Laryngeal disease diagnosis assistance system and method

Abstract

The present invention relates to a system and method for analyzing, by using an algorithm, image data obtained by photographing the larynx of a patient and speech data obtained by recording the voice of the patient to preemptively diagnose laryngeal disease, if a result other than a normal result is output, requesting reading according to whether a user wants the reading, and accurately diagnosing the laryngeal disease.

Description

Laryngeal disease diagnosis assistance system and method

[0001] The present invention relates to a laryngeal disease diagnosis assistance system and method using an algorithm, and more specifically, to a diagnosis assistance system and method in which an algorithm analyzes image data and audio data taken of the larynx to make a preemptive diagnosis, and when a result other than normal is output, the laryngeal disease diagnosis assistance system requests an interpretation to make an accurate diagnosis of the laryngeal disease.

[0002] The larynx is the vocal organ located between the pharynx and the trachea, and plays the most important role in speaking and breathing. It also plays a role in preventing food from entering the airway when eating.

[0003] In this way, the larynx is an organ used for speaking, breathing, and eating, and because it is easily exposed to the outside world, infectious diseases caused by viruses and bacteria often occur.

[0004] When a disease occurs in the larynx, symptoms such as sore throat, cough, phlegm, difficulty breathing, hoarseness, and vocal problems appear. Since these are generally similar to cold symptoms, even general doctors who are not otolaryngologists may mistake them for minor symptoms, which may lead to missing the appropriate treatment period.

[0005] Tumors in the head and neck region are caused by smoking, drinking, and human papillomavirus, and diagnosis is often delayed due to nonspecific symptoms. Even when they progress to cancer, complete resection may be difficult due to the complex anatomical structure. Therefore, rapid diagnosis and appropriate treatment are important.

[0006] As conventional techniques for diagnosing laryngeal diseases, Patent Publication No. 10-2063736 and Patent Publication No. 10-2131419 are known.

[0007] Patent Publication No. 10-2063736 discloses a technology for an endoscopic device that can wirelessly photograph and examine the occipital region of a subject using an ultra-high-speed camera and microphone.

[0008] In addition, Patent Publication No. 10-2131419 discloses a technology for recording and analyzing a user's voice based on an application, calculating a voice disorder index from the user's voice, and determining the type of laryngeal disease corresponding to the voice disorder index.

[0009] According to the prior art disclosed in Patent Publication No. 10-2063736 and Patent Publication No. 10-2131419, a technology capable of diagnosing laryngeal diseases by analyzing a patient's laryngeal image and voice is presented, but medical staff without expensive medical equipment may have difficulty diagnosing patients.

[0010] Therefore, in building a disease diagnosis assistance system, if an algorithm preemptively diagnoses a disease based on a dynamic knowledge base composed of medical data related to laryngeal diseases, and if an abnormality is found based on the diagnosis results, the disease diagnosis assistance system remotely requests a reading from medical staff equipped with expensive medical equipment, and accurately diagnoses the patient's disease based on the reading results, it will be possible to build a system that allows medical staff in areas with poor medical facilities to quickly and accurately diagnose diseases.

[0011] [Patent Document]

[0012] Patent Publication No. 10-2063736 (Registration Date: January 8, 2020, Title: Application-Based Laryngeal Disease Monitoring System and Method)

[0013] Patent Publication No. 10-2131419 (Registration date: July 1, 2020, Title: Smart endoscope device using an ultra-high-speed camera for laryngeal performance inspection and system using the same)

[0014] The technical task of the present invention is to provide a laryngeal disease diagnosis assistance system that can make a preemptive diagnosis by analyzing the patient's image data and voice data through an algorithm.

[0015] Furthermore, another object of the present invention is to provide a method for assisting diagnosis of laryngeal diseases, which analyzes image data and audio data to perform a preemptive diagnosis, and when a result other than normal is output as a result of the diagnosis, a laryngeal disease diagnosis assistance system requests a doctor to interpret the result and receives the interpretation result, and a user can use the interpretation result to make an accurate diagnosis of laryngeal diseases.

