Ai-assisted diagnosis-effect analyzing system and method thereof for ECG assessment

Abstract

An ECG AI-assisted diagnosis-effect analyzing method includes: detecting an abnormal personal ECG time interval in a plurality of personal ECG feature data to obtain an abnormal personal ECG section; processing an abnormal ECG assessment procedure to the abnormal personal ECG section to obtain an assessed abnormal personal ECG feature data; providing an unassessed personal time-interval ECG data; correspondingly connecting the unassessed personal time-interval ECG data with the assessed abnormal personal ECG feature data; operating an AI computing tool the unassessed personal time-interval ECG data and the assessed abnormal personal ECG feature data which are combined with a plurality of abnormal ECG feature data to obtain a plurality of result and analysis data of personal diagnosis assessment data.

Description

[Technical Field]

[0001] This invention relates to a system and method for analyzing the diagnostic effect of electrocardiogram (ECG) images or ECG assessments with AI assistance; in particular, it relates to a system and method for analyzing the diagnostic effect of ECG images or ECG assessments with AI assistance that can be applied to ECG images of the same person or different people generated on various timelines. [Previous Technology]

[0002] Conventional portable electrocardiogram (ECG) monitoring devices include, for example, the Republic of China Patent Publication No. TW-M352354, "Portable Electrocardiogram Monitoring Device," which discloses a portable ECG monitoring device. This portable ECG monitoring device includes at least two conductive patch groups, an internal circuit, a power supply module, and a data transmission module.

[0003] Continuing from the above, the conductive patch group of the aforementioned No. TW-M352354 generates several analog signals, and the internal circuit is used to receive the analog signals of the conductive patch group. The internal circuit has a power supply module, and the power supply module has a low noise voltage regulator circuit and a buck-boost circuit. The low noise voltage regulator circuit has a first ground plane, and the buck-boost circuit has a second ground plane. The first ground plane and the second ground plane are independent.

[0004] As above, the data transmission module of the aforementioned No. TW-M352354 has an operational amplifier circuit and a data processing circuit. The operational amplifier circuit has a third ground plane, and the data processing circuit has an independent fourth ground plane. The third ground plane and the fourth ground plane are independent.

[0005] As above, the aforementioned No. TW-M352354 utilizes the power supply module provided by the power supply module to generate a digital signal required for an electrocardiogram by passing the analog signal through the operational amplifier circuit to the data processing circuit, and only one grounding channel is left between the first grounding layer, the second grounding layer, the third grounding layer and the fourth grounding layer.

[0006] Another conventional invention related to portable electrocardiogram monitoring devices and their detection methods is, for example, US Patent No. US-6871089, "PORTABLE ECG MONITOR AND METHOD FOR ATRIAL FIBRILLATION DETECTION", which also discloses a portable electrocardiogram monitoring device and a method for detecting atrial fibrillation.

[0007] Another conventional invention relating to portable ECG monitoring devices with wireless communication interfaces for remote monitoring and their usage methods is, for example, US Patent No. US-6970737, "PORTABLE ECG DEVICE WITH WIRELESS COMMUNICATION INTERFACE TO REMOTELY MONITOR PATIENTS AND METHOD OF USE", which also discloses a portable ECG monitoring device with a wireless communication interface for remote monitoring and its usage methods.

[0008] Another commonly used portable ECG and blood pressure monitoring system and its detection method is, for example, US Patent No. US-5322069 "AMBULATORY ECG TRIGGERED BLOOD PRESSURE MONITORING SYSTEM AND METHOD THEREFOR", which also discloses a non-bedridden (portable) ECG-triggered blood pressure monitoring system and its detection method.

[0009] However, although the aforementioned Republic of China Patent Publication No. TW-M352354, US Patent No. US-6871089, US-6970737, and US-5322069 have disclosed various portable ECG monitoring devices and their detection methods, they are not suitable for accurately analyzing the diagnostic effects of ECG patterns or ECG assessments. Therefore, there is still a potential need for further improvement (e.g., checking angina, heart health, heart rate, or changes in cardiac potential).

[0010] Another conventionally used automotive electrocardiogram (ECG) monitoring system and its detection method is, for example, the Republic of China Patent Publication No. TW-201328667, "Automotive ECG Detection Device and Automotive ECG Monitoring System," which discloses an automotive ECG detection device. This automotive ECG detection device is suitable for mounting on a steering wheel of a vehicle.

[0011] As above, the aforementioned vehicle ECG detection device No. TW-201328667 is used to detect the cardiac activity status of one of the drivers of the vehicle and then transmit a detection result to a monitoring and warning device. The vehicle ECG detection device includes a steering wheel cover body, a measurement module, a processing module, and a signal output module.

[0012] As above, the steering wheel cover body of the aforementioned No. TW-201328667 is fitted onto the steering wheel, and the measurement module includes two sensing units. The sensing units are disposed in an insulated manner on an outer surface of the steering wheel cover body and cover the outer surface of the steering wheel cover body. The two sensing units are used to capture a first body surface potential signal and a second body surface potential signal of the driver, respectively.

