Atrial fibrillation rotor high-density analysis system and control method

Abstract

The invention relates to the technical field of atrial fibrillation, in particular to an atrial fibrillation rotor high-density analysis system and a control method. The atrial fibrillation rotor high-density analysis system comprises an embedded data acquisition system and an upper computer, and is characterized in that the embedded data acquisition system processes acquired data and uploads the processed data to the upper computer for analysis. Compared with the prior art, the atrial fibrillation rotor high-density analysis system and the control method provided by the invention are used for accurately mapping and recording electric signals in a cardiac cavity, deeply learning a special software algorithm through an AI loaded machine, and analyzing the atrial fibrillation rotor by adopting professional equipment and technologies; therefore, the obtained real-time electrocardiogram data extraction can evaluate direct information and indexes of ablation treatment sites for clinicians, and the treatment accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of atrial fibrillation technology, specifically to a high-density analysis system and control method for atrial fibrillation rotors. Background Technology

[0002] Atrial fibrillation is the most common arrhythmia, causing 287,000 deaths worldwide in 2017, accounting for 0.51% of the global cumulative mortality burden. Patients with atrial fibrillation have a 4-5 times higher risk of ischemic stroke than those without. Heart failure combined with atrial fibrillation increases all-cause mortality by 8% and the risk of cardiovascular death by 10%.

[0003] Atrial fibrillation (AF) can be initially paroxysmal, then persistent, and eventually permanent. In China, the prevalence of persistent AF is higher than that of paroxysmal AF, and it poses the greatest risk. Drug treatment has limited effectiveness and a high recurrence rate. The latest ESC guidelines have upgraded catheter ablation for drug-resistant PsAF to a Class I recommendation, but the proportion of catheter ablation for persistent AF is far lower than for paroxysmal AF. Furthermore, the widely used pulmonary vein electrical isolation for paroxysmal AF is far from sufficient for ablation treatment of persistent AF. Surgical treatment is also ineffective, with a success rate of only 43% to 67%. Experts and scholars have explored the pathogenesis of persistent AF for decades but have yet to achieve a significant breakthrough. Improving the surgical success rate for patients with persistent AF is a major clinical problem that urgently needs to be addressed.

[0004] The atrial fibrillation rotor, a major breakthrough in the study of atrial fibrillation mechanisms, provides a new perspective for its treatment. The concept of the rotor was first discovered in 1998 by researchers such as Skanes using optical mapping techniques. During atrial fibrillation, the electrical activity of the atria is not disordered, but exhibits a relatively regular and rapid microreentry phenomenon, figuratively termed a "rotor." The rotor is essentially a special type of functional reentry, manifesting as a spiral wave. It is mobile, rotating around a core with an uncertain circumference. During the rotor wavefront motion, if it encounters an anatomical barrier, it splits and generates multiple sub-spiral waves. This periodicity of electrical activity is significant in both time and space and is a crucial mechanism for maintaining atrial fibrillation. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention provides a high-density analysis system and control method for atrial fibrillation rotors. This system is used to accurately map and record intracardiac electrical signals. Through a special machine deep learning software algorithm loaded with AI, and using professional equipment and technology, atrial fibrillation rotor analysis is performed. The resulting real-time electrocardiogram data extraction can provide clinicians with direct information and indicators for evaluating ablation treatment sites, thereby improving the accuracy of treatment.

[0006] To achieve the above objectives, a high-density analysis system and control method for atrial fibrillation rotor are designed, including an embedded data acquisition system and a host computer. The embedded data acquisition system processes the acquired data and then uploads it to the host computer for analysis. The control flow of the embedded data acquisition system is as follows: S1-1, Power on and start; S1-2, Initialize the network interface required for subsequent data transmission; S1-3, configures the registers via the control bus to prepare for data output; S1-4, load the FPGA bitstream and enable the FPGA logic required for the interface and data buffer; S1-5 sends a command or signal to the register, causing it to actively convert analog signals into digital data and output them. S1-6, the FPGA logic directly receives high-speed digital data from the register; S1-7 writes digital data into the designated MCU FIFO buffer for asynchronous transmission across PS-PL boundaries; S1-8, the MCU actively reads or reads digital data from the FIFO buffer via an interrupt; S1-9, The MCU checks the current status of the network connection and determines whether the network connection is established. If the network connection is established, proceed to step S1-10; otherwise, return to step S1-6. S1-10, the MCU formats data packets and transmits them to the host computer via the network; S1-11, After data transmission, jump to step S1-6 to collect the next piece of data in a loop; The control flow of the host computer is as follows: S2-1, Start the application in the host computer and initialize the main window and core modules; S2-2, loads or allows users to configure connection parameters and data processing options; S2-3, according to the configuration, attempts to establish a connection with the embedded data acquisition system through the network communication module; S2-4, Determine if the network connection is normal. If yes, continue to step S2-5; otherwise, jump to step S2-2. S2-5: Receives data packets from the embedded data acquisition system; the digital data is parsed and transmitted to the data processing module. S2-6, Determine whether entropy calculation is required. If yes, continue to step S2-7; otherwise, jump to step S2-11. S2-7, the data processing module buffers the continuously received digital data until the sample size required for the 2.5s time window is reached; S2-8, using the accumulated 2.5s of data, executes the entropy calculation process; S2-9 displays the calculated entropy value on the UI interface and updates the relevant trend charts or analysis graphs; S2-10, return to step S2-5, and process the next batch of incoming data in a loop; S2-11, real-time display of EMG, the data is directly used to update the waveform, and then return to step S2-5 to process the next batch of incoming data in a loop.

