TW202416290A - System and method for recommending a treatment plan for obstructive sleep apnea - Google Patents

Abstract

A system and method for recommending a treatment plan for obstructive sleep apnea, including: an artificial intelligence module; and a processor connected to the artificial intelligence module; wherein the processor performs the following operations: inputting multiple patient physiological feature parameters, treating solution parameters and efficacy judgment indicators to the artificial intelligence module to train the efficacy prediction model; and inputting a plurality of treatment solutions into the efficacy prediction model to generate a ranking result of the plurality of treatment solutions. In this way, individual treatment solutions are recommended and prioritized to patients and caregivers.

Description

Obstructive sleep apnea treatment plan recommendation system and method

The present invention relates to a system and method for recommending treatment plans for obstructive sleep apnea (OSA).

The current treatment of obstructive sleep apnea can be divided into surgical treatment and non-surgical treatment, both of which aim to increase the airway space and reduce the risk of sleep apnea. However, there are many treatment options for both surgical and non-surgical treatments, and there is a need to provide the best treatment plan for each patient's symptoms.

The present invention provides an obstructive sleep apnea treatment plan recommendation system and method, which can collect the symptoms of each patient to provide a ranking result of treatment plans and a better treatment plan.

The obstructive sleep apnea treatment plan recommendation system provided by the present invention includes an artificial intelligence module and a processor connected to the artificial intelligence module. The processor performs the following operations: inputting multiple patient physiological characteristic parameters, treatment solution parameters and efficacy judgment indicators to the artificial intelligence module to train the efficacy prediction model; and inputting multiple treatment solutions to the efficacy prediction model to generate a ranking result of multiple treatment solutions.

The obstructive sleep apnea treatment plan recommendation method provided by the present invention is applicable to an obstructive sleep apnea treatment plan recommendation system having a processor, wherein the obstructive sleep apnea treatment plan recommendation method is executed by the processor and includes: inputting multiple patient physiological characteristic parameters, treatment solution parameters and efficacy judgment indicators into an artificial intelligence module to train an efficacy prediction model; and inputting multiple treatment solutions into the efficacy prediction model to generate a ranking result of multiple treatment solutions.

In one embodiment of the present invention, multiple patient physiological characteristic parameters include non-time series characteristic parameters and time series characteristic parameters. The non-time series characteristic parameters at least include patient respiratory imaging parameters, patient medical history parameters, and patient genetic parameters. The time series characteristic parameters at least include patient wearable device physiological parameters, patient apnea/hypopnea index (Apnea/Hypopnea Index; AHI) parameters, surface patch electrode electrical stimulation parameters, and braces parameters.

In one embodiment of the present invention, the electrical stimulation frequency in the electrical stimulation parameters is a low frequency input in the range of 20-100 Hz and at intervals above 1 Hz, and in the range of 1 k-10 kHz and at intervals above 50 Hz, and is arranged and combined at various positions of the surface patch electrode.

In one embodiment of the present invention, the processor performs data filling on non-time series feature parameters and time series feature parameters to generate time series parameters. The data filling at least includes deleting outliers and filling missing values.

In one embodiment of the present invention, the processor performs data integration on the time series parameters to sort them into three-dimensional time series parameters, and the data integration at least includes segmenting the time series parameters and arranging the segmented time series parameters.

In one embodiment of the present invention, the processor performs feature extraction on the three-dimensional time series parameters to generate therapeutic efficacy judgment indicators, and the feature extraction at least includes time domain-frequency domain conversion and convolution operation on the three-dimensional time series parameters.

The present invention collects the patient's medical history and physiological data, and uses these cases and physiological data as input to train an artificial intelligence module to generate a treatment effect prediction model. Therefore, when all treatment solutions are input, all treatment solutions can be ranked to obtain the best treatment solution.

In order to make the above and other purposes, features and advantages of the present invention more clearly understood, embodiments are specifically cited below and described in detail with reference to the accompanying drawings.