[0013] As above, the processing module of the aforementioned No. TW-201328667 is used to receive the first body surface potential signal and the second body surface potential signal, so as to process the first body surface potential signal and the second body surface potential signal, and then output an electrocardiogram signal, and the signal output module is used to transmit the electrocardiogram signal to the monitoring and warning device.

[0014] Obviously, although the aforementioned Republic of China Patent Publication No. TW-201328667 has disclosed the vehicle ECG monitoring system and its detection method, it is not suitable for accurate diagnostic effect analysis when performing ECG graphics or ECG assessment. Therefore, there is still a potential need for further improvement (e.g., checking angina, heart health, heart rate or changes in cardiac potential).

[0015] Another conventionally used wireless transmission electrocardiogram (ECG) monitoring device is, for example, the invention patent No. TW-I274270 of the Republic of China, entitled "Wireless Transmission ECG Monitoring Structure," which discloses a wireless transmission ECG monitoring structure. This wireless transmission ECG monitoring structure includes at least three conductive fabric electrodes, a transmission conversion unit, a wireless transmission module, and a transmission antenna.

[0016] As above, the conductive fabric electrode of the aforementioned No. TW-I274270 includes a conductive fiber and a non-conductive fiber, and the transmission conversion unit is connected to the three conductive fabric electrodes through at least three conductive yarns, and the transmission conversion unit is used to receive an input signal and convert the input signal into a transmission signal.

[0017] As above, the aforementioned wireless transmitting module of No. TW-I274270 receives the transmitting signal and generates a corresponding wireless signal, and the transmitting antenna is used to receive the wireless signal and transmit the wireless signal to a wireless receiving device. In addition, an electromagnetic wave isolation layer, an insulating layer and a wire fixing layer are disposed from the outside to the inside on the outside of the conductive fabric electrode and the conductive yarn.

[0018] Another commonly used intelligent network electrocardiogram monitoring device, for example, is the invention patent No. TW-I315979 of the Republic of China, "Networked Intelligent Electrocardiogram Monitoring Device", which discloses a networked intelligent electrocardiogram monitoring device, and the networked intelligent electrocardiogram monitoring device is a single-chip electrocardiogram collection device system.

[0019] As above, the networked intelligent electrocardiogram monitoring device of the aforementioned No. TW-I315979 includes a heartbeat signal collection device, an analog circuit, a digital circuit, analysis software and a multimedia memory card. The analog circuit includes an amplifier circuit, a filter circuit and a compensation circuit. The digital circuit includes a single chip, and the single chip has an RS232 circuit and Serial Peripheral Interface (SPI) technology and a Multi Media Card (MMC).

[0020] As above, the heartbeat signal collection device of the aforementioned No. TW-I315979 is used to collect the heartbeat signal of a subject. The heartbeat signal is amplified by the amplification circuit and filtered for noise by the filtering circuit before entering the single chip for analog-to-digital conversion. The heartbeat signal is transmitted to an RS232 to TCP / IP device using the RS232 circuit of the single chip, so that the heartbeat signal can be transmitted through the RS232 to TCP / IP device and a network, and then received by a remote computer.

[0021] As above, the aforementioned No. TW-I315979 can use the serial peripheral interface technology of the single chip to store data in the multimedia memory card, and use a card reader to read the heartbeat signal to a computer, and use the analysis software to perform calculations in order to obtain an RR interval diagram of the heartbeat signal.

[0022] Obviously, although the aforementioned Republic of China Patent Publication No. TW-I274270 and No. TW-I315979 have disclosed the wireless transmission ECG monitoring device and the smart network ECG monitoring device, they are not suitable for accurate diagnostic effect analysis when performing ECG graphs or ECG assessments. Therefore, there is still a potential need for further improvement (e.g., checking angina, heart health, heart rate or changes in cardiac potential).

[0023] In short, the aforementioned Republic of China Patent Publication No. TW-M352354, US Patent Nos. US-6871089, US-6970737, US-5322069, Republic of China Patent Publication No. TW-201328667, Republic of China Patent Publication Nos. TW-I274270 and TW-I315979 are only for reference and explanation of the technical background of this invention and the current state of technological development, and are not intended to limit the scope of this invention.

[0024] In view of the above, in order to meet the above-mentioned needs, the present invention provides a system and method for analyzing the diagnostic effect of electrocardiogram (ECG) assessment with the assistance of artificial intelligence. The system detects at least one abnormal ECG time interval of an individual from several individual ECG characteristic data to obtain at least one abnormal ECG region of an individual, and performs an abnormal ECG assessment operation in the abnormal ECG region of the individual to obtain assessed abnormal ECG characteristic data of an individual. The system also provides at least one time interval of individual ECG data to be assessed, and associates the assessed abnormal ECG characteristic data of an individual with the time interval of individual ECG data to be assessed. The system combines the assessed abnormal ECG characteristic data of an individual and the time interval of individual ECG data to be assessed with several abnormal ECG characteristic parameters and performs artificial intelligence calculations using several artificial intelligence computing tools to obtain at least one ECG diagnostic assessment result data, thereby improving the shortcomings of conventional techniques that cannot accurately assess and analyze the diagnostic effect when using ECG graphics or ECG assessments. [Summary of the Invention]

