A smart device that can predict the diagnosis of disease with intraoral odor and image data

Abstract

The invention relates to a system based on an artificial intelligence model developed with machine learning that can predict the likelihood of a mouth and throat-related disease that may be present in the patient with the help of intraoral odor and image data, whether it is viral or bacterial, and which type of bacterial infection if it is bacterial.

Description

[0001] A SMART DEVICE THAT CAN PREDICT THE DIAGNOSIS OF DISEASE WITH INTRAORAL ODOR AND IMAGE DATA

[0002] Technical Field

[0003] The invention relates to a system based on an artificial intelligence model developed with machine learning that can predict the likelihood of a mouth and throat-related disease that may be present in the patient with the help of intraoral odor and image data, whether it is viral or bacterial, and which type of bacterial infection if it is bacterial.

[0004] State of the Art

[0005] Today, diseases such as throat-related tonsillitis (tonsil infection) can be viral or bacterial. Treatment also varies depending on the cause of the infection. Apart from this, the separation of throat-related diseases such as pharyngitis and laryngitis is a subject that requires expertise. It is a subjective decision based on the experience of the physician to diagnose without laboratory test from the throat culture to be taken from the patient. These decisions can cause misdiagnosis many times.

[0006] In the present art, diseases of the throat or mouth occur with the complaints of the patient, which are usually symptoms such as pain, difficulty in swallowing, fever. Diseases such as tonsillitis, pharyngitis, laryngitis can be viral or bacterial. In present applications, this distinction is diagnosed with the experience and opinion of a physician, but the final result can be obtained by laboratory examination. Since laboratory work takes time, the start of treatment is delayed.

[0007] There is no direct solution for the diagnosis of oral and throat diseases in the present art. The following studies were found as a result of the research. Although the relevant studies have similar claims, they are different systems in terms of working principle and application.

[0008] Application W02017010683A1 relates to a medical device, the invention of which comprises a stethoscope, a thermometer, a diagnostic imager, an emergency alarm generator, a blood pressure monitor, and an ultrasonic measuring device for connection to a remote medical device, and more particularly to a smartphone; however, it does not use image and odor data.

[0009] Application US20210358582A1 is an imaging apparatus for obtaining images for the diagnosis of the condition of the subject. The system generally relates to various systems, tools, and methods for obtaining diagnostic information, including medical information for a user, transmitting the information to a remote location, evaluating the information, and transmitting the obtained diagnostic and therapeutic information to the user and / or a person. The system has no smell processing capability and no claim to be able to distinguish between specific bacteria and viruses.

[0010] Application US9101384B2 relates to methods, devices, and systems for the diagnosis and / or treatment of sinusitis or other ear-, nose-, or throat-related diseases. However, the invention does not make a viral-bacterial distinction and does not include a machine learning or artificial intelligence module based on the recommended odor and image data.

[0011] As a result, due to the above-mentioned disadvantages and the insufficiency of the current solutions regarding the subject matter, a development is required to be made in the relevant technical field.

[0012] Object of the Invention

[0013] The object of the invention is to solve the above-mentioned disadvantages by being inspired from the current conditions.

[0014] The main object of the invention is to evaluate the data received orally with the embedded camera and odor sensors through the previously trained artificial intelligence unit and to create a system that tells the user the type of infection and the type of agent causing it with a certain percentage.

[0015] Another object of the invention is to provide a system that allows the diagnosis to be made more accurately without the need for a laboratory result and to start treatment without losing time with this prediction, which will provide a supportive feature for the decision of the physician. Another object of the invention is to create a request that can make decisions by using both image and odor data together.

[0016] Another object of the invention is to create a decision support mechanism that can be used in the training of otorhinolaryngologists.

[0017] In order to fulfill the above-mentioned object, the invention is a system that allows to predict the likelihood and details of a mouth- and throat-related disease that may be present in the patient with the help of the intraoral odor ad image data. Accordingly, the system comprises the following:

[0018] • a main body, which contains all the parts that make up the system,

[0019] • a blow rod that allows the patient to blow into the main body,

[0020] • a propeller, which is located in said main body and allows to distribute the air resulting from the blowing into the system and / or to clean the internal air,

[0021] • a motherboard, which provides the management of all elements in the system and transmits the data collected from the patient to the relevant elements,

[0022] • at least one air sensor that measures the components in the air taken from the patient's mouth by means of said blow rod and sends them to said motherboard,

