Artificial intelligence based, wearable technology-supported telemedicine and remote patient management system for peripheral arterial diseases and method thereof

Abstract

The invention relates to an artificial intelligence based, wearable technology-supported telemedicine and remote patient management system for peripheral arterial diseases, which enables patient management by means of remote monitoring and early diagnosis in peripheral arterial diseases, and which ensures the reduction or prevention of complications, mortality and morbidity risks arising from peripheral arterial diseases.

Description

[0001] DESCRIPTION

[0002] ARTIFICIAL INTELLIGENCE BASED, WEARABLE TECHNOLOGY-SUPPORTED TELEMEDICINE AND REMOTE PATIENT MANAGEMENT SYSTEM FOR PERIPHERAL ARTERIAL DISEASES AND METHOD THEREOF Technical Field of the Invention

[0003] The invention relates to an artificial intelligence based, wearable technology-supported telemedicine and remote patient management system for peripheral arterial diseases, which enables patient management by means of remote monitoring and early diagnosis in peripheral arterial diseases, and which ensures the reduction or prevention of complications, mortality and morbidity risks arising from peripheral arterial diseases.

[0004] State of the Art

[0005] The anatomy and physiology of the normal circulatory system provide the transportation of oxygenated / oxygen-rich blood to tissues and cells through the arterial system, and the return of blood from tissues and cells back to the heart through the venous system. The function of the circulatory system is to transport oxygen to tissues with haemoglobin molecules through arteries by means of blood pumped from the heart, and at the same time to deliver heat and other metabolites to the periphery. The kinetic energy that enables the blood to reach the cells within the body is the contraction of the heart muscle during systole.

[0006] In peripheral occlusive arterial diseases, there occurs an insufficiency of regional blood circulation and of oxygen transport to cells at varying degrees. As a result of peripheral arterial disease, blood ejection volume and pressure in the distal circulation decrease. When there is a reduction or obstruction in blood flow in the arterial system due to atherosclerotic peripheral vascular disease, diabetes mellitus, hypertension, hyperlipidaemia, thromboangiitis obliterans (Buerger syndrome), systemic vasculitis, and other rheumatic and autoimmune diseases, stenosis (narrowing) or occlusion (blockage) occurs in arteries in the pathological process. In particular, in patients with atherosclerosis (arterial stiffness) and type 2 diabetes mellitus, it becomes impossible to ensure tissue nourishment since blood circulation is impaired due to the chronic, progressive disease in the blood vessels. In addition, in vascular lesions such as Buerger disease (thromboangiitis obliterans), scleroderma, Raynaud’s syndrome, dermatomyositis, and peripheral embolism, which disrupt local circulation, there also exist chronic pathological processes that reduce or obstruct blood flow within the arterial lumen. The reflection of decreased peripheral blood circulation manifests itself as the weakening or absence (non-palpable) of the pulse palpable over the artery, the decrease of peripheral oxygen saturation, and the occurrence of pain (claudication / rest pain) due to the inability to provide oxygen, as well as regional heat loss and coldness findings since heat cannot be transmitted to peripheral tissues. As tissue damage develops, the vicious cycle (acidosis, release of local chemotactic factors, increase in viscosity) progresses, and with vasoconstriction, blood flow decreases and tissue nourishment deteriorates further.

[0007] In addition to chronic vascular diseases, in acutely progressing pathologies such as thromboembolism and dissection, the sudden interruption of blood flow through the arterial lumen leads to tissue acidosis and tissue necrosis. In acute arterial occlusions, ensuring the continuation of blood flow by means of emergency surgical treatment within the first four-hour critical period is of vital importance. In chronic peripheral arterial diseases, the same pathophysiological mechanism is valid in the development of acute occlusion. In acute arterial occlusions, the clot or obstructive condition located within the artery must be removed and blood flow must be reestablished. The first four-hour critical period for the continuation of blood flow is extremely important to prevent cell death and tissue necrosis.

[0008] Long-term systematic serial follow-ups should be performed for symptomatic patients diagnosed with peripheral arterial disease. Routine follow-ups should also be carried out for asymptomatic patients. Mild symptoms may be precursors of serious arterial occlusions that can progress to severe neurological events and limb losses. In case of a change in the clinical condition, early or emergency intervention may be required. Evaluation should be made through patient history, physical examination, and other necessary investigations.