[0025] The main objective of the preferred embodiment of the present invention is to provide a system and method for analyzing the diagnostic effect of electrocardiogram (ECG) assessment with the assistance of artificial intelligence. This system detects at least one abnormal ECG time interval in several individual ECG characteristic data sets to obtain at least one abnormal ECG region. An abnormal ECG assessment is then performed on this abnormal ECG region to obtain assessed individual abnormal ECG characteristic data. At least one time interval of individual ECG data to be assessed is also provided. The assessed individual abnormal ECG characteristic data is correlated with the time interval of individual ECG data to be assessed. Furthermore, the assessed individual abnormal ECG characteristic data and the time interval of individual ECG data to be assessed are combined with several abnormal ECG characteristic parameters and subjected to artificial intelligence calculations using several artificial intelligence computing tools to obtain at least one ECG diagnostic assessment result. Therefore, this system achieves the objective or effect of accurately analyzing the diagnostic effect when providing ECG graphics or ECG assessments.

[0026] To achieve the above objectives, the preferred embodiment of the electrocardiogram assessment system for diagnostic effectiveness analysis with artificial intelligence assistance of the present invention includes:

[0027] An input unit for inputting several personal electrocardiogram characteristic data;

[0028] An AI computing unit having several artificial intelligence computing tools, and the input unit being connected to the AI ​​computing unit; and

[0029] An output unit is connected to the AI ​​computing unit, and the output unit is used to output at least one electrocardiogram diagnostic assessment result data;

[0030] Among them, at least one abnormal ECG time interval of an individual is detected in several individual ECG characteristic data to obtain at least one abnormal ECG region of an individual, and an abnormal ECG assessment operation is performed in the abnormal ECG region of the individual to obtain an assessed individual abnormal ECG characteristic data. At least one time interval of individual ECG data to be assessed is provided, and the assessed individual abnormal ECG characteristic data is matched with the time interval of individual ECG data to be assessed. The assessed individual abnormal ECG characteristic data and the time interval of individual ECG data to be assessed are combined with several abnormal ECG characteristic parameters and several artificial intelligence computing tools to perform artificial intelligence calculations to obtain the individual ECG diagnostic assessment result data.

[0031] The preferred embodiment of the artificial intelligence computing tool of the present invention includes a distributed computing model, a grid computing model, a cloud computing model or an edge computing model.

[0032] In a preferred embodiment of the present invention, the artificial intelligence computing tool is selected from a large language model, a medium language model or a small language model.

[0033] In a preferred embodiment of the present invention, the artificial intelligence computing tool combines a Bayesian inference method or a dynamic Bayesian inference method.

[0034] In a preferred embodiment of the present invention, the individual electrocardiogram diagnostic assessment results data form at least one diagnostic effect analysis and arrangement order.

[0035] To achieve the above objectives, the preferred embodiment of the electrocardiogram assessment method for analyzing diagnostic effectiveness with artificial intelligence assistance includes:

[0036] Detect at least one abnormal ECG time interval from several individual ECG characteristic data in order to obtain at least one abnormal ECG region;

[0037] An abnormal electrocardiogram assessment is performed on the abnormal electrocardiogram region of the individual in order to obtain the abnormal electrocardiogram characteristic data of the assessed individual;

[0038] Provide at least one personal electrocardiogram (ECG) data period to be evaluated;

[0039] The assessed individual abnormal electrocardiogram characteristic data is mapped to the individual electrocardiogram data for the time interval to be assessed; and

[0040] The assessed individual abnormal electrocardiogram (ECG) characteristic data and the individual ECG data of the time interval to be assessed are combined with several abnormal ECG characteristic parameters and several artificial intelligence computing tools to perform artificial intelligence calculations in order to obtain at least one ECG diagnostic assessment result data.

[0041] In a preferred embodiment of the present invention, the individual abnormal electrocardiogram region includes an individual arrhythmia region, an individual angina pectoris region, an individual myocardial infarction region, or any combination thereof.

[0042] In a preferred embodiment of the present invention, the time interval of an individual's abnormal electrocardiogram includes an individual's daily routine, an individual's sleep time, an individual's exercise time, an individual's work time, an individual's outdoor activity time, or any combination thereof.

[0043] In a preferred embodiment of the present invention, the assessed individual abnormal electrocardiogram characteristic data includes seasonal characteristic data, climatic characteristic data, temperature characteristic data, humidity characteristic data, pollutant characteristic data, or any combination thereof.

[0044] In a preferred embodiment of the present invention, the individual electrocardiogram diagnostic assessment data corresponds to treatment recommendation data for at least one heart disease patient.

[0045] In a preferred embodiment of the present invention, the individual electrocardiogram diagnostic assessment data corresponds to the survival rate data of at least one heart disease patient. [Simplified Explanation of the Diagram]

[0097] Figure 1: Block diagram of the ECG assessment system for analyzing diagnostic effects with artificial intelligence assistance according to the first preferred embodiment of the present invention.

