Implantable myocardial contractility regulation qrs-wave detection module and apparatus

Abstract

The application provides a QRS wave detection module and device for implantable myocardial contractility adjustment, and belongs to the technical field of implantable medical devices. Full fixed-point recursion calculation is adopted, including: performing recursive baseline correction processing on the collected electrocardio signals to filter out baseline drift; calculating the slope characteristics of the baseline-corrected electrocardio signals, and dynamically updating the amplitude detection threshold based on the peak value of the historical true QRS wave; when the slope characteristics and the signal amplitude both exceed the corresponding threshold, the current sampling point is determined as a QRS wave candidate point; finally, the candidate point is verified based on the regularity of the time interval between the candidate point and the last confirmed true QRS wave, so as to confirm the true QRS wave. The application has extremely low power consumption, strong anti-interference performance and high adaptability, and can guarantee the real-time performance and effectiveness of treatment.

Description

Technical Field

[0001] This invention relates to the field of implantable medical device technology, specifically to an implantable QRS wave detection module and device for myocardial contractility regulation. Background Technology

[0002] CCM (Contraction Modulation) is a technique for treating heart failure by applying electrical stimulation during the absolute refractory period of the myocardium. Its efficacy is highly dependent on the real-time and accurate detection of the QRS complex in the electrocardiogram (ECG) signal, because the stimulation pulse must be delivered within a very short time window after the QRS complex is detected.

[0003] Several QRS wave detection algorithms exist in the current technology. For example, some methods use digital filters to process ECG signals for baseline correction and combine this with adaptive thresholding. When the signal's amplitude or slope meets preset conditions, the signal is identified as a heartbeat. However, these methods typically involve computationally complex operations such as filtering and squaring, resulting in high power consumption and making it difficult to meet the stringent power requirements of some implantable devices. Other more complex algorithms, such as those based on wavelet transform or machine learning, may improve detection accuracy in certain scenarios, but their enormous computational load and power consumption make them unsuitable for emerging battery-free implantable medical devices that rely solely on real-time wireless power.

[0004] These emerging implantable devices have extremely stringent limitations on power consumption and computing resources. Existing algorithms cannot effectively resist strong interferences such as baseline drift and electromyographic artifacts in the body environment while meeting their ultra-low power consumption requirements, nor can they be deeply integrated with the precise timing requirements of myocardial contractility modulation therapy. Therefore, when applied to battery-free, wirelessly powered implantable devices, existing technologies face technical challenges such as excessive power consumption and computational complexity that prevents stable real-time operation. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the purpose of this invention is to provide an implantable QRS wave detection module and device for myocardial contractility regulation.

[0006] According to the present invention, an implantable QRS wave detection module for myocardial contractility modulation includes: an electrocardiogram acquisition electrode, a preamplifier and filter circuit, an analog-to-digital converter, and a main control microcontroller. The ECG acquisition electrodes input the acquired analog ECG signal into the preamplifier and filter circuit for filtering and amplification. The filtered and amplified analog ECG signal is then input into the analog-to-digital converter for analog-to-digital conversion to obtain a digitized ECG signal. The main control microcontroller is configured to perform the following tasks using fully fixed-point recursive calculations: During the initialization phase, a preset initial amplitude detection threshold is used; The digitized electrocardiogram (ECG) signal is subjected to recursive baseline correction to filter out baseline drift and obtain the baseline-corrected ECG signal. The slope characteristics of the baseline-corrected ECG signal are calculated, and the amplitude detection threshold for determining the QRS wave is dynamically updated based on the peak value of the historical real QRS wave. When the slope feature is greater than the preset slope threshold, and the amplitude of the baseline-corrected ECG signal is greater than the currently used amplitude detection threshold, the current sampling point is determined as a QRS wave candidate point. The QRS wave candidate points are verified to confirm them as real QRS waves. The first QRS wave candidate point is confirmed as the first real QRS wave, and subsequent QRS wave candidate points are verified based on the regularity of the time interval between them and the previously confirmed real QRS wave. After confirming the true QRS wave, the statistical average of the historical true RR intervals is calculated and updated.