Please refer to FIG. 1, which is a schematic diagram of an obstructive sleep apnea treatment plan recommendation system provided by an embodiment of the present invention. The obstructive sleep apnea treatment plan recommendation system 1 provided by the present invention includes an artificial intelligence module 2, a processor 3, a storage device 4, an electrical stimulation unit 5, a physiological sensing module 6, and an electrode battery 7, wherein the processor 3 is connected to the artificial intelligence module 2, the storage device 4, the electrical stimulation unit 5, and the physiological sensing module 6, wherein the artificial intelligence module 2 is used to predict the electrical stimulation parameters and the braces parameters according to the patient's background data combined with the physiological sensing data, and the storage device 4, such as a cloud server or a local server, stores the obstructive sleep apnea treatment plan recommendation system 1 executed by the processor 3. The procedure for recommending a treatment plan for obstructive sleep apnea, as well as the medical history and physiological characteristic parameters of all patients, treatment solution parameters and efficacy judgment indicators, the electrical stimulation unit 5 is used for electrical stimulation, the physiological sensing module 6 is used to detect physiological sensing data, the electrode battery 7 is connected to and provides power to the artificial intelligence module 2, the processor 3, the storage device 4, the electrical stimulation unit 5 and the physiological sensing module 6, among which the processor 3 is particularly used to receive the physiological sensing data and perform real-time calculations through the artificial intelligence module 2 to determine the electrical stimulation parameters and the braces parameters. Specifically, the electrical stimulation parameters are that the capsule electrode patch A, adhesive electrode patch B and C are placed in the mouth as shown in Figure 2 or at the junction of the chin and throat as shown in Figure 3, and the optimal parameters of waveform, size and frequency are adjusted according to the patient's individual parameters. The adhesive electrode patch can be designed with a micro-array electrode surface, using a large-area patch to cover at least 16 areas, with many-to-many positive and negative electrodes, and the position of the positive and negative electrodes is determined and replaced in real time according to the artificial intelligence algorithm; and the brace parameter is to use brace D to drive the front of the lower jaw The system can correct the abnormalities by adjusting the position of the braces and predict the correction parameters according to the patient's personalized physiological parameters and background data. It can also use artificial intelligence to select the best personalized braces. For example, it can determine the positioning points of OSA braces and design adjustable mechanical joints and software-controlled OSA braces. OSA braces are mainly used to position the upper and lower teeth at the position where the mandible extends forward, protrude the mandible, pull the tongue and respiratory muscles, expand the respiratory tract to keep the respiratory tract and throat open, and prevent the tongue root and soft palate muscles from falling backward and downward. The positioning principle of mandibular protrusion and slight opening, the balance position of temporomandibular joint, and the opening action can avoid muscle discomfort caused by excessive opening of the mouth during wearing OSA braces, reduce temporomandibular joint symptoms and discomfort of the masticatory muscles; at the same time, according to the individual difference between the upper and lower jaws of the patient and the horizontal distance between the upper and lower central incisors, the mandibular bone is guided to move horizontally to a sufficient amount to achieve the effect of unobstructed airway and throat. The thickness of OSA braces will refer to the vertical distance between the upper and lower posterior teeth, the distance between the upper and lower front teeth, the guide surface formed by the upper incisor palate and the incisor tip of the lower incisor, the inclination direction of the occlusal plane, and the joint activity coefficient. At the most suitable distance for the mandibular opening and forward extension, the OSA braces are given sufficient thickness. The area and height of the OSA braces on the cheek and lingual side of the teeth will be based on the tooth analysis direction and the insertion path to avoid being difficult to wear or too tight, and to avoid being too loose and falling off during sleep. The braces have a special barbed design. If the tongue is paralyzed here, the barbed design can be used to reflexively retract the tongue to avoid airway obstruction. In one example, the present invention can perform real-time calculations to determine electrical stimulation parameters based on an artificial intelligence model established by personalized information. When sleep apnea occurs, the physiological information (SpO2 changes) is received in real time by the physiological sensing module 6 (such as a blood oximeter), and input into the processor 3 to determine whether to perform electrical stimulation, and adjust the size of the electrical stimulation according to the received physiological information, and then execute the electrical stimulation unit 5. If the electrical stimulation improves sleep apnea, the electrical stimulation can be automatically stopped by feedback based on the real-time physiological information, and the location of the patch is the non-invasive hypoglossal nerve location. In addition, it can be noted that the artificial intelligence module 2 can be included in the processor 3, or it can be a separate hardware, a combination of hardware and software, or a device or equipment that can perform machine learning neural networks. In this embodiment, the artificial intelligence module 2 exists independently of the processor 3 in the obstructive sleep apnea treatment plan recommendation system 1.

