionally remote with respect to the first processor.

In this alternative embodiment, the system comprises different processors embodied by computational systems that may be for example located in different places and working in different timeframes. In such case, a first processor executes the calibration operations needed on-site to calibrate the digital image capturing device before capturing the at least one calibrated image, whereas the second processor carries out the elaboration operations leading to the determination of the positions of contact point of the marker with the relevant body landmarks. For example, the first processor can be located in proximity of the location in which image capturing is carried out, whereas the second processor can be a remote processor located elsewhere.

Preferably, said subject is human.

Preferably, said surface configured to receive the subject receives, in use, the subject in standing position thereon.

In particularly preferred embodiments, said anatomical structure of the subject is the spine. Preferably, said markers of said plurality of markers are configured as plate-like elements.

In such case, the contact face and the exposed face of each marker are substantially parallel to one another.

Preferably, said markers of said plurality of markers are substantially two-dimensional elements. In the present description and in the attached claims, the expression “substantially two- dimensional” is used to indicate flat, thin element having a neglectable thickness compared to a length or width thereof. Preferably, the exposed face of each marker comprises a projected contact point aligned to the contact point on the contact surface along a direction perpendicular to the contact face of the marker.

In such case, preferably the contact point and the projected contact point of each marker are spaced of a distance corresponding to the thickness of the marker.

Preferably, said contact face and said exposed face of each marker are shaped as a regular polygon, more preferably selected from a circle, a square, a hexagon.

More preferably, said contact face and said exposed face of each marker are shaped as a circle.

In such case, preferably a diameter of said contact face and said exposed face of each marker is comprised between 0,5 and 2 cm, more preferably between 0,7 and 1 ,5 cm.

Preferably, said distinctive feature comprises a graphic sign.

Preferably, said graphic sign comprises opaque lines over a reflective background.

In this way, a particularly effective contrast is exerted between the graphic sign and the background.

Preferably, said graphic sign is a polygon, more preferably a regular polygon.

Said regular polygon is preferably arranged so that a centre of symmetry of said graphic sign coincides with a centre of symmetry of said exposed face of the marker.

Preferably, said graphic sign is a triangle, more preferably a regular triangle, or a polygon formed of a combination of a plurality of triangles, more preferably a regular polygon formed of a combination of a plurality of regular triangles.

In particularly preferred embodiments, said graphic sign is a regular hexagon.

Preferably, said regular triangle or said regular hexagon have a side of at least 0,2 cm.

Preferably, said projected contact point of the marker corresponds to the centre of symmetry of the exposed face of the marker.

Moreover, said projected contact point of the marker preferably corresponds to the centre of symmetry of said graphic sign. Preferably, the marker is attached to the body landmark of the subject by means of a medical grade adhesive selected from an adhesive paste and an adhesive sheet.

In embodiments, said adhesive is in the form of a removable adhesive sheet which may be provided on the base of the marker. Preferably, the marker is made of a flexible material, preferably plastic.

Preferably, said calibration reference is used to calibrate the images, namely to determine a [pixel] to [cm] conversion factor needed to extract metric data from the images subsequently captured with the digital image capturing device.

Preferably, said step b1) of rectifying a perspective distortion of the uncalibrated image comprises b1-1) setting at least one geometric constraint in connection with the calibration reference. More preferably, said at least one geometric constraint is selected from:

- the first reference direction and the second reference direction being substantially perpendicular to one another; - the first distance being equal to the second distance; and

- the focal axis of the digital image capturing device having a maximum lateral displacement with respect to a median plane comprised within the range of ± 5 mm.

More preferably, all the aforementioned geometric constraint are set in step b1-1).

Preferably, step b) of defining an absolute calibrated reference system of coordinates, and in particular step b1 ) of rectifying a perspective distortion of the uncalibrated image, involve algorithmic operations including geometric, more preferably perspective transformations and/or trigonometric operations.

Preferably, said step b3) of determining a pixel-to-real-distance conversion factor comprises determining a [pixel] to [cm] conversion. Said step b3) of determining a pixel-to-real-distance conversion factor is preferably carried out based on a known resolution of the digital image capturing device, and on said known first and second distances between first and second tags and the second and third tags, respectively, applied onto said surface. Preferably, step b1) of rectifying a perspective distortion of the uncalibrated image further comprises b1 -2) checking if all the geometric constraints set are met.

Preferably, step b1) of rectifying a perspective distortion of the uncalibrated image) comprises b1 -3) if all the geometric constraints set are met, proceed to step b2) of defining the first and second coordinates; or else b1-4) if at least one geometric constraint set is not met, adjusting the system until all the geometric constraints set are met. By way of example, said step b1 -4) of adjusting the system may include physically adjusting the relative position and/or orientation of the digital image capturing device with respect to the surface configured to receive the subject, or physically adjusting the support on which said surface is designed, when provided for.

Preferably, said absolute calibrated reference system of coordinates is a reference system in two dimensions.

In such case, said image plane is preferably defined by two coordinates of said absolute calibrated reference system.

In embodiments, said plurality of body landmarks are located on a posterior side of the subject, on the back of the subject. Therefore, in embodiments, said at least one calibrated image depicts a posterior side of the subject.

In embodiments, said plurality of body landmarks are located on a front side of the subject.

In embodiments, said plurality of body landmarks are located on a lateral side of the subject.

In embodiments, said image plane substantially coincides with a coronal plane of the subject in standing position on the surface.

In embodiments, said image plane substantially coincides with a sagittal plane of the subject in standing position on the surface.

Preferably, said calibration reference comprises at least three tags applied onto said surface.

Preferably, a first and a second tag are positioned at said first distance from one another along said first reference direction, and said second tag and a third tag are positioned at said second distance from one another along said second direction perpendicular to said first reference direction.