[0098] Figure 2: A schematic diagram of the ECG assessment system and method of the present invention, which uses an artificial intelligence-assisted diagnostic effect analysis system and method, employing a personal ECG feature database and an assessed personal abnormal ECG feature database.

[0099] Figure 2A: A schematic diagram of the ECG assessment system and method of the present invention using artificial intelligence-assisted diagnostic effect analysis system and method based on the ECG characteristic data of a first person.

[0100] Figure 2B: A schematic diagram of the ECG assessment system and method of the present invention using artificial intelligence-assisted diagnostic effect analysis system and method using a second ECG feature data.

[0101] Figure 3: Flowchart of the method for analyzing the diagnostic effect of electrocardiogram assessment with artificial intelligence assistance according to a preferred embodiment of the present invention.

[0102] Figure 4: Block diagram of the ECG assessment system for analyzing diagnostic effects with artificial intelligence assistance according to the second preferred embodiment of the present invention.

[0103] Figure 5: Block diagram of the ECG assessment system for analyzing diagnostic effects with artificial intelligence assistance according to the third preferred embodiment of the present invention.

Implementation Method

[0046] In order to fully understand the present invention, preferred embodiments will be described in detail below with reference to the accompanying drawings, and these are not intended to limit the present invention.

[0047] The ECG pattern or ECG assessment system and method with artificial intelligence assistance for diagnostic effect analysis according to the preferred embodiment of the present invention are suitable for various simple ECG pattern or ECG diagnostic assessment result data diagnostic assessment systems and methods or various special ECG pattern or ECG diagnostic assessment result data diagnostic assessment systems and methods, but are not intended to limit the scope of application of the present invention.

[0048] As above, the ECG graphics or ECG assessment system and method with artificial intelligence assistance for diagnostic effect analysis of the preferred embodiment of the present invention are suitable for application in various cardiac diagnostic analysis systems and methods, various cardiac examination systems and methods (e.g., examination of angina pectoris, heart health, heart rate or cardiac potential changes, etc.), various cardiac patient care systems and methods, or various cardiac postoperative care systems and methods, but are not intended to limit the scope of application of the present invention.

[0049] As above, the ECG graphics or ECG assessment system and method with artificial intelligence assistance for diagnostic effect analysis of the preferred embodiment of the present invention is suitable for various wired systems and methods with artificial intelligence assistance for interpretation or diagnostic effect analysis, and various wireless systems and methods with artificial intelligence assistance for interpretation or diagnostic effect analysis, but it is not intended to limit the scope of application of the present invention.

[0050] The preferred embodiment of the ECG graphics or ECG assessment system and method for analyzing diagnostic effects with artificial intelligence is executed in a computer-executable process step. It can be executed in various medical examination systems, various computing devices (e.g., smartphones) or various computer equipment, such as desktop computers, notebook computers, workstation computers, etc.

[0051] The ECG graphics or ECG assessment system and method for diagnostic effect analysis with artificial intelligence assistance according to the preferred embodiment of the present invention are applicable to various intranet systems or various network systems, such as the Internet, various local area networks (LANs) or various wireless LANs, but are not intended to limit the scope of application of the present invention.

[0052] Figure 1 shows a block diagram of the ECG assessment system for diagnostic effect analysis with artificial intelligence assistance according to a first preferred embodiment of the present invention. Referring to Figure 1, for example, the ECG assessment system for diagnostic effect analysis with artificial intelligence assistance according to a first preferred embodiment of the present invention includes an input unit 1, at least one or more individual ECG data 10 for time intervals to be evaluated, an AI computing unit 2, at least one or more artificial intelligence computing tools 20, an individual ECG feature database 21, an evaluated individual abnormal ECG feature database 22, and an output unit 3.

[0053] Please refer to Figure 1 again. For example, the input unit 1 can be selected as a predetermined input port device, and the predetermined input port device can be selected to be connected to a predetermined device (not shown). The predetermined device can be selected from an electrocardiogram (ECG) detection device, a device with similar ECG detection function, a storage device, a device with similar storage function, a personal ECG characteristic database, an assessed personal abnormal ECG characteristic database, or any combination thereof.

[0054] Please refer to Figure 1 again. For example, the predetermined input device can be selected from a wired communication input device, a wireless communication input device or any combination thereof, and the input unit 1 or the predetermined device can be selected to connect to an Internet or other network.

[0055] Please refer to Figure 1 again. For example, the AI ​​computing unit 2 can be selectively connected to the input unit 1 to input the personal electrocardiogram data 10 of the time interval to be evaluated or the personalized personal electrocardiogram data of the time interval to be evaluated. The AI ​​computing unit 2 can be selected from a device with AI computing function (e.g., AI smartphone, AI computer device or AI server computer device), and the AI ​​computing unit 2 can be selected to have at least one computing module (e.g., convolutional network processing module, classification module or other modules with similar processing functions).