[0007] The register mentioned is the ADC1298 register, and the interface is the ADC interface.

[0008] The host computer is a PC based on Qt5 C++ application software; the core modules include a GUI / UI module, a network communication module, and a data processing module.

[0009] In step S2-2, the parameters include IP address and port number; the data processing options include whether to start entropy calculation.

[0010] The embedded data acquisition system is compatible with different brands of electrophysiological multichannel recorders that have analog electrocardiogram signal output.

[0011] The embedded data acquisition system includes 24 independent amplifiers, which can be used for bipolar 24-lead or unipolar 24-lead ECG signal input; and a 24-bit analog-to-digital converter (ADC) that can accurately display atrial fibrillation waveform details.

[0012] The embedded data acquisition system communicates with the interface control module via an isolator, connects to the host computer network via a network card module, and communicates with the display via the interface control module; the embedded data acquisition system and the interface control module are connected to the power module via an isolated power supply module.

[0013] Compared with existing technologies, this invention provides a high-density analysis system and control method for atrial fibrillation rotors, which is used to accurately map and record intracardiac electrical signals. Through a special machine deep learning software algorithm loaded with AI, and using professional equipment and technology, atrial fibrillation rotor analysis is performed. The resulting real-time electrocardiogram data extraction can provide clinicians with direct information and indicators for evaluating ablation treatment sites, thereby improving the accuracy of treatment. Attached Figure Description

[0014] Figure 1 This is a flowchart of the embedded data acquisition system of the present invention.

[0015] Figure 2 This is a flowchart of the host computer data processing of the present invention.

[0016] Figure 3This is a schematic diagram of the medical system framework for which the present invention is applied. Detailed Implementation

[0017] The present invention will now be further described with reference to the accompanying drawings.

[0018] like Figure 1 , Figure 2 As shown, a high-density analysis system and control method for atrial fibrillation rotor includes an embedded data acquisition system and a host computer. The embedded data acquisition system processes the acquired data and uploads it to the host computer for analysis. The control flow of the embedded data acquisition system is as follows: S1-1, Power on and start; S1-2, Initialize the network interface required for subsequent data transmission; S1-3, configures the registers via the control bus to prepare for data output; S1-4, load the FPGA bitstream and enable the FPGA logic required for the interface and data buffer; S1-5 sends a command or signal to the register, causing it to actively convert analog signals into digital data and output them. S1-6, the FPGA logic directly receives high-speed digital data from the register; S1-7 writes digital data into the designated MCU FIFO buffer for asynchronous transmission across PS-PL boundaries; S1-8, the MCU actively reads or reads digital data from the FIFO buffer via an interrupt; S1-9, The MCU checks the current status of the network connection and determines whether the network connection is established. If the network connection is established, proceed to step S1-10; otherwise, return to step S1-6. S1-10, the MCU formats data packets and transmits them to the host computer via the network; S1-11, After data transmission, jump to step S1-6 to collect the next piece of data in a loop.

[0019] Design specifications for an embedded data acquisition and transmission system based on Zynq System-on-Chip (SoC) and ADC1298 analog-to-digital converter. The main functions of the system are to initialize the hardware, acquire ADC data through the FPGA, buffer the data in a FIFO accessible by the MCU, and transmit the acquired data to the host computer via a network connection.

[0020] The system architecture is based on the Xilinx Zynq platform, which integrates the processing system (MCU / PS) and programmable logic (FPGA / PL).