[0023] • a trained gas circuit, pre-trained with machine learning controlled algorithms, which evaluates the measurement data sent by air sensors to said motherboard with machine learning algorithms, predicts with probabilities whether the infection is viral or bacterial and the type of mouth and / or throat disease present in the patient, and communicates the prediction results to the relevant elements,

[0024] • a light that illuminates the inside of the patient's mouth for a clearer image,

[0025] • a camera for obtaining images from the inside of the patient's mouth,

[0026] • an image processing module, which is pre-trained with convolutional artificial neural networks and similar machine learning algorithms, evaluates the images taken by said camera with machine learning algorithms, predicts whether the infection is viral or bacterial and the type of mouth and / or throat disease present in the patient with probabilities, and transmits the prediction results to the relevant elements,

[0027] • a decision-making module that makes the final evaluation with machine learning algorithms by taking the predictions created by said trained gas circuit and image processing module and can transfer the decision created as a result of its own evaluation to a display and / or wirelessly to external systems.

[0028] The structural and characteristic features and all the advantages of the invention will be understood more clearly by means of the figures and the detailed description with reference to these figures given below and therefore, the evaluation should be made by taking these figures and the detailed description into consideration.

[0029] Figures for a Better Understanding of the Invention

[0030] Figure l is a mounted and exploded view of the system of the invention.

[0031] Description of References of the Parts

[0032] 1. Main body

[0033] 2. Handle

[0034] 3. Blow rod

[0035] 4. Tube

[0036] 5. Propeller

[0037] 6. Air sensors

[0038] 7. Motherboard

[0039] 8. Trained gas circuit

[0040] 9. Decision-making module

[0041] 10. Light

[0042] 11. Camera

[0043] 12. Image processing module

[0044] 13. Display

[0045] 14. Bluetooth module

[0046] 15. Wi-Fi module

[0047] 16. On / off button

[0048] 17. Battery

[0049] Detailed Description of the Invention In this detailed description, preferred embodiments of the system of the invention are described for the sole purpose of clarifying the subject matter.

[0050] The invention is a system that allows to predict the likelihood and details of a mouth- and throat-related disease that may be present in the patient with the help of intraoral odor and image data. Figure 1 is a mounted and exploded view of the system according to the invention. Accordingly, the system consists of the following: the main body (1) comprising all the parts that make up the system, the blow rod (3) that allows the patient to blow their breath into the main body (1), the propeller (5) in said main body (1) that distributes the air coming from the blowing into the system and / or cleans the internal air, the motherboard (7) that provides the management of all elements in the system and transmits the data collected from the patient to the relevant elements, the motherboard (7) that measures the components in the air taken from the patient's mouth through said blow rod (3) and sends it to said motherboard (7), an air sensor (6), a trained gas circuit (8) that predicts the infection's viral or bacterial status by evaluating the measurement data sent by the air sensors (6) to said motherboard (7) with machine learning algorithms, which is trained with machine learning controlled algorithms, predicts the probability of the infection being viral or bacterial and what the oral and / or throat disease is in the patient, transmits the prediction results to the relevant elements, a light (10) that illuminates the inside of the patient's mouth to obtain a clearer image, a camera (11) that allows the image to be taken from the inside of the patient's mouth, convolutional artificial neural networks and similar machine learning algorithms includes an image processing module (12) that predicts the viral or bacterial status of the infection and what the oral and / or throat disease is in the patient by evaluating the images taken by said camera (11) with machine learning algorithms and transmits the prediction results to the relevant elements, a decision-making module (9) that makes the final evaluation with machine learning algorithms by taking the predictions created by said trained gas circuit (8) and the image processing module (12), which can transfer the decision created as a result of its own evaluation to a display (13) and / or wirelessly to external systems.

[0051] The invention has a structure consisting of an interlocking handle (2) and a main body (1). It has interconnected parts that enable the operation of the product in the main body (1) and the handle (2). There is a blow rod (3) mounted on the front end of the main body (1). The blow rod (3) allows the breath or air blown by the patient to enter the main body (1). For hygiene reasons, the disposable tube (4) is attached to the blow rod. The tube (4) is ergonomically shaped to match the patient's lips and provide the best blowing.