[0056] Please refer to Figure 1 again. For example, the AI ​​computing unit 2 has at least one or more artificial intelligence computing tools 20, which can be selected from artificial intelligence computing tools provided by Meta Platforms, Google, Apple, Microsoft, DeepSeek or other companies, such as Llama3.2, mediatron, medllama2, mistral, phi3-medium, phi3-mini or other AI tools. The AI ​​computing unit 2 can be selectively connected to the personal electrocardiogram feature database 21 and the evaluated personal abnormal electrocardiogram feature database 22.

[0057] Referring again to Figure 1, for example, the artificial intelligence computing tool 20 can be selected from a large language model (LLM), a medium language model (MLM), or a small language model (SLM). In another preferred embodiment of the present invention, the artificial intelligence computing tool 20 can be selected to combine a Bayesian inference method or a dynamic Bayesian inference method.

[0058] Please refer to Figure 1 again. For example, the artificial intelligence computing tool 20 may include a distributed computing model, a grid computing model, a cloud computing model, or an edge computing model.

[0059] Figure 2 illustrates a preferred embodiment of the ECG assessment system and method for AI-assisted diagnostic effect analysis, which employs a personal ECG feature database and an assessed database of abnormal personal ECG features. Referring to Figures 1 and 2, for example, the personal ECG feature database 21 can be configured on a cloud server or a device with similar cloud server functionality, and the personal ECG feature database 21 is used to provide several personal ECG feature data 21a.

[0060] Please refer to Figures 1 and 2 again. For example, the personal electrocardiogram feature data 21a can be selected as a P waveform data, a Q waveform data, an R waveform data, an S waveform data, a T waveform data or any combination thereof, and each personal electrocardiogram feature data 21a has a predetermined characteristic time length (e.g., 30 seconds or other characteristic time lengths).

[0061] Please refer to Figures 1 and 2 again. For example, the assessed personal abnormal electrocardiogram feature database 22 can also be configured on a cloud server or a device with similar cloud server functions, and the assessed personal abnormal electrocardiogram feature database 22 can appropriately convert several personal electrocardiogram feature data 21a into several assessed personal abnormal electrocardiogram feature data 22a.

[0062] Please refer to Figures 1 and 2 again. For example, the assessed individual abnormal electrocardiogram feature data 22a may be selected to adopt a Q wave large drop model, a T wave large rise model, a T wave inversion model or any combination thereof, and each of the assessed individual abnormal electrocardiogram feature data 22a has a predetermined characteristic time length (e.g., 30 seconds or other characteristic time lengths).

[0063] Please refer to Figures 1 and 2 again. For example, the assessed individual abnormal electrocardiogram characteristic data 22a may include a P wave parameter, a PR segment parameter, a QRS complex parameter, an ST segment parameter, a T wave parameter, a QT interval parameter, or other parameters.

[0064] Please refer to Figure 1 again. For example, the output unit 3 can be selected as a predetermined output port device, and the predetermined output port device can be selected to be connected to a predetermined device (not shown) or the AI ​​computing unit 2. The predetermined device can be selected from a display device, a device with similar display functions, a printer device, a device with similar printing functions, a storage device, a device with similar storage functions, or any combination thereof.

[0065] Please refer to Figures 1 and 2 again. For example, the output unit 3 is used to control and output at least one electrocardiogram diagnostic assessment result data 30 and its related auxiliary data. The predetermined output port device of the output unit 3 can also be selected from a wired communication output device, a wireless communication output device or any combination thereof. The output unit 3 or the predetermined device can be selected to connect to an Internet or other network.

[0066] Figure 2A illustrates a preferred embodiment of the ECG assessment system and method for diagnostic effect analysis with artificial intelligence assistance, using a first person's ECG characteristic data, which corresponds to the personal ECG characteristic database in Figure 1 (as shown in Figure 1, 21). Referring again to Figure 2A, for example, the preferred embodiment of the ECG assessment system and method for diagnostic effect analysis with artificial intelligence assistance uses a first person's ECG characteristic data 211, and this first person's ECG characteristic data 211 can be selected from a first person's abnormal ECG time interval (e.g., 30 seconds or other time intervals).

[0067] Figure 2B illustrates a schematic diagram of the ECG assessment system and method for diagnostic effect analysis with artificial intelligence assistance according to a preferred embodiment of the present invention, which uses a second ECG feature data, corresponding to the first ECG feature data in Figure 2A. Referring again to Figure 2B, for example, the ECG assessment system and method for diagnostic effect analysis with artificial intelligence assistance according to a preferred embodiment of the present invention uses a second ECG feature data 212, and this second ECG feature data 212 can be selected from a second person's abnormal ECG time interval (e.g., 30 seconds or other time intervals).