Preferably, said first reference direction and said second reference direction are parallel, when the at least one calibrated image is captured by the digital image capturing device, to a latero-lateral direction and, respectively, to a postero-anterior direction of the subject.

Preferably, the calibration operation is carried out based on said first and second distances between tags arranged along said first and second reference directions.

Preferably, said first and second distances are predetermined distances.

Preferably, said first and second distances are equal to each other. In such a case, the calibration operation is thus particularly simplified.

In this case, preferably said first and second distances are comprised between 20 and 60 cm, more preferably they are comprised between 35 and 45 cm.

Preferably, the surface configured to receive the subject is square-shaped and said first, second and third tags are positioned at three corners thereof.

More preferably, said calibration reference comprises a fourth tag positioned at the fourth corner of the squared surface.

Preferably, said at least one uncalibrated image acquired in step a) is captured by the digital image capturing device with a focal axis thereof substantially perpendicular to said image plane.

Moreover, preferably, said at least one calibrated image acquired in step c) is captured by the digital image capturing device with the focal axis thereof substantially perpendicular to said image plane.

Preferably, said at least one uncalibrated image acquired in step a) and said at least one calibrated image acquired in step c) are captured by the digital image capturing device with a lens thereof positioned at a same fixed distance along the focal axis from said image plane. The distance between the lens of the camera and the image plane is chosen based on the lens of the digital image capturing device that is used and on the height of the subject.

The distance between said lens and the at image plane is preferably comprised between 210 and 290 cm. More preferably, said distance between the lens of the camera and the image plane is comprised between 220 and 250 cm.

Preferably, said step c) of acquiring at least one calibrated image of a plurality of markers comprises: c1) acquiring, from said digital image capturing device, a first calibrated image of a first plurality of markers applied on a corresponding first plurality of body landmarks of the anatomical structure of the subject at respective first contact points with the body landmarks, said first calibrated image depicting a spatial arrangement of the first plurality of markers in said image plane; and c2) acquiring, from said digital image capturing device, a second calibrated image of a second plurality of markers applied on a corresponding second plurality of body landmarks of the anatomical structure of the subject at respective second contact points with the body landmarks, said second calibrated image depicting a spatial arrangement of the second plurality of markers in said image plane.

Preferably, the method further comprises, between steps d) and c2) of acquiring the first and second calibrated images, a step of: dbis) rotating the subject about a vertical axis.

Preferably, said surface configured to receive the subject is an upper surface of a support rotatably mounted onto a stationary stand configured to be laid on the ground.

Preferably, said support is rotatable between fixed angular positions angularly spaced of 90°.

In such case, in said step dbis) the subject is preferably rotated about the vertical axis of an angle of 90° or 180°.

In this way, the occurrence of positional changes of the subject due to the rotation, that might cause macroscopic differences in the different images captured, are minimized. Also, in this way the position, rotation or arrangement of the support can be easily adjusted as needed before the image capturing step or based on the subject to be analyzed.

Preferably, said support comprises a guide configured to set up an angle between feet of the subject positioned thereon. In this case, said guide preferably comprises an angled spacer having two oblique sides configured to abut, in use, to an inner side of the feet of the subject, said sides diverging from one another moving from a posterior to an anterior side of the support and forming an angle between each other.

Such guide both fixes the angle between the feet and fixes a lateral displacement of the subject with respect to the support.

In the present description and in the attached claims, the terms “anterior”, “posterior”, "front", "back" and similar will be used with reference to the subject positioned on the surface configured to receive the subject or to upper surface of the above-described support.

In the present description and in the attached claims, the terms “horizontal”, “vertical”, “lateral” and similar terms will be used with reference to configuration of the system when the same is in use.

Preferably, said angle is comprised between 25° and 35°.

More preferably, said angle is of about 30°.

The standing position with feet spaced apart at 30° is indicated as an optimal position to evaluate postural issues. Moreover, the guide allows the position of the subject to be repeatable between subsequent image capturing steps in a same execution of the method, for example between capturing different calibrated images, if the subject needs for any reason, in-between, to leave the support.

Preferably, the guide comprises a length limiter comprising an anterior bar and a posterior bar. Such bars advantageously limit the position of the feet of the subject in an antero posterior direction.

More preferably, a distance between the anterior and the posterior bars can be adjusted so as to adapt to different feet dimensions. Preferably, the guide aids in correctly positioning the subject at a symmetrically centered position onto the support, so that the coronal plane and the sagittal plane of the subject are respectively substantially parallel to or coincide with median planes of the support.

In the present description and in the attached claims, by “substantially parallel”, an angle of 0° ± 1° is meant to be indicated.

Preferably, said image plane corresponds to a median plane of the support.

Preferably, said median planes is a plane passing from a centre of symmetry of the upper surface of the support and perpendicular to the upper surface of the support.

Preferably, said at least one uncalibrated image acquired in step a) and said at least one calibrated image acquired in step c) are captured by the digital image capturing device with the optic axis thereof positioned at a same vertical distance from the surface.

The vertical distance between the focal axis of the camera and the surface configured to receive the subject is selected based on the height of the subject.

Preferably, such vertical distance corresponds to the height of the iliac crests of the subject.

Such vertical distance is preferably comprised between 30 and 130 cm.

Preferably, said at least one uncalibrated image acquired in step a), and said at least one calibrated image acquired in step c), are captured by the digital image capturing device with the focal axis thereof substantially parallel to said surface.

In this way, distortion of the image is minimized or avoided, and as a consequence measurement errors in the calibrated images subsequently captured are advantageously minimized.

Preferably, said at least one calibrated image acquired in step c) are captured by the digital image capturing device with the focal axis thereof substantially parallel either to the postero- anterior direction or to the latero-lateral direction of the subject.

Moreover, preferably said at least one uncalibrated image acquired in step a) is captured by the digital image capturing device with the focal axis thereof substantially parallel either to the postero-anterior direction or to the latero-lateral direction of the subject.