[0009] The invention disclosed in application number CN112185499A in the prior art provides a solution to the problems of inability to remotely and regularly monitor patients with hypertension. Said invention describes a blood pressure data management method, device, terminal device, and computer storage medium, which enable the automatic collection, evaluation, storage, and management of blood pressure data of hypertension patients, as well as the remote monitoring of patients’ dynamic blood pressure data and the updating of the electronic record file.

[0010] The invention disclosed in application number FR2498921A1 in the prior art refers to a device worn on the wrist like a watch, containing an adjustable audible alarm, for continuously displaying heart rate. An ultrasonic transceiver measures the Doppler frequency of continuous or pulsed ultrasonic waves reflected by the blood flowing in an artery or vein within the wrist.

[0011] In the prior art, there is no wearable technology product, software, or artificial intelligence study developed specifically for peripheral vascular diseases. By means of the application, an integrated solution specific to the field is provided. Although the invention is designed for peripheral vascular diseases, it can also be used for cerebrovascular diseases by means of enabling measurement from the neck.

[0012] As a result, due to the drawbacks mentioned above and the inadequacy of current solutions regarding the subject matter, a development in the relevant technical field has become necessary.

[0013] Brief Description and Aims of the Invention

[0014] The invention relates to an artificial intelligence based, wearable technology-supported telemedicine and remote patient management system for peripheral arterial diseases, which enables patient management by means of remote monitoring and early diagnosis in peripheral arterial diseases, and which ensures the reduction or prevention of complications, mortality and morbidity risks arising from peripheral arterial diseases.

[0015] The main aim of the invention is to reduce risk factors by means of early diagnosis using machine learning algorithms and to ensure that the processed data is presented to the physician and the patient.

[0016] Another aim of the invention is to ensure health education and awareness regarding the disease, to reduce cardiovascular risk factors, to maintain regular treatment, to detect the need for emergency or urgent vascular intervention and perform the necessary procedure, to prevent adverse events, to ensure the continuation of patient care outside the healthcare institution, to reduce the rate and duration of hospitalisation, to alleviate the healthcare economic burden of peripheral arterial diseases as a chronic condition, and to improve the quality of life of patients by means of remote monitoring methods in peripheral arterial diseases.

[0017] Another aim of the invention is to ensure early prevention and reduction or avoidance of the need for emergency service or intensive care, as well as the reduction of outpatient applications or hospitalisation periods, by means of the regular monitoring of physiological / physio-pathological data through telemedicine and remote patient management systems, and by determining complications before they occur.

[0018] Another aim of the invention is to enable the identification of risk factors through a database to be created by means of recording into the system the data measured with the wearable technology product of peripheral arterial patients.

[0019] Description of Drawings

[0020] Figure 1 is a drawing that shows the schematic view of the invention.

[0021] Reference Numbers

[0022] 110. Wearable device

[0023] 120. Application

[0024] 130. Server

[0025] 140. Database

[0026] Description of the Invention

[0027] The invention relates to an artificial intelligence based, wearable technology-supported telemedicine and remote patient management system for peripheral arterial diseases, which enables patient management by means of remote monitoring and early diagnosis in peripheral arterial diseases, and which ensures the reduction or prevention of complications, mortality and morbidity risks arising from peripheral arterial diseases. The wearable device (110) is a band-shaped device that can be worn on the wrist (over the radial artery), the ankle (over the tibial artery), and the neck (over both carotid arteries), and comprises a pulse sensor and an ultrasonic sensor (Doppler ultrasound) for the flow pattern. The pulse sensor converts the change in light reflected on photodiodes during heartbeats into an analogue signal. The signal movements form upper and lower points on the graph. Since each movement represents one heartbeat, the pulse is detected based on the peaks. In Doppler ultrasound, high-frequency sound waves are used to determine the width (diameter) of the vessels, the blood flow (flow rate), and the factors causing narrowing (plaques and other elements). In addition, the murmur that occurs with the hydrodynamic flow pattern in arteries and veins can also be heard. The wearable device (110), when wrapped around the wrist as a band, measures arterial pressure (blood pressure) through the radial artery with a sphygmomanometer. The wearable device (110) records rhythm (rhythmic NSR, arrhythmic) traces with an ECG sensor and skin temperature over the skin (upper / lower extremities) with an infrared temperature sensor.

[0028] In the invention, the pulse oximeter measures oxygen saturation and pulse rate from the nail bed of the index finger by means of infrared light.