[0068] Figure 3 illustrates a flowchart of a preferred embodiment of the method for analyzing the diagnostic effect of electrocardiogram (ECG) assessment using artificial intelligence (AI) assistance. Referring to Figures 1, 2, and 3, the preferred embodiment of the method for analyzing the diagnostic effect of ECG assessment using AI assistance includes step S1: First, for example, the AI ​​computing unit 2 or other units with similar computing functions automatically, semi-automatically, or manually detects at least one abnormal ECG time interval (i.e., a predetermined abnormal ECG time interval) from several individual ECG feature data 21a using appropriate technical means in an automatic, semi-automatic, or manual manner, in order to obtain at least one abnormal ECG region.

[0069] Please refer again to Figures 1, 2 and 3. For example, the personal electrocardiogram characteristic data 21a includes a static electrocardiogram segment (ECG fragment or EKG fragment), an exercise electrocardiogram segment, a 24-hour electrocardiogram segment (i.e., a 24-hour electrocardiogram segment), or any combination thereof.

[0070] Please refer to Figures 1, 2 and 3 again. For example, the time interval of an individual’s abnormal electrocardiogram includes a person’s daily routine, a person’s sleep time, a person’s exercise time, a person’s work time, a person’s outdoor activity time or any combination thereof.

[0071] Please refer again to Figures 1, 2 and 3. The method for analyzing the diagnostic effect of electrocardiogram assessment with artificial intelligence assistance in the preferred embodiment of the present invention includes step S2: Next, for example, the AI ​​computing unit 2 or other units with similar computing functions automatically, semi-automatically or manually use appropriate technical means to perform an abnormal electrocardiogram assessment operation on the individual's abnormal electrocardiogram region for the individual's electrocardiogram feature data 21a, so as to obtain at least one or more of the assessed individual abnormal electrocardiogram feature data 22a.

[0072] Please refer to Figures 1, 2 and 3 again. For example, the abnormal ECG region of the individual ECG characteristic data 21a or the assessed abnormal ECG characteristic data 22a may include an arrhythmia region, an angina region, a myocardial infarction region or any combination thereof.

[0073] Please refer to Figures 1, 2 and 3 again. For example, the personal electrocardiogram characteristic data 21a or the assessed personal abnormal electrocardiogram characteristic data 22a may include a seasonal characteristic data, a climatic characteristic data, a temperature characteristic data, a humidity characteristic data, a pollutant characteristic data or any combination thereof.

[0074] Please refer to Figures 1, 2 and 3 again. The method for analyzing the diagnostic effect of electrocardiogram assessment with artificial intelligence assistance in the preferred embodiment of the present invention includes step S3: Next, for example, the AI ​​computing unit 2 provides at least one time interval of personal electrocardiogram data 10 to be evaluated through the input unit 1 in an automatic, semi-automatic or manual manner using appropriate technical means.

[0075] Please refer to Figures 1, 2 and 3 again. For example, the personal electrocardiogram data 10 of the time interval to be evaluated can be selected from a portable electrocardiogram detection device, a wireless communication electrocardiogram detection device, a mobile electrocardiogram detection device, a smartwatch that supports electrocardiogram detection devices or any combination thereof. The personal abnormal electrocardiogram region of the personal electrocardiogram data 10 of the time interval to be evaluated can be selected to include a person's arrhythmia region, a person's angina region, a person's myocardial infarction region or any combination thereof.

[0076] Please refer to Figures 1, 2 and 3 again. The method for analyzing the diagnostic effect of electrocardiogram assessment with artificial intelligence assistance in the preferred embodiment of the present invention includes step S4: Next, for example, the AI ​​computing unit 2 uses appropriate technical means to automatically, semi-automatically or manually link the assessed individual abnormal electrocardiogram feature data 22a to the individual electrocardiogram data 10 in the time interval to be assessed, so as to perform a data time interval comparison processing operation, that is, time interval correlation comparison processing.

[0077] Please refer to Figures 1, 2 and 3 again. For example, the assessed personal abnormal electrocardiogram characteristic data 22a can be selected to correspond to the personal electrocardiogram data 10 of the time interval to be assessed for the same patient or test subject, or the assessed personal abnormal electrocardiogram characteristic data 22a can be selected to correspond to the personal electrocardiogram data 10 of the time interval to be assessed for different patients or test subjects.

[0078] Please refer again to Figures 1, 2 and 3. The method for analyzing the diagnostic effect of electrocardiogram assessment with artificial intelligence assistance in the preferred embodiment of the present invention includes step S5: Next, for example, the AI ​​computing unit 2 uses appropriate technical means to appropriately combine the assessed individual abnormal electrocardiogram feature data 22a and the individual electrocardiogram data 10 of the time interval to be assessed with several abnormal electrocardiogram feature parameters and several artificial intelligence computing tools 20 to perform artificial intelligence calculations (e.g., time interval artificial intelligence calculations) in an automatic, semi-automatic or manual manner, so as to obtain at least one electrocardiogram diagnostic assessment result data 30, and may choose to use several artificial intelligence computing tools 20 to perform artificial intelligence calculations in order to perform artificial intelligence-assisted diagnostic effect analysis.