Please refer to FIG. 4, which is a flow chart of a method for recommending a treatment plan for obstructive sleep apnea provided by an embodiment of the present invention. The method for recommending a treatment plan for obstructive sleep apnea provided by the present invention includes step S1: the processor 3 inputs a plurality of patient physiological characteristic parameters, treatment solution parameters and efficacy judgment indicators to the artificial intelligence module 2 to train the efficacy prediction model, wherein the plurality of patient physiological characteristic parameters include non-time series characteristic parameters and time series characteristic parameters, and the efficacy judgment indicators at least include the patient's apnea hypopnea index (AHI), blood oxygen concentration index, snoring index, etc., but the present invention includes the above judgment indicators but does not limit other judgment indicators. Non-time series characteristic parameters at least include patient respiratory imaging parameters, patient medical history parameters, and patient genetic parameters, such as medical imaging related information (computer tomography, ultrasound and X-ray images of the nasal cavity, nasopharynx, pharynx, head, craniofacial and upper respiratory tract), medical history data (height, age, weight, BMI, other comorbidities such as diabetes, hypertension, chronic kidney disease, heart disease and depression, etc.), medical records (neck size, medication records of sleeping pills or sedatives), genetic data (congenital central hypoventilation, bacterial strain genes, skeletal occlusal abnormalities of upper and lower jaws), blood tests (cardiovascular disease data CRP, hsCRP and homocysteine, etc.; diabetes data glycosylated hemoglobin and blood sugar level, etc.); Renal function data: urine albumin, creatinine, blood urea nitrogen and glomerular filtration rate; blood lipid index data (low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides and total cholesterol), etc. However, the present invention includes the above-mentioned non-time series characteristic parameters but does not limit other non-time series characteristic parameters. Time series characteristic parameters at least include physiological parameters of the patient's wearable device (electroencephalogram, eye movement parameters, mandibular electromyogram parameters, heart rate variability and electrocardiogram, such as blood oxygen concentration, blood pressure (systolic pressure, diastolic pressure or mean arterial pressure)), patient respiratory arrest hypopnea index The clinical parameters of the patient were as follows: respiratory parameters (number of hypoventilation caused by respiratory arrest and airway stenosis in each hour, such as respiratory airflow (airflow size and obstruction location), respiratory dynamics (airway dilation size, frequency and length) and respiratory noise (snoring sound size, snoring length and frequency), electrical stimulation parameters of the surface patch electrode (duration, wavelength, intensity, surface electrode stimulation location and electrical stimulation frequency), and brace parameters (U-shaped parameter design, mandibular The present invention includes the above-mentioned time series characteristic parameters but does not limit other time series characteristic parameters. Specifically, the processor 3 performs data filling on the non-time series characteristic parameters and the time series characteristic parameters to generate time series parameters, wherein the data filling at least includes outlier removal and missing value filling on the various parameters after statistics; then, the processor 3 performs data integration on the time series parameters to sort them into three-dimensional time series parameters. The data integration at least includes segmenting the time series parameters and arranging the segmented time series parameters to generate three-dimensional time series parameters; and the processor 3 performs feature extraction on the three-dimensional time series parameters to generate a therapeutic efficacy judgment indicator, wherein the feature extraction at least includes performing a time domain-frequency domain conversion and a convolution operation on the three-dimensional time series parameters to obtain features to define the therapeutic efficacy judgment indicator. Then, the processor 3 uses the artificial intelligence module 2 such as RNN Models or machine learning are used for prediction training (such as SVM, Xgboost, LightGBM) to train efficacy prediction models, which can be divided into AHI numerical prediction models, binary classification prediction models for whether there is efficacy, and complex complication risk assessment predictions. In this way, the electrical stimulation parameters with the lowest damage and no impact on sleep, as well as the most effective and aesthetic braces parameters can be provided. Then, enter step S3.

Step S3: Processor 3 inputs multiple treatment solutions into the treatment effect prediction model to generate a ranking result of the multiple treatment solutions.