[0029] The application (120) operates on a web and / or mobile device and comprises at least one interface that allows users to enter demographic information. The application (120) enables users to enter demographic data. The application (120) communicates with the wearable device (110) by means of wireless communication protocols and receives the data measured by the wearable device (110). In addition, through the application (120), users can enter their own measurements that are not measured by the wearable device (110). The application (120) allows the entry of additional data reflecting the condition of the patient’s disease, such as blood test results, medication reports, and similar information. The application (120) comprises an interface that enables physicians and patients to monitor, on a patient-specific basis, the data identified by the server (130). The application (120) comprises an interface that enables patients and physicians to communicate with each other through a shared interface.

[0030] The server (130) communicates with the application (120) and determines the standard pulse, peripheral oxygenation values, and extremity end-part temperature values that should correspond to the patient’s demographic information. The server (130) ensures that alerts are issued through the application (120) when the standard pulse, peripheral oxygenation values, and extremity end-part temperature exceed their lower or upper limits, in proportion to the degree / level of the exceedance. In this way, it will be possible to establish continuous and accurate communication between physician and patient based on objective data, to inform and raise awareness among patients, to reduce treatment and follow-up costs, and to improve patients’ quality of life. Data transfer to smartphones and / or portable smart devices and daily data storage can be performed through the application (120). The server (130) detects in real time the changes in the characteristics of the measurement parameters (pulse, blood pressure, peripheral oxygen saturation, body temperature) measured by the wearable device (110). The server (130) calculates the patient’s disease risk status by means of machine learning, based on the data entered through the application (120), the measurement data in the database (140) related to the diagnosis and follow-up of peripheral arterial diseases (blood pressure, pulse, blood oxygen level, temperature), and the data concerning various factors known to affect the disease (age, weight, smoking habits, medical history, etc.). The server (130), by means of machine learning models that learn from past data in the database (140), performs disease diagnosis and patient condition assessment using classification and regression algorithms (supervised learning), and detects patterns among the data using clustering algorithms (unsupervised learning). With the results obtained through machine learning, patients and physicians are informed through the application (120). The physician views critical conditions in the system, the patient is informed about their current condition, and all units are alerted to possible emergencies. The server (130) uses 80% of the data in the database (140) as training data and 20% as test data in the machine learning process. Since the data is collected for a specific purpose, it is expected to have minimal structural problems. However, as the measurements will be carried out remotely by patients, incorrect measurements or incorrect data entry may occur. Therefore, it is beneficial for the data to undergo certain processing steps. Outliers and extreme values are detected, duplicate observations are examined, missing values are completed considering the nature of the data, and extreme numerical values are normalised by means of the min-max method. For the detection and monitoring of the disease, the server (130) employs supervised learning algorithms for classification and regression, including the k-Nearest Neighbour algorithm based on distance-based classification (using Euclidean distance), the Naive Bayes and Logistic Regression algorithms based on statistical methods, and the C4.5 Decision Tree algorithm from decision tree methods. To understand the patterns (correlations) within the data, the k-Means algorithm, which is a partitioning method, is used as an unsupervised learning algorithm.

[0031] The database (140) ensures that the information entered through the application (120) is recorded. The database (140) records the disease risk status data calculated by the server (130). The database (140) ensures that the measurement data related to the diagnosis and monitoring of peripheral arterial diseases (blood pressure, pulse, blood oxygen level, temperature), as well as data regarding various factors known to affect the disease (age, weight, smoking habits, medical history, etc.), are recorded. The method of remotely detecting peripheral arterial diseases based on artificial intelligence comprises the following steps:

[0032] entry of users’ demographic information through the application (120), measurement of the person’s blood pressure, Doppler ultrasound, pulse, and skin temperature by means of the wearable device (110) worn on the wrist, ankle, and neck,

[0033] communication of the application (120) with the wearable device (110) by means of wireless communication protocols and the transmission of the data measured by the wearable device (110) to the server (130),

[0034] determination of the standard pulse, peripheral oxygenation values, and extremity end-part temperature information corresponding to the patients’ demographic information by the server (130), and the issuance of alerts through the application (120) in proportion to the degree / level of exceedance in case the lower or upper limits of the standard pulse, peripheral oxygenation values, and extremity end-part temperature are exceeded,

[0035] the calculation of the patient’s disease risk status by the server (130) by means of machine learning, based on the data entered through the application (120) and the measurement data in the database (140) related to the diagnosis and follow-up of peripheral arterial diseases (blood pressure, pulse, blood oxygen level, temperature), as well as the data on age, weight, smoking habits, and medical history known to affect the disease, and

[0036] the presentation of the data identified by the server (130) to physicians and patients through an interface in the application (120).