[0007] Furthermore, the recursive baseline correction process includes: Update the baseline drift estimate b(n) for the current sampling point according to the formula: b(n) = b(n-1) + (y(n-1) - b(n-1)) / 2 k Where n is the current sampling point number, b(n-1) is the baseline drift estimate of the previous sampling point, y(n-1) is the original ECG signal value of the previous sampling point, and k is a positive integer.

[0008] Furthermore, the dynamically updated amplitude detection threshold includes: Update the amplitude detection threshold Th(n) for the current sampling point according to the formula: Th(n) = Th(n-1) + (A peak_hist -Th(n-1)) / 2 m Where m is a positive integer, A peak_hist The statistical average of the historical true QRS peak values ​​is a preset number.

[0009] Furthermore, methods for verifying QRS candidate points include: The absolute value of the difference between the time interval RR(n) between the current QRS candidate point and the previous confirmed true QRS wave and the statistical average of the historical true RR intervals is less than the statistical average of the historical true RR intervals divided by 2. p When the result is obtained, the candidate points of the QRS wave are confirmed as true QRS waves; Where p is a positive integer.

[0010] Furthermore, the main control microcontroller is also configured as follows: After confirming the true QRS wave, a calibrated trigger delay time is calculated based on the preset target trigger delay and the time consumed by determining the current sampling point as a true QRS wave as the algorithm's detection delay. The calibrated trigger delay time is equal to the preset target trigger delay minus the algorithm's detection delay. Based on the calibrated trigger delay time, a trigger signal for CCM treatment using myocardial contractility modulation is generated.

[0011] Furthermore, the target triggering delay is 30 to 40 milliseconds, and the detection delay is no more than 5 milliseconds.

[0012] Furthermore, the recursive baseline correction process, the dynamic update of the amplitude detection threshold, and the division operation in the verification of QRS wave candidate points are all implemented through shift operations.

[0013] An implantable myocardial contractility modulation device according to the present invention includes an in vivo implantation unit, wherein the in vivo implantation unit includes the QRS wave detection module for implantable myocardial contractility modulation.

[0014] Furthermore, it also includes an external wireless power supply unit that supplies power to the implanted unit via electromagnetic coupling.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Ultra-low power consumption: This invention adopts full fixed-point recursive calculation and replaces multiplication and division operations with computationally efficient shift operations, which greatly reduces the computational complexity and power consumption of the algorithm, enabling it to run stably on resource-constrained implantable devices that are powered only by wireless technology and have no battery.

[0016] 2. Strong anti-interference and high adaptability: Through recursive baseline correction, baseline drift of the in vivo environment can be effectively suppressed; through the dual feature fusion of slope and amplitude, misjudgment caused by single feature artifacts is effectively avoided; the dynamic adaptive threshold mechanism can automatically adjust according to the individual ECG signal characteristics of the patient, which is highly adaptable and together improves the accuracy of detection.

[0017] 3. High reliability and safety: Based on the artifact elimination step of the regularity of the RR interval, the final verification is carried out by using the physiological regularity of the heart rhythm, which can effectively filter out transient strong interference caused by exercise, etc., and further improve the reliability and safety of the detection.

[0018] 4. Ensuring treatment effectiveness: The algorithm has extremely low overall detection latency and can be combined with the timing calibration mechanism of myocardial contractility regulation therapy to accurately compensate for this latency, ensuring that the stimulation pulse can accurately fall within the absolute refractory period of the myocardium, thereby ensuring the effectiveness of the treatment. Attached Figure Description

[0019] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the implantable system architecture provided in an embodiment of the present invention.

[0020] Figure 2 This is a flowchart of the QRS wave detection method provided in an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram of the signal processing process provided in an embodiment of the present invention.

[0022] Figure 4 This is a schematic diagram of CCM stimulation timing calibration provided in an embodiment of the present invention. Detailed Implementation

[0023] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0024] Example 1 In one embodiment of the present invention, a low-power QRS wave detection module for implantable devices and its application system are disclosed. This method aims to achieve real-time and accurate detection of QRS waves in electrocardiogram signals with extremely low computational resources and power consumption, and is particularly suitable for battery-free implantable devices that regulate myocardial contractility and rely solely on wireless power.