[0016] In order to solve such technical problems, the laryngeal disease diagnosis assistance system according to an embodiment of the present invention may include a data receiving unit that receives image data of the larynx of a patient to be diagnosed, voice data of recorded voice, and clinical information of the patient to be diagnosed, a laryngeal disease diagnosis unit that analyzes the characteristics of the laryngeal image and voice of the patient to be diagnosed from the received image data and voice data using an inference engine to diagnose one or more laryngeal diseases among normal, laryngitis, benign laryngeal lesion, and malignant laryngeal lesion, a diagnosis request unit that receives an input of whether or not the user wishes to make a diagnosis and requests a diagnosis when a result other than normal is output as a diagnosis result of the laryngeal disease diagnosis unit, and a remote diagnosis result communication unit that, when the diagnosis is requested by the diagnosis request unit, transmits the diagnosis result of the laryngeal disease diagnosis unit, the image data, the voice data, and the clinical information of the patient to be diagnosed to a corresponding hospital server.

[0017] The above image data may be characterized as an image taken using a laryngoscope including a rigid endoscope and a fiber endoscope capable of taking pictures of the larynx of the patient to be diagnosed.

[0018] The above voice data may be characterized as data that is a recording of what the patient to be diagnosed reads about sentences or words set in advance.

[0019] The above clinical information may include at least one of the personal information, medical history, drug allergies, and family history of the patient to be diagnosed.

[0020] The above laryngeal disease diagnosis assistance system further includes a dynamic knowledge base composed of medical data related to the laryngeal disease, wherein the dynamic knowledge base can transmit medical data related to the laryngeal disease to the inference engine for learning of the inference engine.

[0021] The above inference engine can diagnose the laryngeal disease of the patient to be diagnosed using a CNN (Convolutional Neural Network) or Transformer model.

[0022] The above inference engine can diagnose the laryngeal disease by analyzing the image data and the voice data to determine at least one of edema, mucosal discoloration, lesion, vocal cord instability, and voice disorder in the larynx of the patient to be diagnosed.

[0023] The above remote reading result communication unit can receive reading result data read from the corresponding hospital server and transmit it to a user device.

[0024] Meanwhile, a method for assisting in diagnosing laryngeal disease according to another embodiment of the present invention,

[0025] The method may include the steps of receiving image data of the larynx of a patient to be diagnosed, voice data of recorded voice, and clinical information of the patient to be diagnosed from a data receiving unit, the steps of constructing a dynamic knowledge base with medical data related to laryngeal diseases, the steps of analyzing the characteristics of the laryngeal images and voices of the patient to be diagnosed from the image data and voice data using an inference engine of a laryngeal disease diagnosis unit that has learned the medical data related to laryngeal diseases of the dynamic knowledge base, and diagnosing one or more laryngeal diseases among normal, laryngitis, benign laryngeal lesions, and malignant laryngeal lesions, the steps of requesting a reading through a reading request unit upon receiving a result other than normal as a diagnosis result of the laryngeal disease diagnosis unit, and, when the reading request unit requests the reading, the steps of transmitting the diagnosis result of the laryngeal disease diagnosis unit, the image data, the voice data, and the clinical information of the patient to be diagnosed to a corresponding hospital server using a remote reading result communication unit, and the steps of receiving reading result data from the hospital server through the remote reading result communication unit and transmitting it to a user device.

[0026] The above image data may be characterized as an image taken using a laryngoscope including a rigid endoscope and a fiber endoscope capable of taking pictures of the larynx of the patient to be diagnosed.

[0027] The above voice data may be characterized as data that is a recording of what the patient to be diagnosed reads about sentences or words set in advance.

[0028] The above clinical information may include at least one of the personal information, medical history, drug allergies, and family history of the patient to be diagnosed.

[0029] The above dynamic knowledge base is composed of medical data related to the above laryngeal disease, and can be updated with additional data from existing data through at least one medical professional.

[0030] The above inference engine can diagnose the laryngeal disease of the patient to be diagnosed using a CNN (Convolutional Neural Network) or Transformer model.

[0031] The process of diagnosing the laryngeal disease using the inference engine of the laryngeal disease diagnosis unit may include a step of extracting the medical data on images and voices that are correlated with the laryngeal disease from the dynamic knowledge base, a step of training the inference engine based on the medical data on images and voices that are correlated with the laryngeal disease, a step of analyzing the image data and voice data of the patient to be diagnosed using the trained inference engine, a step of determining at least one of edema, mucosal discoloration abnormality, lesion, vocal cord instability movement, and voice disorder in the larynx of the patient to be diagnosed using the analyzed image data and voice data through the inference engine, and a step of diagnosing the laryngeal disease as one or more of normal, laryngeal inflammation, benign laryngeal lesion, and malignant laryngeal lesion using the determined result.