[0021] The use of an MCU FIFO is crucial for decoupling the high-speed continuous data acquisition from the FPGA from the potentially variable-rate network data transmission managed by the MCU. The FIFO depth must be sufficient to prevent buffer overflows during temporary network bottlenecks or processing delays.

[0022] Network status checks (i.e., steps S1-9 above) provide a basic recovery mechanism. If a network failure occurs, the system will prioritize continuous acquisition and retry transmission when the next data block arrives, rather than completely halting data acquisition.

[0023] The control flow of the host computer is as follows: S2-1, Start the application in the host computer and initialize the main window and core modules; S2-2, loads or allows users to configure connection parameters and data processing options; S2-3, according to the configuration, attempts to establish a connection with the embedded data acquisition system through the network communication module; S2-4, Determine if the network connection is normal. If yes, continue to step S2-5; otherwise, jump to step S2-2. S2-5: Receives data packets from the embedded data acquisition system; the digital data is parsed and transmitted to the data processing module. S2-6, Determine whether entropy calculation is required. If yes, continue to step S2-7; otherwise, jump to step S2-11. S2-7, the data processing module buffers the continuously received digital data until the sample size required for the 2.5s time window is reached; S2-8, using the accumulated 2.5s of data, executes the entropy calculation process; S2-9 displays the calculated entropy value on the UI interface and updates the relevant trend charts or analysis graphs; S2-10, return to step S2-5, and process the next batch of incoming data in a loop; S2-11, real-time display of EMG, the data is directly used to update the waveform, and then return to step S2-5 to process the next batch of incoming data in a loop.

[0024] The design and operation flow of the host computer application software (based on Qt5 C++) are described. This software is mainly responsible for establishing a network connection with the embedded data acquisition system, receiving continuous raw ADC data streams, and performing real-time EMG (electromyography) display or entropy analysis according to user selection, and displaying the results graphically.

[0025] The Qt5 framework is used, leveraging its signal / slot mechanism to decouple modules and enable event-driven processing. To ensure that data reception and computation do not block the user interface (UI), network communication and data processing will employ multithreading or Qt asynchronous mechanisms.

[0026] To prevent UI freezing due to network I / O or complex entropy calculations, network reception and data processing must run in a separate thread from the main thread. Qt's QThread or QtConcurrent can be used to achieve this.

[0027] In the entropy calculation path, the cumulative 2.5s of data is critical. This means the system must maintain a sliding or fixed buffer. If a sliding window is used, the starting point of the calculation every 2.5s needs to be precisely located to ensure the continuity and real-time nature of the calculation results.

[0028] The register is an ADC1298 register, and the interface is an ADC interface.

[0029] The host computer is a PC based on Qt5 C++ application software; the core modules include a GUI / UI module, a network communication module, and a data processing module.

[0030] In step S2-2, the parameters include IP address and port number; the data processing options include whether to start entropy calculation.

[0031] The characteristics of an embedded data acquisition system are as follows: 1. Compatible with various brands of electrophysiological multichannel recorders that have analog ECG signal output; 2. 3D mapping system: These compatible devices need to be equipped with a 1.5mm ECG output tail cable, with a maximum of 48 channels and methods; 3. 24 independent amplifiers, which can be used for ECG signal input in either bipolar 24-lead or unipolar 24-lead mode; 4. A 24-bit analog-to-digital converter (ADC) can accurately display the details of atrial fibrillation waveforms; 5. It can achieve a sampling rate of 8K / second for all 24 leads, and fully record the propagation characteristics of atrial fibrillation.

[0032] like Figure 3 As shown, the embedded data acquisition system communicates with the interface control module through an isolator, connects to the host computer network via a network card module, and communicates with the display through the interface control module; the embedded data acquisition system and the interface control module are connected to the power module through an isolated power supply module.

[0033] The features of the system of this invention are as follows: 1. Based on the algorithm, the machine will display the characteristics of the driving rotor on the UI: focal type and rotary type; 2. Multiscale entropy MSE scale: The spatiotemporal dispersion of each electrode is represented by entropy values; 3. Display the entropy value on the electrode: Identify the entropy value on each electrode to provide an intuitive distribution of dispersion for clinical use; 4. Display the average entropy of the site of interest (ROI): Based on clinical experience, only sites with entropy values ​​reaching a certain average value can be selected as ROIs; 5. High-entropy electrode combination: Clinical experience shows that a site with at least three consecutive high-entropy electrodes is a valuable ROI; 6. Real-time EGM period: The calibration instrument analyzes the average period and shortest period of the real-time EGM as AI learning factors; 7. Real-time EGM: A single 2.5-second EGM will be used as data for calculating the entropy value of the mapping site. The amplitude, interval and entropy values ​​of multiple EGMs displayed by 20 independent unipolar mappings will be used as data to determine the direction of atrial fibrillation wave conduction and to determine the driving type. 8. ROI Site Recording: Sites identified as ROIs through calculation and clinical experience will be recorded by the machine and used as learning data for AI.

