Method and Apparatus for Sample Analysis

Field of the Invention

This invention relates to a method, apparatus and system for sample analysis and refers particularly, though not exclusively, to such an apparatus using a microfluidic device for analyzing a fluid sample, and to a method and system for using that apparatus for analyzing the fluid sample.

Background of the Invention

In the case of sample analysis, particularly in microfluidic systems, changes in light signals result from a reaction or multiple reactions in a reaction chamber. The changes in the light signals represent the nature of the reaction(s) and can be detected and processed. As the reaction involves a liquid, the detection of any light signal changes through a cover (covering the reaction chamber situated above the fluidic level) poses a challenge if the reaction involves a heating process. Prior to detecting the light signals, the fluid in the reaction chamber may be heated to above its vaporization point. This heating may be carried out by contact heating by contacting the base of the device or a reaction chamber with a non-transparent heating surface. Heating may be part of the reaction conditions necessary for analyzing the sample. If there is a sufficient gap between the top surface of the fluid and the inner surface of the cover, this may cause partial evaporation of the fluid to form a vapour or misting of the chamber and/or condensation on the inner surface of the top cover if the cover is of lower temperature than the vapour. Both the misting and condensation could significantly interfere with the detection of weak light signals.

In cases where the light signals from the reaction are strong, the misting or condensation could cause interference that reduces accuracy in direct correlation of the signals with the extent of the reaction(s) and therefore the quantitative capability of the device.

Also, bubbles may be formed when the fluid samples are fed into the reaction chamber or may be formed during the reaction in the reaction chamber. The presence of bubbles will interfere with the light signal and, thus, affect the accurate detection of light signals.

Summary of the Invention

In accordance with a first preferred aspect there is provided apparatus for analyzing a sample fluid, the apparatus comprising:

(a) at least one reaction chamber, each of the at least one reaction chambers being defined by a plurality of sides, a base, and a cover transparent to light;

(b) each of the at least one reaction chambers being for receiving therein the sample fluid for analysis such that the sample fluid at least substantially fills the reaction chamber with there being little or no gap between a top surface of the sample fluid and an inner surface of the cover.

In accordance with a second preferred aspect there is provided apparatus for analyzing a sample fluid, the apparatus comprising:

(a) at least one reaction chamber, each of the at least one reaction chambers being defined by a plurality of sides, a base, and a cover transparent to light;

(b) the plurality of sides of the at least one reaction chamber being of a depth to substantially eliminate the detection of light emanating from the sides;

(c) each of the at least one reaction chambers being for receiving therein the sample fluid for analysis.

For the second aspect, the sample fluid is able to at least substantially fill the reaction chamber with there being little or no gap between a top surface of the sample fluid and an inner surface of the cover.

The apparatus may be mountable on a thermal input device for heating and/or cooling the sample fluid. The at least one reaction chamber may comprise at least two reaction chambers connected by a microfludic circuit. The at least one reaction chamber may be of a length greater than its depth, and a width greater than its depth. Contacting the base of larger area with the heating surface may provide a greater area for heat transfer and hence more efficient heating. Further, a larger base provides greater stability and simplifies engineering design for achieving better contact heating and/or cooling. Heating and/or cooling may be by conduction and/or convection

According to a third preferred aspect there is provided apparatus for analyzing a sample fluid in at least one reaction chamber of at least one mirofluidic device, the apparatus comprising:

(a) a thermal input device having an upper surface for receiving thereon the at least one microfluidic device for heating and/or cooling the fluid in the at least one reaction chamber;

(b) at least one light source mounted for direct illumination of the at least one microfluidic device;

(c) an image capturing system mounted above the thermal input device for capturing at least one digital image of the upper surface of the heater, the at least one microfluidic device and the at least one reaction chamber;

(d) an image processing system for processing the at least one digital image and for providing data relating to the at least one microfluidic device, the at least one reaction chamber and a reaction in the at least one reaction chamber.

The at least one light source may be above the thermal input device and/or to at least one side of the thermal input device.