Preferably, the digital image capturing device is a digital camera. Preferably, the camera is provided with a commercial lens selected from a 18-55 mm lens and a 24-105 mm lens.

Preferably, said digital image capturing device has a resolution of at least 12 megapixel. Preferably, said step d) comprises: d1 ) determining a position of a projected contact point on the exposed face of the marker. Preferably, the method further comprises the step of: e) fitting the positions of markers of said plurality of markers, more preferably the positions of the contact points of said markers, with a curve.

Preferably, said step e) of fitting the positions of markers, preferably of the contact points of markers, with a curve is carried out by means of an interpolation operation selected from polynomial interpolation and spline interpolation, preferably by means of cubic spline interpolation.

Preferably, the method further comprises the step of: f) calculating, from the curve fitted in step e), values of parameters of interest related to the anatomical structure of the skeleton of the subject, and/or displaying said curve.

Preferably, the method further comprises the step of: g) comparing said values of parameters of interest calculated in step f) with reference values.

Such reference values preferably include standard or literature values, or previously calculated values of the same parameters of interest for the same subject.

The comparison of the calculated values of the parameters of interest with reference values allows to extract diagnostic data related to the posture of the subject.

Preferably, the method further comprises the step of: h) displaying visual indicators of said parameters of interest and/or displaying said curve (RC).

Preferably, the method further comprises the step of: i) generating a report including an indication of deviations of the calculated values of the parameters of interest with respect to the reference values.

Preferably, said values of parameters of interest calculated in step f) are selected among distances, areas, angles, curvatures.

Preferably, the single processor or the second processor of the system is programmed to carry out one or more of steps e), f), g), h), i) of the method outlined above.

Preferably, said parameters of interest calculated in step f) comprise a Cobb angle.

Preferably, said plurality of markers comprises from 2 to 40 markers.

Preferably, said plurality of markers comprises at least four markers, and said plurality of body landmarks comprises at least:

- a first body landmark corresponding to cervical vertebra C7;

- a second body landmark selected among dorsal vertebrae D4, D5, D6 and D7;

- a third body landmark selected among dorsal vertebrae D8, D9, D10, D11 and D12;

- a fourth body landmark selected among lumbar vertebrae L1 , L2, L3, L4 and L5.

More preferably, said plurality of markers preferably comprises at least six markers, and said plurality of body landmarks further comprises a fifth and a sixth body landmark respectively corresponding to the left and right acromions.

In other embodiments, said plurality of markers preferably comprises up to twenty markers, said plurality of body landmarks further comprising additional body landmarks preferably selected from: apex of left and right axillary cavities, lower apex of left and right scapulae, left and right posterior superior iliac spines, left and right olecranons, left and right lateral knee, left and right outer malleolus, left and right anterior superior iliac spines, left and right radial styloid processes, left and right medial malleolus.

Brief description of the drawings

Additional features and advantages of the present invention will be better illustrated by the following detailed description of some of preferred embodiments thereof, made with reference to the accompanying drawings, in which structural or functional elements having the same or similar function are indicated by identical or similar reference numbers.

In the drawings:

- FIG. 1 is a back view of a first embodiment of a marker according to the present disclosure;

- FIG. 2 is a front view of the marker of FIG. 1 ;

- FIG. 3 is a front view of a second embodiment of a marker according to the present disclosure;

- FIG. 4 illustrates an exemplary distribution of the markers on specific body landmarks of an anatomical structure, in particular the spine, of a human subject;

- FIGs. 5-6 illustrate a system according to a preferred embodiment of the invention, in two different image capturing configurations;

- FIG. 7 illustrates a first embodiment of a support according to a preferred embodiment of the invention; - FIG. 8 illustrates a second embodiment of a support according to a preferred embodiment of the invention;

- FIG. 9 is a flowchart of a method according to a preferred embodiment of the invention for obtaining a spatial pattern of an anatomical structure of a subject;

- FIG. 10 schematically illustrates geometric operations carried out in a calibration procedure of a method according to a preferred embodiment of the invention;

- FIG. 11 schematically illustrates the calculation of parameters of interest on the curve fitted according to a method according to a preferred embodiment of the invention; and

- FIGs. 12-13 schematically illustrate the markers of FIGs. 1-3 with geometric features highlighted. Detailed description of currently preferred embodiments

Throughout the various embodiments disclosed herein, similar elements or elements having a similar function are indicated by the same reference numbers. In the following description, certain specific details are set forth in order to provide a thorough understanding of various disclosed embodiments. However, one skilled in the relevant art will recognize that embodiments may be practiced without one or more of these specific details, or with other methods, components, materials, etc. In other instances, well- known structures associated with the various embodiments have not been shown or described in detail to avoid unnecessarily obscuring descriptions of the embodiments.

With reference to FIGs. 1 -3, preferred embodiments of markers 10, 40 according to the invention, used to carry out a related preferred method later outlined, are now described.

The marker 10 illustrated in FIGs. 1 -2 is configured as a thin and preferably flexible foil, and comprises a contact face 12 (FIG. 1) and an exposed face 14 (FIG. 2).

The contact face 12 is configured to contact and be attached to the skin of a human or animal subject at specific locations, corresponding to predefined body landmarks of an anatomical structure of the skeleton of the subject of interest. To this end, a suitable hypoallergenic adhesive is provided on such contact face 12. By way of a non limiting example, the contact face 12 may be autoadhesive or, alternatively, a double sided adhesive sheet made of a polymeric material. Preferably, the contact face 12 is equipped with a release sheet to be peeled off before applying the marker 10 onto the subject.

The exposed face 14 is the face that remains visible once the marker 10 is attached to the skin of the subject. The exposed face 14 of the marker 10 has a distinctive feature 16 provided thereon. The distinctive feature 16 is in particular a graphic sign, and more in particular a regular triangle (referred to using the same reference number 16) recognizable by processing an image of the marker 10 according to a suitable image recognition algorithm, as better detailed hereinafter.