[0079] Please refer again to Figures 1, 2 and 3. For example, in another preferred embodiment of the present invention, the assessed individual abnormal electrocardiogram feature data 22a and the individual electrocardiogram data 10 of the time interval to be assessed are appropriately combined with several abnormal electrocardiogram feature parameters and artificial intelligence computing tools 20 (e.g., AI tools provided by different companies) to perform artificial intelligence computing, so as to obtain several different individual electrocardiogram diagnostic assessment result data 30 and their differences, so as to further compare and evaluate and personalize the different individual electrocardiogram diagnostic assessment result data 30.

[0080] Please refer again to Figures 1, 2 and 3. For example, several of these artificial intelligence computing tools 20 may be selected from artificial intelligence computing tools provided by Meta Platforms, Google, Apple, Microsoft or other companies, such as Llama 3.2, mediatron, medllama 2, mistral, phi3-medium, phi3-mini or other AI tools, and the results produced by different artificial intelligence computing tools 20 will necessarily be different.

[0081] Please refer to Figures 1, 2 and 3 again. For example, several individual electrocardiogram diagnostic assessment results 30 may include a dyspnea diagnostic assessment result, a palpitation diagnostic assessment result, a weakness diagnostic assessment result, a dizziness diagnostic assessment result, or a syncope diagnostic assessment result, as well as differences generated by different artificial intelligence computing tools 20.

[0082] Please refer to Figures 1, 2 and 3 again. For example, the personal electrocardiogram data 10 or abnormal electrocardiogram characteristic parameters of the time interval to be evaluated can be selected to include a dyspnea electrocardiogram, a palpitation electrocardiogram, a weakness electrocardiogram, a dizziness electrocardiogram, a syncope electrocardiogram or any combination thereof.

[0083] Please refer to Figures 1, 2 and 3 again. For example, the individual ECG data 10 or abnormal ECG characteristic parameters of the time interval to be evaluated can be selected to include ECG data with a large drop in Q wave, ECG data with a large rise in T wave, ECG data with an inverted T wave or any combination thereof.

[0084] Please refer to Figures 1, 2 and 3 again. For example, the abnormal electrocardiogram characteristic parameters of the individual electrocardiogram diagnostic assessment results data can be selected from a family medical history parameter, a genetic parameter, a body response to drugs parameter or other relevant parameters.

[0085] Please refer to Figures 1, 2 and 3 again. For example, the abnormal electrocardiogram characteristic parameters of the individual electrocardiogram diagnostic assessment results data can be selected from an air pollution parameter, a water pollution parameter, a soil pollution parameter, a person's industry parameter or other relevant parameters.

[0086] Please refer to Figures 1, 2 and 3 again. For example, the abnormal electrocardiogram characteristic parameters of the individual electrocardiogram diagnostic assessment results data can be selected from a person's dietary habit parameters, and the individual's dietary habit parameters can be selected from a high-fat parameter (e.g., fried food, high-sugar food, meat food or other high-fat food), an alcohol consumption parameter or other related parameters.

[0087] Please refer to Figures 1, 2 and 3 again. For example, the abnormal electrocardiogram characteristic parameters of the individual electrocardiogram diagnostic assessment results data can be selected from a medical record parameter, an ethnic parameter, a genetic parameter, an environmental factor parameter, an individual's dietary habit parameter or an individual's lifestyle habit parameter.

[0088] Please refer again to Figures 1, 2 and 3. For example, in another preferred embodiment of the present invention, several of the individual electrocardiogram diagnostic assessment results 30 can be selected to form at least one diagnostic effect analysis arrangement order in order to provide at least one optimal diagnostic assessment result data, and the diagnostic effect analysis arrangement order corresponds to different several of the artificial intelligence computing tools 20.

[0089] Please refer again to Figures 1, 2 and 3. For example, in another preferred embodiment of the present invention, the artificial intelligence computing tool 20 may optionally be combined with a Bayesian inference method or a dynamic Bayesian inference method.

[0090] Figure 4 illustrates a block diagram of the ECG assessment diagnostic effect analysis system assisted by artificial intelligence according to the second preferred embodiment of the present invention. Referring to Figure 4, compared to the first embodiment, the ECG assessment diagnostic effect analysis system assisted by artificial intelligence according to the second preferred embodiment of the present invention can optionally integrate a personal ECG feature database 21 and an assessed personal abnormal ECG feature database 22 into a single-type personal ECG feature database 200.

[0091] Please refer to Figure 4 again. For example, the single-type personal electrocardiogram feature database 200 can be selected from a personal electrocardiogram feature cloud database or other cloud databases. Different personal electrocardiogram feature cloud databases can be selected to be applicable to different artificial intelligence computing tools 20, so as to perform various artificial intelligence-assisted diagnostic effect analysis.

[0092] Figure 5 illustrates a block diagram of the ECG assessment system for AI-assisted diagnostic effect analysis according to a third preferred embodiment of the present invention. Referring to Figure 5, compared to the first embodiment, the ECG assessment system for AI-assisted diagnostic effect analysis according to a third preferred embodiment of the present invention can selectively correspond several individual ECG diagnostic assessment results 30 to at least one heart disease patient's treatment recommendation data 31, at least one heart disease patient's survival rate data 32, or both, in order to perform AI-assisted diagnostic effect analysis.