[0032] Specific details of other embodiments are included in the detailed description and drawings.

[0033] According to the technical problem solving means described above, the laryngeal disease diagnosis assistance system and method according to the embodiment of the present invention can assist in the diagnosis of laryngeal disease by using an algorithm based on at least image data and audio data, and thus has the following advantages.

[0034] First, the diagnostic assistance system and method for laryngeal diseases can assist doctors in areas with underdeveloped medical facilities.

[0035] Second, the laryngeal disease diagnosis assistance system and method analyzes the patient's larynx using an algorithm, and if a result other than normal is output among normal, laryngitis, benign laryngeal lesion, and malignant laryngeal lesion, the system can remotely request a reading from a hospital equipped with medical facilities to accurately determine the laryngeal disease.

[0036] Figure 1 illustrates a laryngeal disease diagnosis assistance system according to an embodiment of the present invention.

[0037] Figure 2 illustrates the diagnostic process of a laryngeal disease diagnosis assistance system.

[0038] Figure 3 illustrates a method for assisting in the diagnosis of laryngeal diseases.

[0039] Figure 4 illustrates a process for diagnosing laryngeal diseases by analyzing image data and audio data using an inference engine.

[0040] The following detailed description of the present invention refers to the accompanying drawings, which illustrate specific embodiments in which the present invention may be practiced, in order to clarify the objects, techniques, solutions, and advantages of the present invention. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. Furthermore, throughout the detailed description and claims, the word "comprise" and variations thereof are not intended to exclude other technical features, additions, components, or steps. Other objects, advantages, and features of the present invention will become apparent to those skilled in the art, in part from this description, and in part from practice of the present invention. The examples and drawings below are provided by way of illustration and are not intended to limit the present invention. Moreover, the present invention encompasses all possible combinations of the embodiments set forth herein. It should be understood that the various embodiments of the present invention, while different from one another, are not necessarily mutually exclusive. It should also be understood that the positions or arrangements of individual components within each disclosed embodiment may be varied without departing from the spirit and scope of the present invention. Accordingly, the detailed description set forth below is not intended to be limiting, and the scope of the present invention is defined solely by the appended claims, along with the full scope equivalent to which such claims are entitled, if properly described. Furthermore, unless otherwise indicated herein or clearly contradicted by context, items referred to in the singular encompass the plural unless the context otherwise requires. Furthermore, in describing the present invention, if a detailed description of a related known structure or function is determined to obscure the gist of the present invention, the detailed description will be omitted.

[0041] Fig. 1 illustrates a laryngeal disease diagnosis assistance system according to an embodiment of the present invention, and Fig. 2 illustrates a diagnosis process of the laryngeal disease diagnosis assistance system.

[0042] Referring to FIGS. 1 and 2, the laryngeal disease assistance system (100) can be configured to include a data receiving unit (110), a dynamic knowledge base (120), a laryngeal disease diagnosis unit (130), a reading request unit (140), a remote reading result communication unit (150), a data output unit (160), and a control unit (170).

[0043] The data receiving unit (110) can receive image data (111) of the larynx of a patient to be diagnosed, voice data (112) of recorded voice, and clinical information (113) of the patient to be diagnosed.

[0044] Image data (111) may include images taken using various laryngoscopes, such as a rigid endoscope and a fiber endoscope, which can take pictures of the larynx of a patient to be diagnosed.

[0045] Voice data (112) may include data recorded in which a patient to be diagnosed reads sentences or words set in advance.

[0046] Clinical information (113) may include at least one of the personal information, medical history, drug allergies, and family history of the patient being diagnosed.

[0047] In an embodiment of the present invention, clinical information (113) may include at least one question item inquiring about daily inconvenience and psychological anxiety caused by voice disorder of a patient to be diagnosed, and answers to the questions may be collected to calculate a voice disorder index caused by laryngeal disease of the patient to be diagnosed.

[0048] The dynamic knowledge base (120) is composed of medical data (20) related to laryngeal diseases, and the dynamic knowledge base (120) can transmit the medical data (20) related to laryngeal diseases to the inference engine (131) for learning of the inference engine (131).

[0049] The laryngeal disease diagnosis unit (130) can diagnose one or more laryngeal diseases among normal, laryngeal inflammation, benign laryngeal lesion, and malignant laryngeal lesion by analyzing the characteristics of the laryngeal image and voice of the patient to be diagnosed from the received image data (111) and voice data (112) using the inference engine (131).