[0025] Please see Figure 1 The diagram illustrates the overall architecture of an implantable system according to an embodiment of the present invention. The system mainly consists of two parts: an external wireless power supply unit 10 and an internal implantation unit 20. The external wireless power supply unit 10 is located outside the body and is responsible for providing all the energy required for continuous operation of the internal implantation unit 20 implanted in the patient's body through wireless means such as electromagnetic coupling.

[0026] The implantable unit 20 is the core component for QRS wave detection and subsequent treatment. Its hardware structure includes: an ECG acquisition electrode 21, a preamplifier and filter circuit 22, an analog-to-digital converter 23, a main control microcontroller 24, a myocardial contractility modulation stimulation module 25, and a trigger output interface 26. Specifically, the ECG acquisition electrode 21 can be implanted in a specific location within the patient's body (e.g., the subclavian pouch) to continuously acquire intracardiac or surface ECG signals close to the myocardial tissue. The acquired weak analog ECG signal is first sent to the preamplifier and filter circuit 22, which not only amplifies the signal with low noise to improve the signal-to-noise ratio but also filters out noise outside the ECG signal frequency band, such as power line interference, high-frequency electromyographic noise, and some low-frequency baseline drift, through its built-in bandpass filter (e.g., the passband frequency range is set to 0.5 Hz to 40 Hz).

[0027] The filtered and amplified analog ECG signal is input to the analog-to-digital converter 23. As an optional implementation, a 12-bit analog-to-digital converter can be used to digitize the signal at a sampling frequency (fs) of 250 Hz, generating a 12-bit precision digital sample value x(n) every 4 milliseconds, where n is the index of the current sampling point. This sampling rate and precision setting effectively controls the amount of data while ensuring the capture of the rapid changes in the QRS wave, thus helping to reduce the power consumption of subsequent processing.

[0028] The digitized raw electrocardiogram signal x(n) is then transmitted to the main control microcontroller 24, which serves as the core of the entire detection algorithm. This main control microcontroller 24 integrates a processor and a memory, with the memory containing computer program instructions that implement the low-power QRS wave detection method described in this invention. The processor executes these instructions to process each input sampling point x(n) in real time. It should be noted that this processing is entirely based on fixed-point arithmetic, without any floating-point operations, and all multiplication and division operations are replaced with highly efficient and low-power addition, subtraction, and shift operations. Once the main control microcontroller 24 successfully detects and confirms a real QRS wave according to the algorithm, it sends a precisely timed trigger signal to the myocardial contractility modulation stimulation module 25 via the trigger output interface 26. Correspondingly, upon receiving the trigger signal, the myocardial contractility modulation stimulation module 25 immediately generates and outputs the high-energy electrical stimulation pulse required for treatment.

[0029] Please refer to the following: Figure 2 This is a detailed flowchart of the QRS wave detection method provided in this embodiment of the invention. The method is executed by the main control microcontroller 24 and specifically includes the following steps: Step S101: Signal Acquisition. As mentioned above, the main control microcontroller 24 receives a 12-bit digital electrocardiogram signal sequence, denoted as x(n), obtained from the analog-to-digital converter 23 at a sampling rate of 250 Hz.

[0030] Step S102: Baseline Correction. Due to factors such as patient respiration and positional changes, the raw ECG signal x(n) often exhibits significant baseline drift, such as... Figure 3 As shown in box A. To eliminate the impact of this low-frequency interference on subsequent amplitude determination, this step employs an efficient recursive baseline correction algorithm. This algorithm requires no large amount of caching, only storing the calculation results from the previous moment. Specifically, the main control microcontroller 24 calculates the baseline-corrected ECG signal y(n) by executing the following two formulas: Where n is the current sampling point number, x(n) is the current input raw ECG signal sample value, b(n) is the algorithm's estimate of the baseline drift at the current sampling point, b(n-1) is the baseline drift estimate at the previous sampling point, and y(n-1) is the baseline-corrected ECG signal value at the previous sampling point. This recursive formula essentially constitutes a first-order low-pass filter for tracking the signal baseline. The parameter k is a positive integer that determines the filter's cutoff frequency; in this preferred embodiment, k is set to 6. It is worth noting that the division by... The operation of dividing by 64 is implemented in the main microcontroller 24 through a low-power right shift of 6 bits, rather than executing a high-power division instruction. By subtracting the estimated baseline value b(n) from the original signal x(n), the following can be obtained: Figure 3 The baseline-stable electrocardiogram signal y(n) is shown in box B.