According to a fourth preferred aspect there is provided a method for analyzing a sample fluid, the method comprising:

(a) placing the sample fluid in at least one reaction chamber of at least one microfluidic device until the sample fluid at least substantially fills the at least one reaction chamber with there being little or no gap between a top surface of the sample fluid and an inside surface of a cover of the at least one reaction chamber; (b) heating and/or cooling the sample fluid:

(c) directly illuminating the sample fluid in the at least one reaction chamber using at least one light source;

(d) capturing light emanating from the at least one reaction chamber using an image capturing system, the image capturing system outputting image data to an image processing system.

According to a fifth preferred aspect there is provided a method of analyzing a sample fluid in at least one reaction chamber of at least one microfluidic device, the method comprising: (a) using a digital image capturing system to capture at least one digital image of the at least one reaction chamber, the at least one image including light emanating from the at least one reaction chamber;

(b) the digital image capturing system sending image data to an image processing system;

(c) scanning the image data to detect intensity variation;

(d) determining potential areas for reaction chamber detection; (e) determining those of the potential areas that are reaction chambers;

(f) comparing the determined reaction chambers to a library of possible reaction chambers; and

(g) labeling the at least one microfluidic device and the at least one reaction chambers.

The determination of (e) may be by.

(a) measuring the intensity of a candidate point;

(b) comparing the intensity of points neighboring the candidate point and determining if their intensity is within a predetermined range compared to that of the candidate point;

(c) repeating (b) until all candidate points have been considered.

The method may further comprise at least one of: (h) within each reaction chamber determining an average intensity and eliminating any value that is outside a predetermined range of values based on the average intensity;

(i) within each reaction chamber discounting any intensity value that is outside a selected range of intensity values; (j) within each reaction chamber discounting any measurement of a different colour tonality; and

(k) within each reaction chamber discounting any measurement that moves during the measurement.

According to a sixth preferred aspect there is provided a method for analyzing a sample fluid, the method comprising:

(a) placing the sample fluid in at least one reaction chamber of at least one microfluidic device until the sample fluid at least substantially fills the at least one reaction chamber with there being little or no gap between a top surface of the sample fluid and an inside surface of a cover of the at least one reaction chamber;

(b) subjecting the sample fluid to at least one of heating and cooling by a thermal input device:

(c) directly illuminating the sample fluid in the at least one reaction chamber using at least one light source;

(d) capturing at least one digital image of the upper surface of the thermal input device, the at least one microfluidic device and the at least one reaction chamber;

(e) processing the at least one digital image and for providing data relating to the at least one microfluidic device, the at least one reaction chamber and a reaction in the at least one reaction chamber.

For the third to sixth aspects, heating and/or cooling may be by a thermal input device. The thermal input device may be a resistive heater, a peltier heater or thermal electric cooler, and may not be transparent to light. The thermal input device may be used to heat or cool or to provide thermal cycling. The thermal input device may have an upper surface with a non-reflective surface for minimizing light reflection by the upper surface. The thermal input device may be mounted on a slider for movement between a retracted position and an extended position. The apparatus may be able to operate only when the thermal input device is the retracted position.

For the third to sixth aspects the at least one light source may be located above and/or to at least one side of the microfludic device, and the image capturing system is above the microfludic device. There may be two light sources. A first light source may be mounted above the thermal input device and to a first side of the contact heater, and a second light source that may be mounted above the thermal input device and to a second side of the contact heater. The first and second sides may be opposed. Each light source may be an array of light emitting diodes, halogen light or any suitable illumination devices. There may be a plurality of microfluidic devices placed randomly on the thermal input device with no overlap of the plurality of microfluidic devices. The image capturing system may be for capturing a plurality of sequential images for enabling the image processing system to provide data relating to the dynamics of the reaction. The image processing system may be for at least one of: a) tolerating artifacts;

(b) image analysis for identifying reaction areas from non-reaction areas;

(c) recognizing boundaries of the microfludic device and the least one reaction chamber; and

(d) detecting a region of the sample fluid in the reaction chamber containing no artifacts and/or bubbles.

According to a seventh preferred aspect there is provided a computer usable medium comprising a computer program code that is configured to cause a processor to execute one or more functions for the performance of one or more of the above methods.

Brief Description of the Drawings

In order that the present invention may be fully understood and readily put into practical effect, there shall now be described by way of non-limitative example only preferred embodiments of the present invention, the description being with reference to the accompanying illustrative drawings.