In the preferred embodiment illustrated, the marker 10 has a circular shape, both the contact face 12 and the exposed face 14 being circular.

Although not visible in the figures, the contact face 12 and the exposed face 14 of the marker 10 are substantially parallel to one another. A thickness (not indicated in the figures), measured perpendicularly to the contact face 12 and the exposed face 14 of the marker 10, is of neglectable entity compared to a diameter D of the marker 10, so that the marker 10 can be considered substantially two-dimensional. The contact point C of the marker 10 with the relevant body landmark to which it is meant to be attached corresponds to a centre of symmetry of the circular contact face 12.

A projected contact point C’ is defined on the exposed face 14 of the marker 10, aligned to the contact point C on the contact face 12 along a thickness direction, namely a direction perpendicular to the contact face 12 and to the exposed face 14.

FIG. 3 shows a marker 40 according to a second preferred embodiment of the invention. The marker 40 differs from marker 10 only in that the exposed face 44 has a different distinctive feature 46 with respect to marker 10 of the previous figures.

The distinctive feature 46 is a graphic sign, in this case a regular hexagon (referred to using the same reference number 46), thus formed of a combination of six regular triangles 47. The regular hexagon 46 is recognizable by processing an image of the marker 40 according to a suitable image recognition algorithm, as better detailed hereinafter.

Apart from the different graphic sign on the exposed face 44, marker 40 is identical to marker 10 and in particular its contact face (not shown) is similar to contact face 16 shown in FIG. 1 for marker 10.

The markers 10, 40 can be manufactured from any suitable hypoallergenic material, preferably flexible so as to better adapt to the skin of the subject, when attached thereon. Suitable materials are for example plastic or polymeric materials.

Preferably, the markers 10, 40 are made of a polymeric material. Different marker sizes, along with dimensions of the graphic signs, can be provided. Preferred sizes are mentioned in the introductory part of the present disclosure.

The markers 10, 40 are employed in the method of the invention, as later described in detail with reference to FIGs. 5-13, to determine a pattern of points of a wide variety of different anatomical structures of human or animal subjects, and to subsequently extract biometric data from such pattern of points. To this end, a plurality of markers 10, 40 is distributed on the subject at specific body landmarks.

The number of markers 10, 40 employed and the selection of body landmarks depends on the complexity of the anatomical structure under analysis and on the desired biometric parameters to be subsequently estimated. For example, for dimensional analysis of simple “linear” structures such as long bones (e.g. femur, tibia), two markers, positioned at the bone ends, are sufficient. For example, for leg length analysis, relevant landmarks are the trochanter and the outer malleolus.

For more complex structures, such as the spine, more than two markers are generally needed.

FIG. 4 schematically illustrates an exemplary distribution of markers on the skin of a human subject FI for biometric analysis of the spine, according to a preferred embodiment of the invention.

The markers may be markers according to any embodiment of markers 10, 40 described above with reference to FIGs. 1-3, preferably all having a same size. The markers may be of a same type or of a different type. In FIG. 4, markers 10a, 10b, 10c and 10d are shown merely as an example.

As shown in FIG. 4, four markers 10a, 10b, 10c and 10d are applied on the back B of the subject FI at a corresponding plurality of body landmarks 50a, 50b, 50c, 50d located along the spine S.

In particular, a first body landmark 50a corresponding to cervical vertebra C7 (approximately within region 52a); a second body landmark 50b is selected among dorsal vertebrae D4, D5, D6 and D7 (approximately within region 52b); a third body landmark 50c is selected among dorsal vertebrae D8, D9, D10, D11 and D12 (approximately within region 52c); and a fourth body landmark 50d is selected among lumbar vertebrae L1 , L2, L3, L4 and L5 (approximately within region 52d).

Additional markers 10e, 10f can be optionally provided to improve the analysis by correlating the morphology and position of the spine S with the surrounding bone structures of the torso. For example, as illustrated, additional markers 10e, 10f are provided respectively positioned at body landmarks 50e, 50f, corresponding to the left and right acromions.

In an even more complex scenario, for analyzing the overall skeleton, (“total body” analysis), up to forty markers may be needed, according to other embodiments (not shown) of the invention.

Besides the six markers 10a-1 Of illustrated in FIG. 6, positioned along the spine and at the acromions, further markers (not shown) may be added, the related body landmarks being for example selected from: apex of left and right axillary cavities, lower apex of left and right scapulae, left and right posterior superior iliac spines, left and right olecranons, left and right lateral knee, left and right outer malleolus, left and right anterior superior iliac spines, left and right radial styloid processes, left and right medial malleolus.

FIGs. 5-6 schematically illustrate a system 100 according to a preferred embodiment of the invention, configured to carry out a preferred method, later described, for obtaining spatial patterns of anatomical structures of a subject H, in particular a human subject H.

The system 100 preferably comprises a plurality 110 of markers (which may be the markers 10, the markers 40 or any suitable combination thereof), a digital image capturing device 112, in particular a camera, a computational device 113 including a processor 114, and a support 116.

In the preferred embodiment illustrated in FIGs. 5-6 and, once again, merely as an example, the plurality 110 of markers includes the markers 40 described above.

In particular, in the acquisition configuration of FIG. 5, a first plurality 110a of markers is provided, including markers 40a-40d as previously described, applied on the back B of the subject FI and distributed along the spine S, in a same manner as previously illustrated in FIG. 4.

The system 100 as set in FIG. 5 is thus configured for extracting measurements related to the spine S of the subject FI, according to a preferred embodiment of the method of the invention.

Preferably, the support 116 is adapted to receive the subject FI, having the markers 40a- 40d attached on the spine at selected body landmarks, in a standardized standing position. The camera 112 is configured and positioned to capture images of the plurality 110 of markers, applied onto the subject FI, from at least two different angles.