[0093] Please refer to Figure 5 again. For example, there is at least one correlation between the heart disease patient treatment recommendation data 31 and the heart disease patient survival rate data 32, and the heart disease patient survival rate data 32 corresponds to the heart disease patient treatment recommendation data 31 with a relationship coefficient obtained by artificial intelligence calculation.

[0094] Please refer to Figure 5 again. For example, the ECG assessment system of the third preferred embodiment of the present invention, which uses artificial intelligence to assist in the diagnosis effect analysis, can select to correspond several individual ECG diagnostic assessment results 30 to at least one post-cardiac care data 33, and the survival rate data 32 of the heart disease patient to the treatment recommendation data 31 or the post-cardiac care data 33 of the heart disease patient have a relationship coefficient through artificial intelligence calculation.

[0095] The foregoing preferred embodiments are merely illustrative of the present invention and its technical features. The technology of these embodiments can still be implemented with various substantially equivalent modifications and / or substitutions. Therefore, the scope of the present invention shall be determined by the scope defined in the appended claims. The copyright of this case is limited to the use of the patent application in the Republic of China.

Claims

1. A system for analyzing the diagnostic effectiveness of electrocardiogram (ECG) assessment with artificial intelligence (AI) assistance, comprising: an input unit for inputting several individual ECG characteristic data; an AI computing unit having several AI computing tools, and the input unit being connected to the AI ​​computing unit; and an output unit connected to the AI ​​computing unit, and the output unit being used to output at least one individual ECG diagnostic assessment result data; wherein at least one individual abnormal ECG time interval is detected from the several individual ECG characteristic data to obtain at least one individual abnormal ECG region, and an abnormal ECG assessment is performed on the individual abnormal ECG region to obtain assessed individual abnormal ECG characteristic data; at least one individual ECG data for a time interval to be assessed is provided, and the assessed individual abnormal ECG characteristic data is correlated with the individual ECG data for the time interval to be assessed; and the assessed individual abnormal ECG characteristic data and the individual ECG data for the time interval to be assessed are combined with several abnormal ECG characteristic parameters and subjected to AI calculation using several AI computing tools to obtain the individual ECG diagnostic assessment result data.

2. The ECG assessment system for diagnosis effect analysis assisted by artificial intelligence as described in claim 1, wherein the artificial intelligence computing tool includes a distributed computing model, a grid computing model, a cloud computing model, or an edge computing model.

3. The ECG assessment system for diagnosis effect analysis assisted by artificial intelligence as described in claim 1, wherein the artificial intelligence computing tool is selected from a large language model, a medium language model or a small language model.

4. The ECG assessment system for diagnosis effect analysis with artificial intelligence assistance as described in claim 1, wherein the artificial intelligence computing tool combines a Bayesian inference method or a dynamic Bayesian inference method.

5. The ECG assessment system for diagnosis effect analysis with artificial intelligence assistance as described in claim 1, wherein the individual's ECG diagnostic assessment results data form at least one diagnostic effect analysis sorting order.

6. A method for analyzing the diagnostic effectiveness of electrocardiogram (ECG) assessment with artificial intelligence assistance, comprising the following steps: detecting at least one abnormal ECG time interval in several individuals' ECG characteristic data to obtain at least one abnormal ECG region in an individual; Steps: Perform an abnormal electrocardiogram assessment on the individual's abnormal electrocardiogram area to obtain the assessed abnormal electrocardiogram characteristic data; Steps: Provide personal electrocardiogram (ECG) data for at least one time interval to be evaluated; Steps: 1) The assessed individual abnormal ECG characteristic data is mapped to the individual ECG data for the time interval to be assessed; 2) The assessed individual abnormal ECG characteristic data and the individual ECG data for the time interval to be assessed are combined with several abnormal ECG characteristic parameters and several artificial intelligence computing tools to perform artificial intelligence calculations in order to obtain at least one ECG diagnostic assessment result data.

7. The method for analyzing the diagnostic effect of electrocardiogram assessment with artificial intelligence assistance as described in claim 6, wherein the abnormal electrocardiogram region of an individual includes an arrhythmia region, an angina pectoris region, a myocardial infarction region, or any combination thereof.

8. The method for analyzing the diagnostic effectiveness of electrocardiogram assessment with artificial intelligence assistance as described in claim 6, wherein the individual electrocardiogram diagnostic assessment results correspond to treatment recommendation data for at least one heart disease patient.

9. The method for analyzing the diagnostic effectiveness of electrocardiogram assessment with artificial intelligence assistance as described in claim 6, wherein the individual electrocardiogram diagnostic assessment data corresponds to the survival rate data of at least one heart disease patient.

10. The method for analyzing the diagnostic effect of electrocardiogram assessment with artificial intelligence assistance as described in claim 6, wherein the individual electrocardiogram diagnostic assessment results data form at least one diagnostic effect analysis sorting order, and the diagnostic effect analysis sorting order corresponds to different numbers of the artificial intelligence computing tools.

Images