Claims

CLAIMS1. The system for the remote detection of peripheral arterial diseases based on artificial intelligence, comprising:- at least one wearable device (110) that can be worn on the wrist, ankle, and neck, and comprises a pulse sensor, an ultrasonic sensor for the flow pattern, a sphygmomanometer for measuring blood pressure through the radial artery, an ECG sensor, and an infrared temperature sensor for measuring skin temperature,- at least one application (120) that operates on a web and / or mobile device, enables users to enter demographic information, comprises an interface allowing physicians and patients to monitor the data identified by the server (130), communicates with the wearable device (110) by means of wireless communication protocols, and ensures the transmission of the data measured by the wearable device (110) to the server (130),- at least one server (130) that communicates with the application (120), determines the standard pulse, peripheral oxygenation values, and extremity end-part temperature information corresponding to the patients’ demographic information, ensures the issuance of alerts through the application (120) in proportion to the degree / level of exceedance in case the lower or upper limits of the standard pulse, peripheral oxygenation values, and extremity end-part temperature are exceeded, calculates the patient’s disease risk status by means of machine learning based on the data entered through the application (120) and the measurement data in the database (140) related to the diagnosis and monitoring of peripheral arterial diseases (blood pressure, pulse, blood oxygen level, temperature), as well as on the data concerning age, weight, smoking habits, and medical history known to affect the disease, and detects disease diagnosis and patient condition by means of classification and regression algorithms through machine learning models that learn from past data in the database (140), and detects patterns among the data by means of clustering algorithms, and- at least one database (140) that ensures the recording of the information entered through the application (120) and the recording of the measurement data related to the diagnosis and monitoring of peripheral arterial diseases (blood pressure, pulse, blood oxygen level, temperature), as well as the data concerning age, weight, smoking habits, and medical history known to affect the disease.

2. The method for the remote detection of peripheral arterial diseases based on artificial intelligence, comprising the process steps of:- entry of users’ demographic information through the application (120),- measuring the person’s blood pressure, Doppler ultrasound, pulse, and skin temperature by means of the wearable device (110) worn on the wrist, ankle, and neck,- communication of the application (120) with the wearable device (110) by means of wireless communication protocols and the transmission of the data measured by the wearable device (110) to the server (130),- determination of the standard pulse, peripheral oxygenation values, and extremity end-part temperature information corresponding to the patients’ demographic information by the server (130), and the issuance of alerts through the application (120) in proportion to the degree / level of exceedance in case the lower or upper limits of the standard pulse, peripheral oxygenation values, and extremity end-part temperature are exceeded,- the calculation of the patient’s disease risk status by the server (130) by means of machine learning, based on the data entered through the application (120) and the measurement data in the database (140) related to the diagnosis and follow-up of peripheral arterial diseases (blood pressure, pulse, blood oxygen level, temperature), as well as the data on age, weight, smoking habits, and medical history known to affect the disease, and- the presentation of the data identified by the server (130) to physicians and patients through an interface in the application (120).

3. The system for the remote detection of peripheral arterial diseases based on artificial intelligence according to claim 1, comprising a database (140) that records the disease risk status data calculated by the server (130).

4. The system for the remote detection of peripheral arterial diseases based on artificial intelligence according to claim 1, comprising a server (130) that detects, in real time, the changes in the characteristics of the measurement parameters of pulse, blood pressure, peripheral oxygen saturation, and body temperature measured by the wearable device (110).

5. The system for the remote detection of peripheral arterial diseases based on artificial intelligence according to claim 1, comprising a server (130) that detects outliers and extreme values in the data contained in the database (140), completes missing values, and normalises extreme numerical values by means of the minimum-maximum method.

6. The system for the remote detection of peripheral arterial diseases based on artificial intelligence according to claim 1, comprising a server (130) that uses supervised learning algorithms for classification and regression, including the k- Nearest Neighbour algorithm from distance-based classification algorithms, the Naive Bayes and Logistic Regression algorithms from statistical algorithms, and the C4.5 Decision Tree algorithm from decision tree algorithms, for the detection and monitoring of the disease.

7. The system for the remote detection of peripheral arterial diseases based on artificial intelligence according to claim 1, comprising an application (120) having an interface that enables patients and physicians to communicate with each other through an interface.

8. The system for the remote detection of peripheral arterial diseases based on artificial intelligence according to claim 1, comprising an application (120) that enables the entry of additional data reflecting the condition of the patient’s disease, such as blood test results and medication reports.

9. The system for the remote detection of peripheral arterial diseases based on artificial intelligence according to claim 1, wherein the wearable device (110) is in the form of a band.

Images