In the drawings:

Figure 1 is a top perspective view of a first preferred embodiment of an assortment of microfluidic devices on a thermal input device of a preferred analysis apparatus;

Figure 2 is an illustration of different microfluidic devices having different numbers of reaction chambers;

Figure 3 is an enlarged vertical cross-sectional view along the lines and in the direction of arrows 3-3 on Figure 1 ;

Figure 4 is a partial perspective view of a reaction chamber;

Figure 5 is a front schematic view of a preferred embodiment of analysis apparatus;

Figure 6 is a front view (not to scale) corresponding to Figure 5;

Figure 7 is a flow chart of the method for microfluidic device detection;

Figure 8 is a flow chart for reaction chamber detection;

Figure 9 is a flow chart for artifact detection; Figure 10 is two graphs of intensity distribution;

Figure 11 is a flow chart of the analysis process for candidate points in accordance with the fourth preferred aspect of the invention; and

Figure 12 is a flow chart of the analysis process for chamber detection in accordance with the fourth preferred aspect.

Detailed Description of the Preferred Embodiments

To first refer to Figures 1 to 4, there is shown a thermal input device such as a heater and/or cooler 14 that forms part an apparatus for analyzing a fluid. This may be by conduction heating through the base and/or by convection through the atmosphere. By "thermal input device", it is meant to include any device for temperature cycling, i.e. the device 14 may be set to periodically heat and cool the apparatus. The device 14 may not only heat but also may cool the apparatus, depending on the conditions required to carry out the reaction in the reaction chamber. The device 14 may be a contact heater.

Placed on the device 14 are three microfluidic devices 8, 10, 12. Each microfluidic device 8, 10, 12 contains at least one reaction chamber 6 and a microfludic structure 4. The microfluidic devices 8, 10, 12 are for analysis of a fluid 2 in the reaction chamber 6. The number of reaction chambers 6 in each microfluidic device 8, 10, 12 may be any whole number from 1 up to about 12, as in shown in Figure 2. The number of reaction chambers may be any number beyond 12, depending on the desired number of chambers 6 required to analyse a fluid 2. They may be in any suitable, required or desired configuration.

Each chamber 6 of each microfluidic device 8, 10, 12 has a cover 16 that is transparent to light. Each chamber 6 has sides 18 of a length L, ends 20 of depth D, and is of a width W. It is preferred that D be minimized so that W > D. Preferably, W is significantly greater that D. Also, it is preferred that L is significantly greater than D. D should be as small as possible, preferably < 1mm. The relationship between L and W is not of relevance. Examples of dimensions are: length - about 5mm, width - about 1mm, depth - about 0.3mm; or length - about 3.5mm, width - about 1.3mm, depth - about 0.5mm. The reason for this is that when a fluid 2 (Figure 3) requiring analysis is inserted into a reaction chamber 6, the fluid 2 will substantially fill the chamber 6 so there is no, or only a minimal, gap between the top surface 36 of fluid 2 and the inside surface 24 of cover 16. It is preferred that the top surface 36 of fluid 2 contacts the inside surface 24 of cover 16 over at least a part of the inside surface 24. This will assist in minimizing difficulties due to evaporation of fluid 2 to form a vapour in the gap (hence misting the chamber 6), and/or condensation on the inside surface 24. The sealing of each reaction chamber 6, once filled with fluid 2, will further assist this objective. This

avoids the need for a heater to be used on the cover 16. Preferably, fluid 2 flows into each chamber 6 by capillary action.

Although each microfluidic device 8, 10, 12 is shown as being rectangular, they may be of any suitable size or shape including triangular, pentagonal, hexagonal, octagonal, and so forth. Similarly, although each reaction chamber is shown as being cuboid in shape, they may be of any suitable size and shape to facilitate filling of the fluid 2 by any filling method, preferably capillary action.