[0050] Here, the inference engine (131) of the laryngeal disease diagnosis unit (130) can diagnose the laryngeal disease of the patient to be diagnosed using a CNN (Convolutional Neural Network) or Transformer model.

[0051] The inference engine (131) learns medical data (20) related to laryngeal disease using a CNN (Convolutional Neural Network) or a Transformer model, and the learned inference engine (131) can diagnose the laryngeal disease of the patient to be diagnosed, but is not limited thereto. According to an embodiment of the present invention, the inference engine (131) can learn medical data (20) related to laryngeal disease and diagnose the laryngeal disease of the patient to be diagnosed using at least one or more of various algorithms such as a CNN (Convolutional Neural Network), a Transformer model, as well as linear regression, logistic regression, a decision tree algorithm, a support vector machine, a naive Bayes, a K-nearest neighbor, a random forest algorithm, an ensemble, and a recurrent neural network.

[0052] The inference engine (131) can diagnose a laryngeal disease by analyzing image data (111) and voice data (112) to determine at least one of edema, mucosal discoloration, lesion, vocal cord instability, and voice disorder in the larynx of the patient to be diagnosed.

[0053] When the diagnosis result of the laryngeal disease diagnosis unit (130) outputs a result other than normal, the interpretation request unit (140) can request interpretation by receiving the user's input of whether or not he or she wishes to interpret the result.

[0054] When a reading request unit (140) requests a reading, the remote reading result communication unit (150) can transmit the diagnosis result of the laryngeal disease diagnosis unit (130), the image data (111), voice data (112) and clinical information (113) of the patient to be diagnosed to the corresponding hospital server (30), and the remote reading communication unit (150) can receive the read reading result data from the corresponding hospital server (30).

[0055] The data output unit (160) can transmit the reading result data received from the remote reading communication unit (150) to the user device (10).

[0056] According to an embodiment of the present invention, the user device (10) can be implemented as a smartphone, tablet PC, laptop, desktop, etc. that can be operated by a user.

[0057] The control unit (170) generally corresponds to a server, and the control unit (170) performs overall control of the data receiving unit (110), dynamic knowledge base (120), laryngeal disease diagnosis unit (130), reading request unit (140), remote reading result communication unit (150), and data output unit (160).

[0058] Figure 3 illustrates a method for assisting in diagnosing laryngeal diseases, and Figure 4 illustrates a process for diagnosing laryngeal diseases by analyzing image data and voice data using an inference engine.

[0059] Referring to FIGS. 3 and 4, a method for assisting diagnosis of laryngeal disease can diagnose laryngeal disease by analyzing image data (111) and voice data (112) of a patient to be diagnosed using an inference engine (131), request a reading when a result other than normal is output, and provide the reading result data to a user through a user device (10).

[0060] The diagnostic aid method for laryngeal disease can proceed as follows.

[0061] The method for assisting in the diagnosis of laryngeal disease can proceed with a step of receiving image data (111) of the larynx of a patient to be diagnosed, voice data (112) of recorded voice, and clinical information (113) of the patient to be diagnosed from a data receiving unit (100) (S100).

[0062] The method for assisting in the diagnosis of laryngeal diseases can proceed with the step of constructing a dynamic knowledge base (120) with medical data (20) related to laryngeal diseases (S110).

[0063] According to an embodiment of the present invention, the dynamic knowledge base (120) is composed of medical data (20) related to laryngeal diseases, and can be updated with data added to existing data through at least one medical staff member.

[0064] The assisting method for diagnosing laryngeal diseases can analyze the characteristics of the laryngeal image and voice of a patient to be diagnosed from image data (111) and voice data (112) using the inference engine (131) of the laryngeal disease diagnosis unit (130) that has learned medical data (20) related to laryngeal diseases of the dynamic knowledge base (120), and proceed with the step of diagnosing one or more laryngeal diseases among normal, laryngeal inflammation, benign laryngeal lesion, and malignant laryngeal lesion (S120).

[0065] Here, the process of diagnosing laryngeal disease using the inference engine (131) of the laryngeal disease diagnosis unit (130) using a CNN (Convolutional Neural Network) or Transformer model is as follows.

[0066] The laryngeal disease diagnosis unit (130) can proceed with a step of extracting medical data (20) on images and voices that are correlated with laryngeal disease from the dynamic knowledge base (120) (S200).

[0067] The laryngeal disease diagnosis unit (130) can proceed with a step of learning the inference engine (131) based on medical data (20) on images and voices that are correlated with laryngeal disease (S210).