Generally speaking, after establishing the efficacy prediction model, the various parameters of the patient's background (image respiratory characteristics, medical history data, genetic data, wearable physiological characteristics data and AHI diagnosis and evaluation) are first input, and then various treatment parameters are input. For example, the input parameters are electrical stimulation parameters such as low frequency in the range of 20-100Hz and intervals above 1Hz, in the range of 1k-10kHz, and intervals above 50Hz, and the various positions of the surface patch electrodes are arranged and combined to find the best efficacy value to determine the best parameter combination. The use of the sensing end example For example, a blood oximeter can be used to measure the state of hypoxia due to respiratory arrest, indicating whether the patient needs electrical stimulation. At the same time, a wireless electrical stimulation feedback system can be used to automatically apply electrical stimulation during sleep to reduce snoring. Barbs may be designed inside the braces to retract the tongue. Electrical stimulation and tongue training are used to train the tongue muscles, shrink the tongue, and increase the flexibility of the tongue, improving The swallowing problem of the elderly is treated, but the parameters of each patient are different. Therefore, the efficacy prediction model is used to predict and conduct personalized treatment. The genetic data, CT, blood oxygen concentration and ultrasound of each patient are used to obtain various values. The artificial intelligence algorithm is used to calculate the best parameters to achieve the right waveform, intensity, electrical stimulation length and frequency (the reason for using genes is that the pain index of each patient is related to genes, blood oxygen concentration can be selected according to the severity of apnea, and CT images can determine the tongue electrical stimulation muscle training mode according to the respiratory condition); and for example, the input parameters are braces parameters, such as the mathematical model formed by the U-shaped tooth arrangement, such as polynomials, and the parameters that constitute the U-shaped will be adjusted, especially for the U-shaped opening coefficient, with a value of 0.1. The above different openings are input into the efficacy model to determine the best U-shaped opening to obtain the parameter combination ranking to determine the best parameter solution. According to the various parameters of the patient including genes, medical data, CT, ultrasound (genes for difficulty in tooth arrangement, respiratory cavity conditions, craniofacial position conditions) to understand the best U-shaped tooth arrangement (such as the shape parameters of braces) to achieve the most effective key prediction and different parameters. It can be noted that the efficacy prediction index of the best solution can be based on a single indicator or multiple indicators to be scored together, so as to obtain various solutions and their ranking recommendations to patients and caregivers.

In summary, the obstructive sleep apnea treatment plan recommendation system and method provided by the present invention can collect the patient's medical history and physiological data, and use these cases and physiological data as input to train the artificial intelligence module to generate a treatment effect prediction model, so that when all treatment solutions are input, all treatment solutions can be sorted to obtain the best treatment solution. Specifically, the treatment solution includes adjusting the patch position, adjusting the electrical stimulation parameters, adjusting the braces treatment parameters, etc. After different combinations of these treatment solutions are combined with the patient's physiological parameters and input into the treatment effect prediction model, the treatment effect prediction results of different combinations of treatment solutions will be obtained. The best treatment plan can be obtained, for example, by ranking different combinations of treatment solutions according to their AHI values, and evaluating the AHI values to obtain the best efficacy prediction result, so as to recommend the parameter combination of the treatment solution with the best efficacy.

Although the present invention has been disclosed as above by way of embodiments, they are not intended to limit the present invention. A person having ordinary knowledge in the technical field to which the present invention belongs may make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined in the attached patent application.

1: Obstructive sleep apnea treatment plan recommendation system 2: Artificial intelligence module 3: Processor 4: Storage device 5: Electrical stimulation unit 6: Physiological sensing module 7: Electrode battery S1, S3: Steps

FIG1 is a schematic diagram of a system for recommending treatment plans for obstructive sleep apnea provided in an embodiment of the present invention; FIG2 is a schematic diagram of the position of a patch inside the oral cavity provided in an embodiment of the present invention; FIG3 is a schematic diagram of the position of a patch outside the throat provided in an embodiment of the present invention; and FIG4 is a flow chart of a method for recommending treatment plans for obstructive sleep apnea provided in an embodiment of the present invention.