[0031] Step S103: Feature Extraction and Threshold Update. This step performs two tasks in parallel. The first is to calculate the slope feature. To capture the steep rising or falling edges of the QRS group, the algorithm uses a simplified difference method to calculate the slope feature k(n), as shown in the formula: Where fs is the sampling rate of 250 Hz, therefore fs / 2 equals 125. This calculation is achieved by taking the absolute value of the difference between the current point and the previous two sampling points, effectively reflecting the rate of change of the signal. For example... Figure 3 As shown in box D, the slope feature k(n) exhibits a significant spike at the location where the QRS wave appears. Secondly, the amplitude detection threshold is dynamically updated. To adapt to changes in ECG signal amplitude under different individuals and physiological conditions, this method employs a dynamically adaptive amplitude detection threshold Th(n). A preset initial value can be used during the method initialization phase. During the detection process, this threshold is updated in each sampling period according to the following recursive formula:

[0032] Where Th(n-1) is the amplitude detection threshold of the previous sampling point, and A peak_histThis is a key variable stored in memory, representing the statistical average of the historical true QRS peak values ​​of a preset number (e.g., the most recent 3). The parameter m is a positive integer; in this preferred embodiment, m is set to 4. Similarly, dividing by... The division by 16 operation is achieved by right-shifting the data by 4 bits. This formula allows the threshold Th(n) to smoothly track the average peak level of recent QRS waves, such as... Figure 3 As shown in box C, the waveform of the threshold Th(n) will fluctuate up and down following the overall amplitude change of the signal y(n).

[0033] Step S104: Dual Feature Determination. This step is the core of identifying QRS wave candidate points. For a sampling point to be determined as a QRS wave candidate point, it must simultaneously meet conditions in both slope and amplitude dimensions. The specific determination logic is: the slope feature k(n) of the current sampling point is greater than a preset slope threshold K if and only if... th (For example, K) th A sampling point is marked as a QRS wave candidate only when its baseline-corrected amplitude y(n) is greater than the currently dynamically updated amplitude detection threshold Th(n), and the process proceeds to the next step S105. If any condition is not met, the current point is considered not to be part of the QRS wave, and the process returns directly to step S101 to continue processing the next sampling point. This dual-feature fusion strategy can effectively distinguish between true QRS waves (which usually have both high peak value and high slope) and various artifacts, such as T waves with high amplitude but gentle changes, or electromyographic noise with high slope but insufficient amplitude.

[0034] Step S105: Artifact Removal. Although the QRS candidate points selected in Step S104 have eliminated a large amount of interference, they may still contain artifacts generated by strenuous exercise or strong electromyographic activity that simultaneously satisfy the dual-feature condition. To further improve the reliability of the detection, this step introduces a verification mechanism based on the physiological regularity of cardiac rhythm. Specifically, the algorithm calculates the time interval between the current QRS candidate point and the previous finally confirmed true QRS wave, denoted as RR(n), and verifies it according to the following inequality:

[0035] Among them, RR avgThis is a statistical average of historical true RR intervals stored in memory, calculated based on a preset number (e.g., the most recent 8), reflecting the patient's recent average heart rate. The parameter p is a positive integer; in this preferred embodiment, p is set to 2. This inequality means that a newly emerging candidate point is considered to conform to cardiac rhythm only if the deviation between its RR interval and the recent average RR interval is within a certain range (here, one-quarter of the average). Otherwise, the candidate point is considered an isolated, non-physiological disturbance and is eliminated, and the process returns to step S101. It should be noted that for the first QRS wave candidate point detected by the algorithm, since there is no "previous confirmed true QRS wave," this verification step is skipped, and the first candidate point is directly confirmed as the first true QRS wave.

[0036] Step S106: Confirmation and Triggering. Once a QRS wave candidate point successfully passes the artifact removal verification in step S105, the main control microcontroller 24 finally confirms it as a real QRS wave. At this time, the main control microcontroller 24 performs two operations: first, updating historical data, that is, updating A with the peak amplitude of the currently confirmed QRS wave. peak_hist And update the RR with the newly calculated RR interval value. avg First, to provide the latest adaptive parameters for subsequent testing; second, to generate and output a trigger signal to the myocardial contractility regulation stimulation module 25 according to the preset treatment timing requirements.