Images captured by the camera 112 are transferred to the processor 114 of the computational device 113 for image processing and subsequent elaboration in order to reconstruct a pattern of points of the spine S of the subject FI and further determine related biometric parameters.

Preferably, the support 116 is rotatably mounted onto a stationary base (underneath the support 116 and not visible) that lays on the ground, so that the support 116 is rotatable with respect to the stationary base. Preferably, the support 116 is rotatable between fixed angular positions angularly spaced at 90° from each other.

Preferably, the support 116 is a squared platform with a substantially flat upper surface 118.

With further reference to FIG. 7, it may be appreciated that the support 116 preferably comprises, on the upper surface 118 thereof, four tags 120a, 120b, 120c, and 120d of a reflective material.

Conveniently, the tags 120a, 120b, 120c, 120d constitute a calibration reference CRef used during a calibration operation of the method of the invention, as later described.

The tags 120a, 120b, 120c, 120d of the calibration reference CRef are positioned at the four corners of the upper surface 118 of the support 116. Therefore, each tag is equally spaced from adjacent ones of a distance di = and couples of tags 120a, 120b; 120b, 120c; 120c, 120d; and 120d, 120a lie along reciprocally perpendicular first and second reference directions (X and Z), respectively parallel to or coinciding with a latero-lateral direction and to an postero-anterior direction of the subject H standing on the support 116. Moreover, the first and second reference directions X, Z are respectively parallel to median planes ti and h of the support 116.

Preferably, the support 116 further comprises, fixed on its upper surface 118, a guide 122 (illustrated only in FIG. 7 for the sake of clarity) comprising angled or oblique sides 124, 126 that diverge from one another moving from a posterior side 128 to an anterior side 130 of the support 116.

The oblique sides 124, 126 of the guide 122 are configured to abut, in use, to inner sides of the feet of the subject (not shown) in order to fix and “standardize” the position of the subject. The angle a defined by the sides 124, 126 is of about 30°, which is considered an optimum angle to evaluate postural issues.

In a different preferred embodiment shown in FIG. 8, a support 216 comprises a guide 222 that incorporates a length limiter formed by anterior and posterior bars 224, 226 connected to the guide 222 by means of connecting shanks 228, 230. Bars 224, 226 can be approached or moved away from each other by acting on a displacing mechanism driven by a suitable controller such as a dial 228, correspondingly reducing or increasing a distance therebetween in order to adapt the bars 224, 226 to different feet sizes. Going back to FIGs. 5-6, the camera 112 is configured to capture images of the plurality 110 of markers 40a-40d applied to the skin of subject H.

The camera 112 is positioned so that its focal axis F is substantially parallel to the upper surface 118 of the support 116.

The focal axis F is also substantially aligned with either median plane h (image capturing configuration of FIG. 5) or median plane ti (image capturing configuration of FIG. 6), respectively, with a maximum lateral displacement with respect to median plane h or median plane ti, respectively, comprised within the range of ± 5 mm.

Preferably, the camera 112 is so positioned that its lens 127 is at a predefined distance D with respect to median plane ti (configuration of FIG. 7), respectively median plane h (configuration of FIG. 8) of the support 116. The distance D is preferably comprised between 210 and 290 cm, more preferably of about 230 cm.

Moreover, based on the height of the subject FI, the camera 112 is positioned so that the focal axis F is at a vertical distance h from the upper surface of the support 116, said vertical distance h preferably substantially corresponding to the height of the subject’s iliac crest.

In this way, the image distortion is advantageously minimized.

In the image capturing configuration of FIG. 5, the camera 112 captures a first calibrated image depicting a spatial arrangement of the first plurality of markers 110a in an image plane P substantially parallel to (or coinciding with) the median plane ti of the support 116.

Preferably, in the configuration of FIG. 5, image plane P corresponds to a coronal plane of the subject FI carrying the first plurality of markers 110a.

In other words, in the image capturing configuration of FIG. 5, the camera 112 captures the back B of the subject FI, carrying the first plurality of markers 110a right from behind, along a direction substantially coinciding with a postero-anterior direction of the subject FI.

In the image capturing configuration of FIG. 6, a second plurality 110b of markers is provided on the subject FI. The second plurality 110b of markers includes markers 40a-40d as previously described, applied in this case on a lateral side L of the subject FI at a corresponding plurality of body landmarks (not labeled in FIG. 6). By way of example, the body landmarks on which the second plurality 110b of markers are applied are: shoulder centre, epicondyle, greater trochanter, and external malleolus. The system 100 set as in FIG. 6 is thus configured for extracting postural measurements of the subject H. The camera 112 captures a second calibrated image depicting a spatial arrangement of the markers 40a-40d in said image plane P, this time arranged substantially parallel to (or coinciding with) the median plane h and perpendicular to the median plane ti of the support 116. Preferably, in this case the image plane P corresponds to a sagittal plane of the subject H carrying the second plurality 110b of markers.

In other words, in the image capturing configuration of FIG. 6, the camera 112 captures the lateral side L of the subject FI, carrying the markers 40a-40d, along a direction substantially coinciding with a latero-lateral direction of the subject FI.

The images captured by the camera 112 are then acquired by the processor 114 of the computational device 113, in order to be processed according to a preferred method according to the invention.

The processor 114 can be in data communication with the camera 112 through a suitable communication link 128, such as a cable or a wireless link.

Alternatively, data transfer between the camera 112 and the processor 114 can take place off-line, by means of movable storage devices such as USB flash drives or flash memory cards to be placed in suitable ports/readers provided in the computational device 113 and in the camera 112.

The processor 114, which is for example a central processor (CPU) of the computational device 113, may be coupled to a random access memory (RAM), preferably of at least 8GB, and to a read-only memory (ROM). The ROM may also be other types of storage media to store programs, such as programmable ROM (PROM), erasable PROM (EPROM), etc.