[0052] The propeller (5) in the main body (1) is used to distribute the incoming air inside or to clean the internal air. As soon as the system is running, the propeller (5) starts to work. There are different numbers of replaceable air sensors (6) mounted on the main body (1). These sensors measure the gas values in the air coming from the patient's mouth and send it to the motherboard (7). The motherboard (7) is a master circuit that controls all components in the system. The trained gas circuit (8) performs the task of evaluating the gas values. The trained gas circuit (8) is pre-trained with machine learning supervised algorithms such as artificial neural networks, support vector machines, gradient boost, decision trees, Naive Bayes, community learning. The trained gas circuit (8) predicts with its probabilities which of the oral or throat diseases such as tonsillitis, laryngitis, pharyngitis, or gingivitis are present in the patient using the incoming gas values. It also has the ability to predict the likelihood of the current infection being viral or bacterial. The trained gas circuit content (8) is updated. The predictions made by the trained gas circuit are sent to the decision-making module (9).

[0053] In the system, the light (10) helps to illuminate the inside of the patient's mouth. The out- mounted camera (11) preferably takes an image of the inside of the patient's mouth. Each image from the camera (11) is sent to the image processing module (12). The image processing module (12) is trained with convolutional artificial neural networks and similar machine learning algorithms and predicts with its probabilities which of the oral or throat diseases such as tonsillitis, laryngitis, pharyngitis, or gingivitis are present in the patient from the image. It also has the ability to predict the likelihood of the current infection being viral or bacterial. The image processing module (12) is updated. The predictions made by the image processing module (12) are also sent to the decision-making module (9).

[0054] The decision-making module (9) makes the final evaluation with machine learning algorithms by taking the predictions made by the trained gas circuit (8) and the image processing module and makes the final decision. The final decisions are reflected on a display (13) of the LCD or similar type. The decision-making module (9) is also capable of connecting to any other computer, printer or display with the connection of the Bluetooth module (14) and / or the Wi-Fi module (15) in the main body (1). In the system, the external access on / off button (16) on the handle (2) is used to turn the system on and off. The battery (17) used in the system is characterized as a rechargeable type battery.

Claims

CLAIMS1. A system that allows to predict the likelihood and details of a mouth- and throat-related disease that may be present in the patient with the help of intraoral odor and image data, characterized in that it comprises the following:• a main body (1), which contains all the parts that make up the system,• a blow rod (3) that allows the patient to breathe into the main body (1),• a propeller (5), which is located in said main body (1) and allows to distribute the air resulting from the blowing into the system and / or to clean the internal air,• a motherboard (7), which provides the management of all elements in the system and transmits the data collected from the patient to the relevant elements,• at least one air sensor (6) that measures the components in the air taken from the patient's mouth by means of said blow rod (3) and sends them to said motherboard (7),• a trained gas circuit (8), pre-trained with machine learning controlled algorithms, which evaluates the measurement data sent by air sensors (6) to said motherboard (7) with machine learning algorithms, predicts with probabilities whether the infection is viral or bacterial and the type of mouth and / or throat disease present in the patient, and communicates the prediction results to the relevant elements,• a light (10) that illuminates the inside of the patient's mouth for a clearer image,• a camera (11) for obtaining images from inside the patient's mouth,• an image processing module (12), which is pre-trained with convolutional artificial neural networks and similar machine learning algorithms, evaluates the images taken by said camera (11) with machine learning algorithms, predicts whether the infection is viral or bacterial and the type of mouth and / or throat disease present in the patient with probabilities, and transmits the prediction results to the relevant elements,• a decision-making module (9) that makes the final evaluation with machine learning algorithms by taking the predictions created by said trained gas circuit (8) and image processing module (12) and can transfer the decision created as a result of its own evaluation to a display (13) and / or wirelessly to external systems.

2. A system according to claim 1, characterized in that it comprises at least one handle (2) connected to the main body (1) and enabling the system to be carried.

3. A system according to claim 1, characterized in that it comprises a tube (4) used specifically for the patient and allows the patient to perform the air blowing process in a hygienic manner.

4. A system according to claim 1, characterized in that it comprises a Bluetooth module (14) that enables the system to make a Bluetooth connection with external elements.

5. A system according to claim 1, characterized in that it comprises a Wi-Fi module (15) that allows the system to make Wi-Fi connection with external elements.

6. A system according to claim 1, characterized in that it comprises an external access on / off button (16) that enables the system to be turned on and off.

7. A system according to claim 1, characterized in that it comprises a rechargeable battery (17) that provides the energy required for the operation of the system.

Images