[0037] Through the above series of fixed-point recursive calculation steps, the method of this embodiment can achieve high-precision detection of QRS waves (detection accuracy of not less than 99.2%) with extremely low power consumption (tested, the overall operating current of the module can be stabilized below 8 microamps), and the delay of the entire detection process is extremely low (not greater than 5 milliseconds), which fully meets the application requirements of battery-free wireless power-embedded devices.

[0038] Example 2 This embodiment provides a more detailed explanation of the dynamic adaptive logic described in Embodiment 1, particularly regarding the historical QRS peak value statistical average value A. peak_hist RR and historical RR interval statistical mean RR avg The algorithm employs a computation and update mechanism that endows it with powerful learning and adaptive capabilities, enabling it to automatically track and adapt to changes in the patient's physiological state (such as fluctuations in heart rate and changes in the amplitude of electrocardiogram signals), thereby significantly improving the algorithm's robustness.

[0039] In the internal memory (e.g., random access memory) of the main microcontroller 24, two circular buffers called circular queues are pre-allocated to store recent electrocardiogram activity characteristics.

[0040] The first is a peak buffer, used to store the peak amplitudes of the most recent N (e.g., N=3) confirmed true QRS waves. The size of this buffer is fixed. Whenever a new QRS wave is finally confirmed in step S106, the main microcontroller 24 records the maximum amplitude of that QRS wave within its group and stores this new peak amplitude in the peak buffer, overwriting the oldest data in the buffer. Subsequently, the main microcontroller 24 immediately recalculates the arithmetic mean of all N amplitudes in the buffer to obtain a new A. peak_hist Value. This newly calculated A peak_hist This will be immediately used in the update calculation of the dynamic amplitude detection threshold Th(n) for subsequent sampling points. In this way, the amplitude detection threshold can closely follow the actual intensity of the most recent heartbeats. If the overall amplitude of the ECG signal increases or decreases due to changes in patient posture or other reasons, the threshold can also be automatically adjusted to a suitable new level within a few heartbeats.

[0041] The second is the RR interval buffer, used to store the duration of the most recently confirmed true RR intervals (e.g., M=8). Its operation is similar to the peak buffer. Whenever a new true QRS wave is confirmed, the main microcontroller 24 calculates the time interval between it and the previous true QRS wave (i.e., the RR interval, typically in milliseconds or sampling points), and stores this new RR interval value in the RR interval buffer, overwriting the oldest data. Then, the main microcontroller 24 recalculates the arithmetic mean of all M RR interval values ​​in the buffer to obtain a new RR interval. avg Value. This new RR avg It will be immediately used for the artifact elimination determination of subsequent QRS wave candidate points (step S105).

[0042] To illustrate its working process more vividly, consider this scenario: After the device is implanted, the patient is at rest with a heart rate of 60 beats per minute, corresponding to an RR interval of approximately 1000 milliseconds. After a period of stable monitoring, the eight values ​​stored in the RR interval buffer are all close to 1000 milliseconds, thus the calculated RR interval... avg It also stabilizes at around 1000 milliseconds. At this point, the artifact removal formula... The defined effective RR interval range is approximately 750 to 1250 milliseconds. Now, the patient begins mild activity, and the heart rate gradually increases to 80 beats per minute, corresponding to an RR interval shortening to approximately 750 milliseconds. As new RR intervals of approximately 750 milliseconds are continuously identified and added to the RR interval buffer, older 1000-millisecond data points are replaced one by one. Therefore, RR... avgThe value will smoothly and gradually transition from 1000 milliseconds to 750 milliseconds. Accordingly, the decision window for artifact removal will also dynamically move to a new range centered on 750 milliseconds (approximately 562.5 milliseconds to 937.5 milliseconds).

[0043] This process ensures that the algorithm correctly identifies the 750-millisecond RR interval, which arises from a normal physiological increase in heart rate, as a valid heartbeat, rather than misclassifying it as artifact simply because it is significantly shorter than the initial 1000 milliseconds. Conversely, the mechanism also adapts to a slower heart rate. This adaptability greatly enhances the algorithm's robustness to individual patient differences and daily physiological changes, avoiding performance degradation caused by fluctuations in heart rate and ECG amplitude, and maintaining consistently high detection accuracy without any manual intervention or parameter recalibration.