The computational device 113 may include a hard drive, preferably of at least 256 GB.

The processor 114 may communicate with other internal and/or external components through input/output (I/O) circuitry and bussing to provide control signals and the like. The processor 114 carries out a variety of functions as is known in the art, as dictated by software and/or firmware instructions.

The computational device 113 may also include one or more data storage devices, such as a hard disk drive, flash memory drive, CD-ROM drive and/or other hardware capable of reading and/or storing information. Software may be stored in the read-only memory of the computational device 113 and/or distributed on a removable memory device or other form of media capable of portably storing information. These storage media may be inserted into, and read by, devices such as the CD-ROM drive, the disk drive, USB drive etc. Alternatively, the software may be stored on the cloud.

The processor 114 may be coupled to a display, which may be any type of known display or presentation screen, such as LCD displays, LED displays, plasma display, cathode ray tubes (CRT), etc. A user input interface may be provided, including one or more user interface mechanisms such as a mouse, keyboard, microphone, touch pad, touch screen, voice-recognition system, etc.

The processor 114 may be coupled to other computing devices, such as the landline and/or wireless terminals via a network.

The computational device 113 may be part of a larger network configuration as in a global area network (GAN) such as the Internet, which allows ultimate connection to the various landline and/or mobile client devices.

For example, the processor 114 of the computational device 113 may communicate with one or more remote processors.

The computational device 113 can be for example a personal computer (PC), a tablet or even a sufficiently performing smartphone.

With reference to FIGs. 9-13, a method 500 for obtaining a spatial pattern of an anatomical structure of a subject according to a preferred embodiment of the invention is now described.

The method 500 is preferably carried out by the system 100 of FIG. 5. In the description below, reference will be made sometimes to the system 100 and related components, indicated by the same reference numbers as used above in the related description.

As already stated, the method 500 of the invention allows to easily acquire photographic images of a subject FI having a plurality 110 of markers positioned on the skin thereof at predefined anatomical landmarks of an anatomical structure of interest, and thereby allows to easily determine a “pattern of points” of such anatomical structure, based on which parameters of interest may be extracted. In FIG. 11 , even blocks 502-520, bound by solid lines, represent the main steps of the method 500, executed by the processor 114. Odd blocks 501-503, bound by dashed lines, are further operations executed by the camera 112, added to better support the description of the method 500. In block 502, an uncalibrated image depicting a calibration reference is acquired. The uncalibrated image depicts in particular the support 116 carrying on its upper surface 118 the calibration reference CRef, constituted by the tags 120a-120d (shown in FIGs. 5-8).

The uncalibrated image is captured (block 501) by the camera 112 positioned with respect to the support 116 in a similar image capturing configuration as that of FIG. 5, but without the subject FI standing over the support 116.

Based on the uncalibrated image acquired in block 502, in block 504 an image calibration operation is carried out, through which an absolute calibrated reference system (x, y) is defined.

In the calibration operation of block 504, first a rectification of the perspective distortion is carried out, thereby obtaining a perspectically rectified uncalibrated image. Such operation is described with further reference to FIG. 10.

Due to perspective, the square-shaped support 116 appears in the uncalibrated image (as well as in all the subsequently acquired calibrated images) in a distorted configuration, having the shape of a trapezium. A vanishing point U is defined in the uncalibrated image (FIG. 10).

Based on the known real geometry of the support 116, in particular on the know geometric relation between the first and second reference directions X, Z and on the known distances di, d2 between tags 120a-120d of the calibration reference on the support 116, geometric constraints are defined and algorithmically set according to the following formulas (1)-(3): 0° < a £ 1° (1) wherein a is the angle between the first and second reference directions X, Z;

PQ = PS = SR (2)

-0,5 mm £ k £ +0,5 mm (3) wherein k is a lateral displacement of the focal axis F with respect to a median plane of the support (median plane h illustrated in FIG. 7).

The latter geometric constraint corresponds to setting a maximum displacement along the first coordinate x (later defined), of the focal axis F of the digital image capturing device with respect to the second coordinate y (later defined).

From the above geometric constraints, further trigonometric relations are obtained according to the following formulas:

PS = V(ST ² + ((PQ - SR)/2) ² ) (4)

ST = V(PS ² - ((PQ - SR)/2) ² ) (5)

((PQ - SR)/2) = V(PS ² - (PS ² - ST ² ) (6)

The algorithm then checks whether the above geometric constraints are met. If such check is positive, the algorithm proceeds with further steps of the calibration operation; if it is negative, physical adjustment of the system (e.g. repositioning of the support 116 with respect of the digital image capturing device, or vice versa) is needed until the outcome of the check is positive.

Directions X, Z are thus defined based on the aligned tags 120a-120d. In particular, a first reference direction X is set as the direction along which tags 120a and 120b lay, and a second direction Z, perpendicular to said first reference direction X, is set as the direction along which tags 120a and 120d lay.

Therefore, a first coordinate x is defined, having an origin O at a centre of symmetry of the support 116 and parallel to said first reference direction X, and a second coordinate y is defined from said origin O and perpendicular to both said first and second reference directions X, Z, in other words along a vertical direction. In this way, an absolute calibrated reference system of coordinates (x, y) is built, in which coordinates x, y define said image plane P.

Moreover, a [pixel] to [cm] conversion factor is determined based on the known resolution of the camera 112, and on the known distance di, d2 between the tags 120a-120b; 120a- 120d positioned on the support 116 by means of geometric, in particular trigonometric, transformations. Such [pixel] to [cm] conversion factor allows to extract metric data from the images subsequently captured with the digital image capturing device, and in particular to determine distances between points in the image plane P defined by the absolute calibrated reference system (x, y).

