FOR DISEASE DIAGNOSIS

TECHNICAL FIELD

The invention generally relates to an artificial intelligence-assisted device that can diagnose Covid from breath data.

STATE OF THE ART

With the impact of the Covid- 19 pandemic, public health has gained importance at every level and every individual has started to pay attention to reducing the risk of transmission and disease. The mask has been tried to prevent contamination with distance and hygiene protection and, at times, complete closure that stops life. It seems that the effects of Covid- 19 will continue for a long time.

It is difficult to determine whether people with Covid-19 disease are sick or not since they may not show symptoms. It is necessary to control the persons entering and exiting collective and closed venues such as airplanes, schools, and hospitals, and this control should be reliable. It is important to determine immediately whether the person is the carrier of the disease with a rapid test.

In the present art, PCR-based diagnostic kits have been developed to diagnose Covid- 19 and similar diseases in individuals. These kits have a long diagnostic time. Rapid diagnostic kits also have low reliability with results in a short time. In addition, methods such as taking swabs from the patient's throat and nose make the patient uncomfortable and the procedure difficult.

In the study titled 'Infectious Disease Detection System' with the patent number US20130130227A1, which is in the state of the art, a solution is presented in which the general disease control of the passengers is carried out within the airline system. On the other hand, the invention offers an electronic nose product that can detect Covid- 19 and similar infectious diseases that can be used by individuals alone and can be applied by companies in every branch of the industry.

In the study titled 'Method of detecting plasmodium infection' with the patent number WO20 15077843 Al, which is in the state of the art, new compounds that can detect diseases such as malaria caused by plasmodium parasites are presented. These compounds can easily and rapidly detect the disease; however, they should be used in the laboratory environment and by specialists. The invention proposes a product that can be used both personally and on a large scale, enabling the rapid and high accuracy detection of Covid and similar infectious diseases.

In the study titled 'System and methods for health monitoring of anonymous animals in livestock groups' with the patent number CN102378981A, which is in the state of the art, a system that enables the determination of the diseases of poultry and barn animals with the help of sensors is proposed. In the study, data such as the animals' temperature, weight, and sounds were collected through sensors and the diseased ones were determined by the system. The invention presents a device that can detect Covid- 19 and similar infectious diseases in humans only with its breath.

In the study titled 'Collaborative electronic nose management in personal devices' with the patent number WO2015028364A1, which is in the state of the art, a system that enables the management of different sensors and electronic nose devices in the environments from a central place is presented. This system does not focus on any disease or problem; it deals with the processing and interpretation of data obtained from many devices. The invention proposes a device that is trained with a large amount of patient data and can specifically detect Covid- 19 and similar diseases with high accuracy.

There is a need for improvements in diagnostic devices, therefore there is a need for new embodiments to eliminate the disadvantages mentioned above and to provide solutions to existing systems.

THE OBJECT OF THE INVENTION

The present invention relates to a diagnostic device that meets the aforementioned needs and eliminates all the disadvantages.

The main objective of the invention is to provide a device that can diagnose Covid- 19 and similar infectious diseases by analyzing breath data by using artificial intelligence modules that have already been trained for these diseases.

With the invention, a device that can be competent for different diseases is provided by updating itself, which has the feature of multiple diagnosing gas sensors by competing with each other.

With the invention, a device is provided, which

• will self-sterilize by trapping air that carries dirt or pathogens

• can be used in crowded open or closed venues

• be able to diagnose very quickly and give the probability of this as a percentage

• be able to work with all kinds of machine learning algorithms

• be able to sterilize itself automatically.

The structural and characteristic features and all the advantages of the invention will be understood more clearly by reference to the following figures and the detailed description thereof.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention should be evaluated together with the figures explained below so that it will be constructed, and its advantages will be understood together with the additional elements in the best way.