[0044] Example 3 This embodiment proposes a parameter configurable scheme to further enhance the clinical application flexibility of the QRS wave detection method described in this invention. By allowing physicians to personalize the key parameters of the algorithm according to the specific patient's clinical condition or electrocardiogram characteristics, an optimal trade-off can be achieved between detection sensitivity (ensuring no missed reports of true QRS waves) and specificity (ensuring no false reports of QRS waves).

[0045] In this embodiment, some key thresholds in the algorithm are designed as configurable variables, rather than constants fixed in the program code. These parameters can be stored in the non-volatile memory (such as electrically erasable programmable read-only memory) of the main microcontroller 24 so that they can be retained after the device is powered off. Specifically, at least the following two parameters are set to be configurable: 1. Slope threshold K th 1. The slope threshold used in step S104 to determine the candidate points of the QRS wave. 2. The lower limit Th of the dynamic amplitude threshold. min Although the amplitude threshold Th(n) is dynamically adaptive, a minimum lower limit Th is set for it. min It is necessary to prevent excessive threshold reduction in cases of extremely weak signals or prolonged absence of QRS waves, which could lead to oversensitivity to noise.

[0046] Based on these configurable parameters, multiple operating modes can be preset for implantable devices. For example, the following two modes can be preset: Standard mode: In this mode, the parameter is set to K. th =200μV / ms, Th min =150μV. This is a well-balanced setting validated by extensive data, suitable for the vast majority of patients with normal ECG signal characteristics, achieving a good balance between sensitivity and specificity. High-sensitivity mode: In this mode, the parameter is adjusted to a lower value, such as K. th=150μV / ms, Th min =120μV. This mode is specifically designed for a special patient group whose ECG signal amplitude and slope are significantly lower than normal. For these patients, using the standard mode may result in missed detections because their QRS waves cannot reach the preset slope or amplitude thresholds. Switching to the high-sensitivity mode significantly improves the detection rate of these weak QRS waves by lowering the judgment threshold.

[0047] The parameter configuration process is as follows: After the device implantation surgery or during subsequent routine follow-up, the physician can use an external programmer that establishes a secure wireless communication link with the implanted internal unit 20. The physician can use the programmer to read the patient's real-time electrocardiogram and evaluate the performance of QRS wave detection. If the physician observes that the patient's electrocardiogram amplitude is low and there are occasional missed QRS waves in standard mode, they can switch the device's operating mode to "high-sensitivity mode" on the programmer's interface. The programmer then applies the new parameter configuration (K... th =150μV / ms,Th min =120μV) is wirelessly transmitted to the implanted unit 20. After receiving the instruction, the main control microcontroller 24 updates the corresponding parameter values ​​in its non-volatile memory, and loads and uses these new parameters to determine the QRS wave starting from the next sampling point.

[0048] Understandably, lowering the threshold to increase sensitivity may have a side effect: more noise or artifact signals may meet the initial dual-feature determination condition (step S104), thus becoming QRS wave candidate points. However, the algorithm design of this invention has inherent robustness. The subsequent artifact removal step (step S105), which verifies the physiological rhythm of the RR interval, acts as a powerful "filter," effectively identifying and filtering out the vast majority of isolated candidate points introduced by noise that do not conform to the heart rhythm. Therefore, even in "high-sensitivity mode," the specificity of the entire algorithm remains at a very high level, ensuring the overall reliability of the detection.

[0049] By providing such a configurable parameter scheme, the technology of this invention greatly expands its clinical applicability, enabling personalized and equally stable and reliable QRS wave detection for a wide range of patients with diverse electrocardiogram signal characteristics, demonstrating high clinical application value.

[0050] Example 4 This embodiment details the process of precisely timing-linking QRS wave detection results with myocardial contractility modulation stimulation output, a crucial step in ensuring the effectiveness and safety of myocardial contractility modulation therapy. Its core technology lies in accurately measuring and compensating for the inherent processing delay of the QRS wave detection algorithm itself, ensuring that the final therapeutic electrical stimulation is accurately applied at the predetermined target time point within the absolute refractory period of the myocardium.