As a result of the calibration operation is carried out in block 504, every position within the images subsequently captured by the camera 112 and acquired by the processor 114 are mapped in the absolute calibrated reference system (x, y), so that coordinates of any point of interest in the calibrates images can be simply “read” in the calibrated image.

In step 506, a first calibrated image of a plurality 110 of markers 40a-40d applied over the skin of the subject H at the relevant body landmarks is acquired.

By way of example, the first calibrated image is captured (block 503) by the camera 112 positioned with respect to the subject H as in the image capturing configuration illustrated in FIG. 5, in which the back B of the subject H is captured right from behind.

In such case, the first calibrated image depicts a marker spatial arrangement in the image plane P substantially parallel to (or coinciding with) the median plane ti of the support 116.

In order to obtain a good reconstruction of the anatomical structure of the subject H, from which parameters of interest may be reliably estimated, each of the markers 40a-40d should be attached at the relevant body landmark precisely enough so that the contact point of the marker - substantially corresponding to the centre of symmetry of the contact face of the marker - matches as closely as possible the body landmark on which the marker is to be applied.

To this end, positioning of the markers 40a-40d is preferably carried out by qualified staff, or at least by specifically trained staff.

In subsequent block 508, positions of contact points of each marker 40a-40d with the respective body landmark 50a-50d of the subject H, expressed as two-dimensional coordinates (Xc-i, Yc-i), are determined in the calibrated image acquired in block 503 and captured in the image plane P. Index “i”, used in connection with coordinates, is an integer spanning between 1 and the total number of markers (4 in the case illustrated in FIG. 5).

In detail, to determination of the positions (Xc-i, Yc-i) of contact points C of the markers includes a first step of recognizing the distinctive feature (in particular the regular hexagon 46) provided on the exposed face of the markers 40a-40d. This recognition step is carried out by running a predetermined recognition algorithm, which may be implemented by any suitable commercial or non-commercial software for optical image recognition configured for contrast and brightness detection.

Thanks to the known geometry of the distinctive feature and its predetermined position within the exposed face of the marker, the position of the contact point C of the marker is then determined.

With further reference to FIGs. 12-13, the determination of contact points is described both for a marker 10, in which the distinctive feature 16 provided on the exposed face 14 is a regular triangle, and for a marker 40, in which the distinctive feature 46 provided on the exposed face 44 is a regular hexagon.

For marker 10 (reference to FIG. 11), first the position of the projected contact point C’ is determined according to the following formulas:

HiC’ = H ₂ C’ = n (7)

DC’ = FC’ = r ₂ (8) wherein n and r ₂ are known,

DF = (2 ^* FH) / V3 (9)

FH = (DF ^* V3) / 2 (10)

GC’ = V(FC’ ² - FG ² ) (11)

HC’ = GC’ = V((X _c --i - XH) ² - (Yc-i - YH) ² ) (12) wherein Xc i, XH, YC i and YH are coordinates of the projected contact point C’ in the absolute calibrated reference system.

From the latter formula (12), coordinates (Xc-i, Yc i) of the projected contact point C’ are extracted.

For marker 40 (reference to FIG. 12), the position of the projected contact point C’ is determined according to the following formulas:

HiC’ = ri (13)

DC’ = r ₂ (14) wherein x^ and r ₂ are known, DC’ = (2 ^* HC’) / V3 (15)

HC’ = (DC’*V3) / 2 (16)

HC’ = V((Xc--i - XH) ² - (Yc-i - YH) ² ) (17)

HC’ = GC’ = V((Xc--i - XH) ² - (Yc-i - YH) ² ) (18) wherein Xc-i, XH, Yc -i and YH are coordinates of the projected contact point C’ in the absolute calibrated reference system.

From the latter formula (18), coordinates (Xc-i, Yc-i) of the projected contact point C’ are extracted.

Since markers 10, 40 are substantially two-dimensional, their thickness is neglectable. Therefore, the position of the contact point C of markers 10, 40 (on the contact face, see e.g. contact face 12 in FIG. 1 ) can be approximated to the position of the respective projected contact point C’. The position (Xc-i, Yc-i) of the contact point C of markers 10, 40 is determined as follows:

(Xc-i, Yc-i) = (Xc-i, Yc-i) (19) Going back to FIG. 9, after step 508 of determining the position of the contact point of the marker according to the procedure described above, in subsequent block 510, the positions (Xc-i, Yc-i) of contact points of all markers 40a-40d are fitted with a curve, which constitutes a pattern of the anatomical structure of the subject H.

The fitting operation of block 510 is carried out by means of a suitable interpolation algorithm such as polynomial interpolation or spline interpolation. Preferably, cubic spline interpolations is used.

As known, given a set of n+1 different data points or nodes falling within a main numerical interval and dividing the main numerical interval in n sub-intervals, the cubic spline s(x) is the function defined piecewise by n third-order polynomials, each defined in each of the n sub-intervals.

The second derivative of each polynomial is set to zero at the endpoints of each of the n sub-intervals, to ensure a smooth curvature of the spline s(x) at the “junction” between each couple of polynomials. In the case analyzed, the positions (Xc-i, Yc-i) of contact points of the four markers 40a-40d are set as nodes, the first and the last position along the spine representing the ends of the main numerical interval.

Further data points may be added, within the main interval, so as to define narrower sub- intervals and thereby obtain a smoother curve.

By way of example, the spline s(x) may be defined by 999 polynomials defined in 999 corresponding sub-intervals delimited by 1000 points. Further data points besides the four positions of the markers may be found by setting up a matrix system.

The fitting operation of block 516 yields a curve that represents a pattern of the anatomical structure (e.g. the spine S) of subject FI, that will be referred to hereinafter as reconstructed curve or RC.

In subsequent block 512, which may also run partially in parallel to block 510, parameters of interest are estimated from the reconstructed curve RC determined in block 510.