[0034] This invention aims to utilize existing clinical data and experience to create a new method for continuous atrial fibrillation ablation therapy. It is used to accurately map and record intracardiac electrical signals, and through a special machine deep learning software algorithm loaded with AI, atrial fibrillation rotor analysis is performed using professional equipment and technology. The acquired real-time ECG data is used to extract direct information and indicators that can be used by clinicians to evaluate ablation treatment sites.

Claims

1. A high-density analysis system and control method for atrial fibrillation rotors, comprising an embedded data acquisition system, an upper computer, characterized in that: The embedded data acquisition system uploads the processed data to the host computer for analysis; The control flow of the embedded data acquisition system is as follows: S1-1, power on; S1-2, initialize the network interface required for subsequent data transmission; S1-3, configure the register through the control bus to prepare for data output; S1-4, load the FPGA bit stream and enable the FPGA logic required for the interface and data buffering; S1-5, send a command or signal to the register to start actively converting analog signals into digital data and outputting them; S1-6, the FPGA logic directly receives high-speed digital data from the register; S1-7, write the digital data into the designated MCU FIFO buffer for asynchronous transmission across the PS-PL boundary; S1-8, the MCU actively reads or reads through an interrupt the digital data in the FIFO buffer; S1-9, the MCU checks the current state of the network connection and determines whether the network connection is established, and if so, proceeds to step S1-10; otherwise, returns to step S1-6; S1-10, the MCU formats the data packet and transmits it to the host computer through the network; S1-11, after data transmission, jump to step S1-6 to collect the next block of data; The control flow of the host computer is as follows: S2-1, start the application program in the host computer, initialize the main window and core module; S2-2, load or allow the user to configure connection parameters and data processing options; S2-3, attempt to establish a connection with the embedded data acquisition system through the network communication module according to the configuration; S2-4, determine whether the network connection is normal, and if so, proceed to step S2-5, otherwise jump to step S2-2; S2-5, receive the data packet from the embedded data acquisition system, and the digital data is parsed and passed to the data processing module; S2-6, determine whether entropy calculation is required, and if so, proceed to step S2-7, otherwise jump to step S2-11; S2-7, the data processing module buffers the continuously received digital data until the required sample size for a 2.5s time window is reached; S2-8, use the accumulated 2.5s data to perform the entropy calculation process; S2-9, display the calculated entropy value on the UI interface and update the relevant trend chart or analysis graph; S2-10, return to step S2-5 to process the next batch of incoming data; S2-11, display the EMG in real time, and the data is directly used for waveform graph updating, then return to step S2-5 to process the next batch of incoming data.

2. The high-density analysis system and control method of a fibrillation rotor according to claim 1, characterized by: The register is an ADC1298 register, and the interface is an ADC interface.

3. The high-density analysis system and control method of a fibrillation rotor according to claim 1, characterized by: The host computer is a PC based on Qt5 C++ application software; the core module includes a GUI / UI module, a network communication module, and a data processing module.

4. The high-density analysis system and control method of a fibrillation rotor according to claim 1, characterized by: In step S2-2, the parameters include IP address and port number; the data processing options include whether to start entropy calculation.

5. The high-density analysis system and control method of a fibrillation rotor according to claim 1, characterized by: The embedded data acquisition system is compatible with different brands of electrophysiological multi-channel recorders that have analog ECG signal output.

6. The high density analysis system and control method of a fibrillation rotor according to claim 1, characterized by: The embedded data acquisition system comprises 24 independent amplifiers, and the input mode of the electrocardio signal can be 24 leads of bipolar or 24 leads of monopolar; 24-bit analog-digital conversion ADCs can accurately display the details of the atrial fibrillation waveform.

7. The high-density analysis system and control method of a rotor of atrial fibrillation according to claim 1 or 2 or 3, characterized in that: The embedded data acquisition system is in communication connection with the interface control module through an isolator, is in network connection with the upper computer via a network card module, and is in communication connection with a display; the embedded data acquisition system and the interface control module are respectively connected with a power module through an isolation power module.

Images