[0051] Please see Figure 4 This is a schematic diagram of CCM stimulation timing calibration provided in an embodiment of the present invention. The diagram clearly depicts the complete time sequence from the physiological appearance of a QRS wave to the final emission of the myocardial contractility modulation stimulation pulse in the form of a time axis.

[0052] In myocardial contractility modulation therapy, a core requirement is that electrical stimulation must be applied within the absolute refractory period following ventricular depolarization (i.e., the QRS wave). This time window is very narrow, typically requiring the stimulation pulse to be delivered at a specific point in time (the target stimulation time point) after the QRS wave initiation; for example, the target trigger delay is set to 35 milliseconds.

[0053] However, all digital signal processing algorithms have inherent processing delays. In the scheme of this invention, from the time the ECG signal is acquired by the analog-to-digital converter 23 at a certain sampling point (corresponding to the true start point of the QRS wave), to the time the signal undergoes a series of judgment steps (S102 to S105) within the main control microcontroller 24, and is finally confirmed as part of the true QRS wave at a certain moment, this process takes time. We define this time as the "algorithm detection delay". Figure 4 The middle label is t detect This delay primarily includes the signal transmission delay in the hardware path and the computation time required for the microcontroller to execute algorithm instructions. Through precise offline testing and calibration during the product development phase, this algorithm detects the delay t. detect It can be determined as a stable and known value. Thanks to the extremely high computational efficiency of the algorithm of this invention, this delay is very short; for example, the calibration results show that t... detect It is a fixed value (e.g., 4 milliseconds) and is guaranteed not to exceed 5 milliseconds.

[0054] If the main microcontroller 24 issues a trigger signal immediately after confirming the QRS wave, the actual stimulation time will be the sum of the QRS wave initiation and the entire detection delay t. detect This will be later than the expected target stimulus time. detect To address this issue, this embodiment introduces a trigger delay calibration mechanism. A calibration formula is embedded in the main control microcontroller 24 to calculate a "calibrated trigger delay." Figure 4The value is marked as T_delay. The formula is:

[0055] Taking the aforementioned values ​​as an example, the calibrated trigger delay is calculated as follows:

[0056] The specific working process is as follows: When the main control microcontroller 24 is at time T... confirm (like Figure 4 After confirming a true QRS wave peak (at the "MCU confirms QRS wave" moment), it does not immediately send a trigger signal through trigger output interface 26. Instead, it starts an internal high-precision hardware timer whose duration is precisely set to the calculated calibrated trigger delay T. delay The timeout period is 31 milliseconds. The main microcontroller 24 then enters a wait state until the timer finishes counting down. When the 31-millisecond timeout period is reached, the timer generates an interrupt signal, and the main microcontroller 24 responds to this interrupt in T... confirm At a precise moment of +31 milliseconds, a high-level pulse is sent as a trigger signal to the myocardial contractility modulation stimulation module 25 via the trigger output interface 26. The myocardial contractility modulation stimulation module 25 is designed to immediately generate and output the bidirectional electrical stimulation pulse required for treatment (e.g., upon receiving the rising edge of this trigger signal) Figure 4 (The "actual stimulus pulse" in the text).

[0057] Through this precise compensation process, the final stimulation pulse emission time is controlled approximately (t) after the QRS wave initiation point. detect +T delay At 4ms + 31ms = 35ms, it is precisely aligned with the preset "target stimulation time point". Considering that the peak point of the QRS wave is usually a few milliseconds later than its starting point, and the time confirmed by the algorithm is usually near the peak point, this compensation method based on calibration delay can very accurately position the stimulation pulse within the absolute refractory period of the myocardium.

[0058] The closed-loop timing calibration mechanism of detection-trigger-stimulation described in this embodiment is a manifestation of the deep integration of the technical solution of this invention with specific clinical applications. It ensures that the myocardial contractility modulation stimulation pulse can be applied accurately and reliably within the optimal treatment time window, thereby maximizing the effectiveness of heart failure treatment and fundamentally avoiding the risk of malignant arrhythmias caused by inaccurate timing leading to stimulation falling into the vulnerable period of the ventricle, thus greatly ensuring the safety of treatment.