A parameter of particular interest that can be extracted from the reconstructed curve of the spine is the Cobb angle g, as schematically illustrated in FIG. 11 .

As known in the art, the Cobb angle is defined as the greatest angle at a certain region of the vertebral column, conventionally measured in X-ray images of the spine in the coronal plane, measured from the superior endplate of a predetermine superior vertebra to the inferior endplate of a predetermined inferior vertebra. The Cobb angle allows to estimate the entity of bending disorders of the spine (e.g. scoliosis), both due to postural issues and/or to traumatic events.

The reconstructed curve RC in FIG. 11 is viewed in the coronal plane of the subject FI, therefore it can be represented, in this case, as a one-variable function f(x).

First, to determine the Cobb angle of the reconstructed curve RC, changes of concavity along the reconstructed curve are preferably determined by finding the zero-derivative points (e.g. P1 and P2 in FIG. 10) of the function f(x), calculated by imposing a null second derivative of the function f(x), according to formula (20): f” = d ² f / dx ² = 0 (20) Thereafter, the equations of straight lines wi and w ₂ orthogonal to the reconstructed curve RC at the zero-derivative points P1 and P2 are determined according to the following formulas (21 ) and (22) (in which the y-intercept is neglected for simplicity): wi = ux (21 )

W2 = m ₂ x (22) with u = tan(oo), m ₂ = tan^)

The values of angles w and f are determined based on the first derivative of the zero- derivative points P1 and P2.

Cobb angle g is thus determined according to the following formula (23): g = arctan|(mi-m ₂ )/(1+mim ₂ )| (23)

According to the knowledge in the art, scoliosis is generally associated to a Cobb angle of more than 10°.

Going back to FIG. 9, after blocks 510 and 512, optionally the reconstructed curve RC, as well as visual indicators of the extracted parameters of interest, can be further displayed in block 514, for example on a display of the computational device 113.

In block 516, the parameters of interest are compared with reference values.

Such reference values may include for example standard or literature values. Alternatively or in addition, previously calculated values of the same parameters of interest for the same subject may be used as reference values for the comparison, in order to monitor in time changes in the morphology of the reconstructed anatomical structure, and in general the postural status of the subject.

In subsequent block 518, a report is generated, including in particular an indication of deviations of the calculated values of the parameters of interest with respect to the selected reference values.

Such deviations are preferably associated with a qualitative evaluation of the severity of the postural issue detected, and represent a useful tool for the medical staff to quickly evaluate the postural status of the subject. Finally, in step 520 data related to the reconstructed curve RC, as well as the parameters extracted in block 518, are preferably stored in the memory of the computational device 113 and/or in a remote central database. The database may be an organized database including also the reference values used for the comparison in block 514.

After block 520, step 503 can optionally be returned, thereby capturing a second calibrated image of a second plurality of markers applied over the skin of the subject H at relevant body landmarks. Such second plurality of markers can be different from the first plurality of markers captured in the first calibrated image, and can be applied to different body landmarks, possibly also with the subject assuming a different rotation about a vertical axis. By way of example, the second calibrated image may be captured by the camera 112 in the image capturing configuration illustrated in FIG. 6, in which the lateral side L of the subject FI faces the camera 112.

Steps 506-520 may thus be repeated onto the second calibrated image as described above in connection with the first calibrated image.

Although block 508 has been described with reference to a marker 40, the same analysis applies, mutatis mutandis, to markers 10 illustrated in FIGs. 1-2, taking into account the different geometry of the distinctive feature depicted thereon.

In particular, the same description made above applies to blocks 506-520 by changing "calibrated image" or "first calibrated image" into "second calibrated image.

In addition, it should be underlined that although blocks 502-520 of the method 500 are illustrated and described in a sequence, the order of some blocks could be modified without departing from the scope of the invention.

For example, blocks 510-520, related to the operations of curve fitting, parameters extraction and estimation, display of the reconstructed curve and parameters, generation of report and data storage, could be carried out in a different reciprocal order or even simultaneously.

Moreover, although the system 100 has been described as comprising a single computational device 113 comprising a processor 114, such system may comprise a second remote processor, part of a second remote computational device, located elsewhere. In particular, in such embodiment (not illustrated) the processor 114 may be programmed to execute blocks 502-506 of method 500 of the invention and to further send the calibrated image to the remote processor (not illustrated), which may be programmed to acquire the calibrated image from the processor 114 and to further execute blocks 508-520 of the method 500.

The foregoing detailed description has set forth various embodiments of the devices, system and methods via the use of block diagrams, schematics, and examples. With particular reference to the method according to the invention, insofar as such block diagrams, schematics, and examples contain one or more functions and/or operations, it will be understood by those skilled in the art that each function and/or operation within such block diagrams, flowcharts, or examples can be implemented, individually and/or collectively, by a wide range of hardware, software, firmware, or virtually any combination thereof.

However, those skilled in the art will recognize that the embodiments of the method disclosed herein, in whole or in part, can be implemented as one or more computer programs running on one or more processors similar to processor 114 described above (e.g., as one or more programs running on one or more computer systems), as one or more programs running on one or more controllers (e.g., microcontrollers), as one or more programs running on one or more processors (e.g., microprocessors), as firmware, or as virtually any combination thereof, and that designing the circuitry and/or writing the code for the software and/or firmware would be well within the skill of one of ordinary skill in the art in light of this disclosure.

Those of skill in the art will recognize that many of the steps and algorithms set out herein may employ additional acts, may omit some acts, and/or may execute acts in a different order than specified.

In addition, those skilled in the art will appreciate that the mechanisms taught herein are capable of being distributed as a program product in a variety of forms, and that an illustrative embodiment applies equally regardless of the particular type of signal bearing media used to actually carry out the distribution. Examples of signal bearing media include, but are not limited to, the following: recordable type media such as hard disk drives, CD ROMs, digital tape, and computer memory.