Figure 1 : Schematic view of the diagnostic device of the invention

REFERENCE NUMBERS

1. Blow lance

2. Blow funnel 3. Oxygen sensor

4. CO sensor

5. CO2 Sensor

6. Hydrogen sensor

7. Nitrogen sensor

8. Argon sensor

9. Helium sensor

10. Ozone sensor

11. Methane sensor

12. Krypton sensor

13. Xenon sensor

14. Nitrogen oxide sensor

15. Digitization module

16. Learned module

17. Learned module update part

18. LCD mini screen

19. Mini fan

20. Air outlet tube

21. UV light

22. Balloon insertion tube

23. Balloon

24. Bluetooth connection

25. USB connection

26. Wi-Fi Connection

27. Mini USB connection

28. Update Module (Inside another computer)

29. Power battery

DETAILED DESCRIPTION OF THE INVENTION

In this detailed description, preferred embodiments of the inventive diagnostic device are only described for a better understanding of the subject of the invention and without any limiting effect. Figure 1 shows a schematic view of the diagnostic device of the invention.

The device user blows air through the blow lance (1) or funnel (2) into the device. The Oxygen sensor (3), CO sensor (4), CO2 Sensor (5), Hydrogen sensor (6), Nitrogen sensor (7), Argon sensor (8), Helium sensor (9), Ozone sensor (10), Methane sensor (11), Krypton sensor (12), Xenon sensor (13), and Nitrogen oxide sensor (14) inside the device measure the relevant part in the entered air. Each sensor sends its measurements to the digitization module (15).

The digitization module (15) creates separate variables by calculating the difference of each value with the other value, the square of its differences, the logarithm, and the sum of the squares of the differences between logarithms. These variables are corrected or normalized with the numerical values of the air in the environment. One or more of the New Min-Max, Z- Score, Logarithmic, and Mastery smoothing methods is used in smoothing. The digitization module (15) transmits the generated data to the learned module (16).

Learned module (16) is a module trained with algorithms such as LSTM, XBOOST, Logistic Regression, linear regression, gradient descent boosted tree, Artificial Neural Networks, Probabilistic Artificial Neural Networks, Naive Bayes learning, Naive Bayes Networks, Random forest, Adaboost, C4.5, and ID3. The trained module (16) predicts the diagnosis of the disease as it has previously learned, using one or more of all the algorithms counted and all the data it encountered. The trained module (16) is a module that can be updated and new algorithms can be installed. Trained module (16) can be connected to the computer via bluetooth (24), USB (25), Wi-Fi (26), and Mini USB connection (27) and can use the update program on the computer.

Update module (28) is software that works on smartphones, tablets, computers, Windows, Linux, MacOS, Android, and iOS operating systems. The update module (28) checks for new training updates from the center at regular intervals. In this way, it can easily diagnose different diseases. The device is updated at any time by connecting to the update module (28). The LCD mini screen (18) writes the diagnosis of the device with the probability value. The LCD mini display (18) shows which disease or diseases it has diagnosed. The mini fan (19) allows the blown air to be distributed homogeneously within the device. The mini fan (19) allows the air to be filled into the balloon (23) after diagnosis. The air outlet tube (20) communicates with the balloon insertion tube (22). The air outlet tube (20) allows some of the air to pass into the balloon (23) during blowing. When a positive diagnosis is made, the mini fan (19) sends all the air to the balloon (23) rope with the air outlet tube (20) and the balloon insertion tube (21). The reason why the blown air is trapped in the balloon (23) is to prevent the possible positive patient breath from contacting the outside environment or the person holding the tool. After the air is sent into the balloon (23), the air outlet tube (20) closes the cover and traps the air in the balloon (23). The balloon insertion tube (21) rotates and bends the mouth of the balloon (23) and prevents air from passing. The UV light (21) becomes active after each operation and sterilizes the air inside.

When a negative diagnosis is made, the air outlet tube (20) expels the air with the help of the mini fan (19). When a negative diagnosis is made, the air outlet tube (20) can trap the air in the balloon (23) for prevention. The power battery (29) supplies electrical power to the device, it operates in the range of 12 - 40 volts. The power battery (29) is rechargeable.