[0059] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features of the present invention can be arbitrarily combined with each other.

Claims

1. An implantable QRS wave detection module for myocardial contractility modulation, characterized in that, include: ECG acquisition electrodes, preamplifier and filter circuit, analog-to-digital converter and main control microcontroller; The ECG acquisition electrodes input the acquired analog ECG signal into the preamplifier and filter circuit for filtering and amplification. The filtered and amplified analog ECG signal is then input into the analog-to-digital converter for analog-to-digital conversion to obtain a digitized ECG signal. The main control microcontroller is configured to perform the following tasks using fully fixed-point recursive calculations: During the initialization phase, a preset initial amplitude detection threshold is used; The digitized electrocardiogram (ECG) signal is subjected to recursive baseline correction to filter out baseline drift and obtain the baseline-corrected ECG signal. The slope characteristics of the baseline-corrected ECG signal are calculated, and the amplitude detection threshold for determining the QRS wave is dynamically updated based on the peak value of the historical real QRS wave. When the slope feature is greater than the preset slope threshold, and the amplitude of the baseline-corrected ECG signal is greater than the currently used amplitude detection threshold, the current sampling point is determined as a QRS wave candidate point. The QRS wave candidate points are verified to confirm them as real QRS waves. The first QRS wave candidate point is confirmed as the first real QRS wave, and subsequent QRS wave candidate points are verified based on the regularity of the time interval between them and the previously confirmed real QRS wave. After confirming the true QRS wave, the statistical average of the historical true RR intervals is calculated and updated.

2. The implantable QRS wave detection module for myocardial contractility modulation according to claim 1, characterized in that, The recursive baseline correction process includes: Update the baseline drift estimate b(n) for the current sampling point according to the formula: b(n)=b(n-1)+(y(n-1)-b(n-1)) / 2 k Where n is the current sampling point number, b(n-1) is the baseline drift estimate of the previous sampling point, y(n-1) is the original ECG signal value of the previous sampling point, and k is a positive integer.

3. The implantable QRS wave detection module for myocardial contractility modulation according to claim 1, characterized in that, The dynamically updated amplitude detection threshold includes: Update the amplitude detection threshold Th(n) for the current sampling point according to the formula: Th(n)=Th(n-1)+(A peak_hist -Th(n-1)) / 2 m Where m is a positive integer, A peak_hist The statistical average of the historical true QRS peak values ​​is a preset number.

4. The implantable QRS wave detection module for myocardial contractility modulation according to claim 1, characterized in that, Methods for verifying QRS candidate points include: The absolute value of the difference between the time interval RR(n) between the current QRS candidate point and the previous confirmed true QRS wave and the statistical average of the historical true RR intervals is less than the statistical average of the historical true RR intervals divided by 2. p When the result is obtained, the candidate points of the QRS wave are confirmed as true QRS waves; Where p is a positive integer.

5. The implantable QRS wave detection module for myocardial contractility modulation according to claim 1, characterized in that, The main control microcontroller is also configured to: After confirming the true QRS wave, a calibrated trigger delay time is calculated based on the preset target trigger delay and the time consumed by determining the current sampling point as a true QRS wave as the algorithm's detection delay. The calibrated trigger delay time is equal to the preset target trigger delay minus the algorithm's detection delay. Based on the calibrated trigger delay time, a trigger signal for CCM treatment using myocardial contractility modulation is generated.

6. The implantable QRS wave detection module for myocardial contractility modulation according to claim 5, characterized in that, The target triggering delay is 30 to 40 milliseconds, and the detection delay is no more than 5 milliseconds.

7. The implantable QRS wave detection module for myocardial contractility modulation according to claim 5, characterized in that, The recursive baseline correction process, the dynamic update of the amplitude detection threshold, and the division operation in the verification of QRS wave candidate points are all implemented through shift operations.

8. An implantable myocardial contractility modulation device, characterized in that, The device includes an implantable unit, which includes the implantable QRS wave detection module for myocardial contractility modulation as described in any one of claims 1-7.

9. The implantable myocardial contractility modulation device according to claim 8, characterized in that, It also includes an external wireless power supply unit that supplies power to the implanted unit via electromagnetic coupling.

Images