[0068] The laryngeal disease diagnosis unit (130) can proceed with a step of analyzing image data (111) and voice data (112) of a patient to be diagnosed using a learned inference engine (131) (S220).

[0069] The laryngeal disease diagnosis unit (130) uses the analyzed image data (111) and voice data (112) to determine at least one of edema, mucosal discoloration, lesion, vocal cord instability, and voice disorder in the larynx of the patient to be diagnosed (S230).

[0070] According to an embodiment of the present invention, the inference engine (131) can determine abnormal findings by searching for edema, mucosal color change, and lesion in the anatomical region of the larynx, including the pharynx, vocal cords, epiglottis, and paranasal sinuses of a patient to be diagnosed, in the image data (111).

[0071] In addition, the inference engine (131) can perform acoustic evaluation, aerodynamic evaluation, and physiological evaluation on voice data (112), and can determine the characteristics of the voice of the patient to be diagnosed and the voice disorder index by analyzing at least one or more of frequency, pitch range, sound intensity, jitter, shimmer, noise-to-harmonics ratio (NHR), normalized noise energy (NNE), and voice energy produced during voice production, voice time, sound frequency, and voice efficiency.

[0072] The laryngeal disease diagnosis unit (130) uses the determined result to proceed with the step of diagnosing one or more laryngeal diseases among normal, laryngeal inflammation, benign laryngeal lesion, and malignant laryngeal lesion (S240), and the inference engine (131) of the laryngeal disease diagnosis unit (130) completes the process of diagnosing the larynx of the patient to be diagnosed.

[0073] In the method for assisting in the diagnosis of laryngeal diseases, when the diagnosis result of the laryngeal disease diagnosis unit (130) outputs a result other than normal, the step of requesting a reading through the reading request unit (140) by receiving an input from the user as to whether or not they wish to read the result can be performed (S130).

[0074] In the case where a reading request unit (140) requests a reading, the assisting method for laryngeal disease diagnosis can proceed with a step of transmitting the diagnosis result of the laryngeal disease diagnosis unit (130), the image data (111), voice data (112) and clinical information (113) of the patient to be diagnosed to the corresponding hospital server (30) using the remote reading result communication unit (150) (S140).

[0075] The method for assisting in diagnosing laryngeal diseases proceeds with a step of receiving reading result data from a hospital server through the remote reading communication unit (150) and transmitting it to a user device through the data output unit (160) (S150), thereby completing the method for assisting in diagnosing laryngeal diseases.

[0076] While the present invention has been described with reference to the embodiments illustrated in the drawings, these are merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible. For example, although the present invention has been described based on the assumption of a system for diagnosing laryngeal diseases, the present invention can be equally applied to all cases where laryngeal diseases can be preemptively diagnosed by analyzing image data and audio data without any particular modification. Therefore, the true technical protection scope of the present invention should be defined solely by the appended claims.

[0077] [Explanation of symbols]

[0078] 100: Laryngeal Disease Diagnosis Assistance System

[0079] 110: Data receiving unit

[0080] 120: Dynamic Knowledge Base

[0081] 130: Laryngeal Disease Diagnostic Department

[0082] 131: Inference Engine

[0083] 140: Reading Request Department

[0084] 150: Remote reading results communication department

[0085] 160: Data output section

[0086] 170: Control unit

Claims

1. A data receiving unit that receives image data of the larynx of a patient to be diagnosed, voice data of recorded voice, and clinical information of the patient to be diagnosed; A laryngeal disease diagnosis unit that analyzes the characteristics of the laryngeal image and voice of the patient to be diagnosed from the received image data and voice data using an inference engine and diagnoses one or more laryngeal diseases among normal, laryngitis, benign laryngeal lesion, and malignant laryngeal lesion; When the above laryngeal disease diagnosis section outputs a result other than normal, a diagnosis request section that receives the user's input of whether or not they wish to have the diagnosis performed and requests the diagnosis; and A laryngeal disease diagnosis assistance system, comprising: a remote interpretation result communication unit that transmits the diagnosis results of the laryngeal disease diagnosis unit, the image data, the voice data, and the clinical information of the patient to be diagnosed to the corresponding hospital server when the interpretation request unit requests the interpretation.

2. In paragraph 1, A laryngeal disease diagnosis assistance system characterized in that the image data is an image captured using a laryngoscope including a rigid endoscope and a fiber endoscope capable of capturing an image of the larynx of the patient to be diagnosed.