As is shown in Figures 5 and 6, the microfluidic devices 8, 10, 12 are placed on the device 14. The device 14 may be a resistive heater, peitier heater, thermal electric cooler or other suitable contact heater. The device 14 may be capable of heating and cooling. It is preferred for the device 14 to be not transparent to light so there is no interference with the light from chambers 6. The upper surface of the device 14 may have a matte finish to reduce the reflection of light by the device 14. Alternatively, the upper surface of the device 14 may be coated with a coating having properties that do not interfere with the detection process. Due to W > D and L > D, the base 26 of each microfluidic device 12 has a large surface area contact with device 14 and thus there will be effective heat transfer to the fluid 2 inside each reaction chamber 6. As the device 14 will radiate heat, there will also be heating through the atmosphere. This will be through the cover 16 and sides 18 of the chambers 6. In this way there will be convection and/or conduction heating of the fluid 2.

As the device 14 is preferably not transparent to light, light generation and detection are both above the device 14, and thus are above all microfluidic devices on the device 14. It is preferred for the light generation to be above the device and to one side of the device 14. Light generation may be above the device 14 and to more than one side of the device 14, if required or desired. Also, as W > D and L > D, D being quite small, W x L (the top and base surface area) is very much greater than L x D or W x D, the side surface area. Therefore, minimal light enters each reaction chamber 6 through the sides 18 or 20. The effect of this light on the reaction in the reaction chamber 6, and the resultant light signal emitted from the reaction chamber 6, can therefore effectively be ignored

Since each reaction chamber 6 is sealed, should any gaseous phase be formed in the reaction chamber 6 either as a result of the reaction (as product or by product)

or due to entrapment of air or other contaminants during the filling of the reaction chamber 6, the gaseous phase will result in bubbles. Bubbles interfere with the emission of light from the chamber 6. Because of the large surface area of the fluid 2 available for detection (as L>D and W>D) and the homogenous nature of the reaction process, the light emission signal can be detected from non-affected areas. Where the reaction involves a change in temperature, such as in a cycling process required for a polymerase chain reaction, the changes in temperature and hence the pressure could result in movement of the bubbles. This movement of the bubbles may not allow selection of a fixed, non-affected (i.e. no bubble) area for light detection.

With the fluid 2 substantially filling each reaction chamber 6, preferably contacting inside surface 24 of the cover 16, there may be a tendency for bubbles rising to the top surface 36 of fluid 2 to move and contact other bubbles and thus may form a smaller number of larger bubbles. Larger bubbles are more easily detected than smaller bubbles. Furthermore, bubbles will expand and contract with temperature changes, again leading to easier detection of bubbles due to the changes in the bubble size or shape as the process continues.

The apparatus shown in Figures 5 and 6 consists of:

(a) at least one digital image capturing system 30 located to collect the light signal emitting from the chambers 6. This will depend on a number of factors including, but not limited to: the number and location of the chambers 6, the location of the light generation, the nature of the light generation, the number of light sources, and so forth. The capturing system 30 may be directly above the top surface 38 of device 14 for taking a sequence of digital images of the device 14 and the microfluidic devices 8, 10, 12 thereon, the digital images including light emanating from the reaction chambers 6;

(b) the device 14 for controlling the temperature. The top surface 38 of device 14 may have a non-reflective surface, for example, a matte finish to reduce light reflection from the surface 38 or the surface 38 may be coated with a coating that has properties that do not interfere with the detection process. Reflection is to be taken as including reflection with different characteristics such as, for example, a different wavelength; (c) an image analysis system 32 for receiving and processing recovered image data sent from the image capturing system 30;

(d) at least one light source 28. It is preferred for the light sources 28 (there may be one, two or more) to be located not directly above the device 14, but above and to the side of the device 14 to allow for direct illumination of the surface 38 as well as microfluidic devices 8, 10, 12. Alternatively, the light source may be situated directly above the device 14, for example, in the form of a ring light surrounding the image capturing system 30. If there are two light sources 28 they are preferably on opposite or adjacent sides of the device 14, and more preferably they are not on the same side. The sideways illumination allows for direct illumination and eliminates the need for one or more diachronic mirrors. It also allows of the image capturing system 30 to detect the signal from above the reaction chambers 6. Therefore, both the light source 28 and the image capturing system 30 are above the device 14. This addresses the problem of using contact heating by a peltier heater or thermal electric cooler where the heating surface is not transparent. Furthermore, by D being small, preferably < 1mm, possible shadows are minimized, as is reflection of light by the sides. The light sources 28 may be for ultraviolet radiation (light), an array of LEDs, or other suitable forms of electromagnetic radiation. Blue LEDs may be used and they may be arranged in a panel of 4x10 LEDs. However, the number of LEDs may be that required to ensure sufficient illumination of the heating surface. The total length of the LED panel is the length of at least one row of LEDs (the short side) rather than the total side length of the heating surface. Suitable filters may be applied to the light source 28 and the detectors 30 for selection of specific wavelengths for illumination and detection respectively.