1: Obstructive sleep apnea treatment recommendation system

2: Artificial intelligence module

3: Processor

4: Storage device

5:Electrostimulation unit

6: Physiological sensing module

7: Electrode battery

Claims (12)

A system for recommending treatment plans for obstructive sleep apnea, comprising: an artificial intelligence module; and a processor connected to the artificial intelligence module; wherein the processor performs the following operations: inputting multiple patient physiological characteristic parameters, treatment solution parameters and efficacy judgment indicators into the artificial intelligence module to train an efficacy prediction model; and inputting multiple treatment solutions into the efficacy prediction model to generate a ranking result of the treatment solutions. The obstructive sleep apnea treatment plan recommendation system as described in claim 1, wherein the patient's physiological characteristic parameters include non-temporal characteristic parameters and temporal characteristic parameters, the non-temporal characteristic parameters at least include patient respiratory imaging parameters, patient medical history parameters, and patient genetic parameters, and the temporal characteristic parameters at least include patient wearable device physiological parameters, patient apnea-hypopnea index parameters, surface patch electrode electrical stimulation parameters, and braces parameters. A system for recommending treatment plans for obstructive sleep apnea as described in claim 2, wherein the electrical stimulation frequency in the electrical stimulation parameters is a low frequency input in the range of 20-100 Hz and at intervals above 1 Hz, and input in the range of 1k-10kHz and at intervals above 50 Hz, and is arranged and combined at different positions of the surface patch electrode. The obstructive sleep apnea treatment plan recommendation system as described in claim 2, wherein the processor performs data filling on the non-time series characteristic parameters and the time series characteristic parameters to generate time series parameters, and the data filling at least includes deleting outliers and filling missing values. The obstructive sleep apnea treatment plan recommendation system as described in claim 4, wherein the processor performs data integration on the time series parameters to sort them into a three-dimensional time series parameter, and the data integration at least includes segmenting the time series parameters and arranging the segmented time series parameters. The obstructive sleep apnea treatment plan recommendation system as described in claim 5, wherein the processor performs feature extraction on the three-dimensional time series parameters to generate the treatment efficacy judgment indicator, and the feature extraction at least includes time domain-frequency domain conversion and convolution operation on the three-dimensional time series parameters. A method for recommending treatment plans for obstructive sleep apnea is applicable to an obstructive sleep apnea treatment plan recommendation system having a processor, wherein the method for recommending treatment plans for obstructive sleep apnea is executed by the processor and includes: Inputting multiple patient physiological characteristic parameters, treatment solution parameters and efficacy judgment indicators into an artificial intelligence module to train an efficacy prediction model; and Inputting multiple treatment solutions into the efficacy prediction model to generate a ranking result of the treatment solutions. A method for recommending a treatment plan for obstructive sleep apnea as described in claim 7, wherein the patient's physiological characteristic parameters include non-temporal characteristic parameters and temporal characteristic parameters, the non-temporal characteristic parameters at least include patient respiratory imaging parameters, patient medical history parameters, and patient genetic parameters, and the temporal characteristic parameters at least include patient wearable device physiological parameters, patient apnea-hypopnea index parameters, surface patch electrode electrical stimulation parameters, and braces parameters. A method for recommending a treatment plan for obstructive sleep apnea as described in claim 8, wherein the electrical stimulation frequency in the electrical stimulation parameters is a low frequency input in the range of 20-100 Hz and at intervals above 1 Hz, and input in the range of 1k-10kHz and at intervals above 50 Hz, and is arranged and combined at different positions of the surface patch electrode. A method for recommending a treatment plan for obstructive sleep apnea as described in claim 8, wherein the processor performs data filling on the non-time series characteristic parameters and the time series characteristic parameters to generate time series parameters, and the data filling at least includes deleting outliers and filling in missing values. A method for recommending a treatment plan for obstructive sleep apnea as described in claim 10, wherein the processor performs data integration on the time series parameters to sort them into a three-dimensional time series parameter, and the data integration at least includes segmenting the time series parameters and arranging the segmented time series parameters. A method for recommending a treatment plan for obstructive sleep apnea as described in claim 11, wherein the processor performs feature extraction on the three-dimensional time series parameters to generate the therapeutic efficacy judgment indicator, and the feature extraction at least includes performing time domain-frequency domain conversion and convolution operation on the three-dimensional time series parameters.