3. In paragraph 1, A laryngeal disease diagnosis assistance system characterized in that the above voice data is data recording what the patient to be diagnosed reads about sentences or words set in advance.

4. In paragraph 1, A laryngeal disease diagnosis assistance system, wherein the above clinical information includes at least one of personal information, medical history, drug allergies, and family history of the patient to be diagnosed.

5. In paragraph 1, The above laryngeal disease diagnosis assistance system further includes a dynamic knowledge base composed of medical data related to the above laryngeal disease, The above dynamic knowledge base is a laryngeal disease diagnosis assistance system that transmits medical data related to the laryngeal disease to the inference engine for learning of the inference engine.

6. In paragraph 1, The above inference engine is a laryngeal disease diagnosis assistance system that diagnoses the laryngeal disease of the patient to be diagnosed using a CNN (Convolutional Neural Network) or Transformer model.

7. In paragraph 1, A laryngeal disease diagnosis assistance system wherein the above inference engine analyzes the image data and the voice data to determine at least one of edema, mucosal discoloration, lesion, vocal cord instability, and voice disorder in the larynx of the patient to be diagnosed, thereby diagnosing the laryngeal disease.

8. In paragraph 1, The above remote reading result communication unit is a laryngeal disease diagnosis assistance system that receives reading result data read from the corresponding hospital server and transmits it to a user device.

9. A step of receiving image data of the larynx of a patient to be diagnosed, voice data of recorded voice, and clinical information of the patient to be diagnosed from a data receiving unit; Step of constructing a dynamic knowledge base with medical data related to laryngeal diseases; A step of analyzing the characteristics of the laryngeal image and voice of the patient to be diagnosed from the image data and voice data using the inference engine of the laryngeal disease diagnosis unit that has learned the medical data related to the laryngeal disease of the dynamic knowledge base, and diagnosing one or more laryngeal diseases among normal, laryngitis, benign laryngeal lesion, and malignant laryngeal lesion; When the diagnosis result of the above laryngeal disease diagnosis department outputs a result other than normal, a step of requesting a reading through the reading request department by receiving the user's input of whether or not they wish to read the result; When the interpretation request department requests the interpretation, the step of transmitting the diagnosis results of the laryngeal disease diagnosis department, the image data, the voice data and the clinical information of the patient to be diagnosed to the corresponding hospital server using the remote interpretation result communication department; and A method for assisting in the diagnosis of laryngeal disease, comprising: receiving reading result data from the hospital server through the remote reading result communication unit and transmitting the data to a user device.

10. In paragraph 9, A method for assisting in the diagnosis of laryngeal disease, characterized in that the image data is an image captured using a laryngoscope including a rigid endoscope and a fiber endoscope capable of capturing an image of the larynx of the patient to be diagnosed.

11. In paragraph 9, A method for assisting in the diagnosis of laryngeal disease, characterized in that the above voice data is data that records what the patient to be diagnosed reads about sentences or words set in advance.

12. In paragraph 9, A method for assisting in the diagnosis of laryngeal disease, wherein the clinical information includes at least one of personal information, medical history, drug allergies, and family history of the patient to be diagnosed.

13. In paragraph 9, The above dynamic knowledge base is composed of medical data related to the above laryngeal disease, and is a laryngeal disease diagnosis assistance method that updates data added to existing data through at least one medical professional.

14. In paragraph 9, The above inference engine is a laryngeal disease diagnosis assistance method that diagnoses the laryngeal disease of the patient to be diagnosed using a CNN (Convolutional Neural Network) or a Transformer model.

15. In paragraph 9, The process of diagnosing the laryngeal disease using the above inference engine of the above laryngeal disease diagnosis department is as follows: A step of extracting the medical data for images and voices having a correlation with the laryngeal disease from the dynamic knowledge base; A step of training the inference engine based on the medical data for images and voices having a correlation with the above laryngeal disease; A step of analyzing the image data and voice data of the patient to be diagnosed using the learned inference engine; A step of determining at least one of edema, mucosal discoloration, lesion, vocal cord instability, and voice disorder in the larynx of the patient to be diagnosed using the above analyzed image data and voice data through the inference engine; and A method for assisting in the diagnosis of laryngeal diseases, comprising: a step of diagnosing one or more of the above laryngeal diseases among normal, laryngitis, benign laryngeal lesion, and malignant laryngeal lesion using the above-determined results.

Images