Advantageously, the digital images of the sequence obtained in step (a) are automatically compared.

It is preferred for the device 14 to be mounted on a slider 22 so that the device 14 can be extended out of the apparatus for placement thereon and removal therefrom of the microfluidic devices 8, 10 and 12; and to be subsequently retracted into the apparatus. Preferably, the apparatus is only able to operate when the device 14 is in the retracted position.

The image analysis system 32 is able to detect areas which are not affected by bubbles, and is able to acquire a signal from that area. The image analysis system

32 is able to differentiate a reaction area from a non-reaction area (background).

Hence it is not only able to eliminate the bubble affected areas, it can also identify

a group of reaction chambers belong to the same experiment (e.g. from the same microfluidic device).

The image analysis method is summarized in Figure 7. The right hand column 75 to 79 shows the device 14 with three microfluidic devices 8, 10, 12 thereon. These devices are detected according to the process described in 70 to 74:

(a) 70, 75 - the digital image of the heating block surface 38 (background) where the microfluidic devices 8, 10, 12 are placed is scanned for reaction area(s) (i.e. enclosed reaction chambers 6 with samples 2 of interest); (b) during the scanning, the intensity of each pixel is recorded;

(c) the differences of the intensity changes of the adjacent pixels and its maximum are obtained (process 111 of Figure 11);

(d) 71 , 76 - the differences in intensity changes are compared and the candidate points (Tp) of each potential area for each potential reaction chamber are differentiated from the background and identified (process 112 of Figure 11);

(e) 72, 77 - the reaction area (chamber) detection algorithm (as shown in Figure 8 and 12) is applied on every candidate point to identify the boundary of the potential reaction chambers;

(f) 73, 78 - from the boundary of each potential reaction chamber identified, its dimension and configuration as well as relative position to its neighboring chambers are compared to a library (i.e., a look up table) of possible detection chambers arrangements of each different microfluidic device (101); and

(g) 74, 79 - each microfluidic device is then identified and labeled (e.g. the microfluidic device 8 consisting of reaction chambers 1 , 2, 3, 4, is labeled as C1 and each of the reaction chambers is further designated as C1_1 , C1_2, C1_3, C1_4. )

The first image is used to determine the boundary of the reaction chamber and the chip boundary. The subsequent images taken are used to establish the reaction process. All the pixels inside the boundary of the reaction chambers established in the first image are examined and pixels of an intensity within an established range when compared to the candidate point are considered as those of signals of the reaction area.

Figure 8 shows the specific reaction area detection mentioned in 72 and 77 of Figure 7. The reaction area detection process is further represented schematically in Figure 12:

(a) 80, 83 - the intensity of each candidate point is measured;

(b) 81 , 84 - all neighboring pixels are measured and compared;

(c) 82, 85,86,87,88. If the intensity of neighbouring pixels is within an established range of that of the candidate point, they are accepted as being a part of the detection chamber. This is repeated until all possible candidate points have been considered. Those outside the established range are not considered for detection; and

(d) 88 - the outline of reaction chamber 6 is shown in relief.

The method of eliminating the artifacts, including bubbles, in the reaction area is shown in Figure 9. For every intensity/ color measurement, the point belonging to the reaction chamber 6 will be scanned. Only those within the selected intensity range will be used in the calculation. Artifacts introducing errors in the measurements are usually brighter or darker than average, or have different color tonality, and can be stationary, or may move during measurement.

As an example, the intensity readings of a homogenous reaction area versus one containing artifacts are shown in Figure 10 with (a) being those of a homogenous reaction (101) and (b) being those with artifacts of lower intensity (102).

The image analysis then identifies the full range of intensities of the signal of the reaction area in chamber 6 and regards those that are of the lowest intensity range of certain predetermined percentages as artifacts.

Similarly, the image analysis may also be set to regard those that are on the highest range of the intensity of certain predetermined percentages as artifacts.

This system therefore enables the comparison of the light signals taken in sequence over the duration of a reaction process (91 to 95 in Figure 9) and to track the movement of the artifacts, for example 901 and 902, which are shown to move from 91 to 95. Such ability is particularly useful for accurate real time reaction detection and tracking, and for determining the dynamics of the reaction process in any one reaction chamber 6, or a part of a reaction chamber 6.

The analyzer is therefore able to: a) tolerate artifacts. Artifacts such a bubbles and dirt that may affect the sequential signal detection process and hence introduce errors. For example, bubbles could change position between a first detection and a subsequent detection thereby causing a change in the signals (90 to 95);

(b) identify reaction areas (903) from non reaction area (901 , 902);

(c) recognize the boundary of the reaction chamber (904); and

(d) detect a region of the sample fluid in the reaction chamber containing no artifacts and/or bubbles.

Figure 11 shows the process for determination of candidate pixels p. At the start the derivatives D along the x axis of the intensities of all of the pixels in the image is obtained, and the maximum intensity derivate value Dm is also obtained (1101 ). From there a determination is made whether or not all pixels have been scanned (1102). If yes, the process ends (1103). If not, a pixel p is obtained (1104) and a determination made (1105) whether the derivative of pixel p is greater than its neighbouring pixels: D(p) > D(p-1) and D(p) > D(p+1 ). If no, the process reverts back to (1102). If yes, a determination is made whether D(p) > Dm/5. If no, the process reverts back to (1102). If yes, the empty edge pixel p is designated as Ep1 or Ep2. A further determination is then made (1108) whether two edge pixel positions have been found. If not, the process reverts back to (1102). If yes the target (T) candidate pixel Tp is at the middle of a first edge (E) pixel Ep1 of a reaction area or chamber and a second edge pixel Ep2 that is also at an edge of a reaction area or chamber (1109). A determination is then made (1110) if the intensity of the target pixel Tp > Dm/5. If yes, Tp is inserted into the detection list (1111). The process of Figure 12 is then performed (1112) to flag pixels as potentially valid pixel members of a reaction chamber. A determination is then made (1113) if there has been a detection of a valid reaction area/chamber by comparing with a library/look-up table. If not at (1113) the pixels that were flagged as potentially valid pixels are cleared (1114) and Ep1 is set as being equal to Ep2, and that Ep2 = 0 (1115). If no at (1110) or yes at (1113), the process proceeds directly to (1115). The process then reverts to (1102).

In Figure 12, the term "point" is used to designate a pixel or any values derived therefrom. The candidate point is set as a target point for detection (1201) and recursive detection is commenced (1202). The neighbouring points for the target point are inserted into the queue (1203) and a point obtained from the queue

(1204). A determination is made whether this is the end of the queue (1205) and, if not, a point is retrieved from the queue (1206). If the point density is within an estimated range compared to the candidate point (1207) a determination is made whether the point is in the detection list (1208). If no at (1207) or yes at (1208), the process reverts back to (1204). If no at (1208) the point is marked as a potentially valid pixel member of the reaction area (1209) and inserted into the detection list, and the detection counter increased by 1 (1210). If the detection counter = the end counter, this marks the end of the recursive detection (1211 ) and the process ends (1212). If not at (1211 ) the next point from the detection list is set as the target point and the end counter increased by 1 (1213). The process then reverts back to (1202).

The number of microfluidic devices able to be placed on device 14 is limited only the area of the top surface 38 of device 14. Any number of microfluidic devices may be placed thereon provided the microfluidic devices do not overlap. They may contact each other, but should not overlap. The microfluidic devices may be for different reactions, or may be for the same reaction. The microfluidic devices may be placed anywhere on the device 14 and in any orientation. For a plurality of microfluidic devices, the placement and orientation may be random.

There may also be provided a computer usable medium comprising a computer program code that is configured to cause a processor to execute one or more functions for the performance of the methods described above.

Apart from auto detection of the well and chip, the present invention also enables a user to select and process the detection signals of a particular reaction chamber manually and determine the results manually.

Whilst there has been described in the foregoing description preferred embodiments of the present invention, it will be understood by those skilled in the technology concerned that many variations or modifications in details of design or construction may be made without departing from the present invention.