A voltage control method and system for pulse sequence-based electroporation ablation

Abstract

This invention discloses a pulse sequence-based method and system for controlling the voltage of electroporation ablation, involving key technologies such as intelligent target region identification, electric field modeling, dose control, real-time monitoring, and feedback. AI technology is used to accurately identify tumors and plan treatment, optimizing the electric field distribution to ensure precise application of the electroporation pulses. Closed-loop control dynamically adjusts pulse parameters, and sensor fusion technology provides treatment efficacy evaluation. The system integrates data analysis and machine learning to intelligently adjust treatment strategies, enabling personalized treatment and complication prediction, thereby improving treatment effectiveness and patient safety. This invention is applicable to precision tumor treatment, ensuring efficacy while protecting healthy tissues.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a voltage control method and system for electroporation ablation based on pulse sequences. It is suitable for the precise location and treatment of diseases such as tumors, especially in achieving precise control of the target area in electroporation treatment, while maximizing the protection of surrounding healthy tissues from damage. Background Technology

[0002] Traditional cancer treatments, such as surgical resection, chemotherapy, and radiotherapy, often come with certain side effects and risks. For example, surgery may lead to wound infection, and chemotherapy and radiotherapy may damage normal tissue. Electroporation ablation, as a novel cancer treatment technique, creates temporary pores in the cell membrane, causing leakage of internal cellular substances and leading to cell death. It is particularly suitable for tumors that are difficult to surgically remove or are insensitive to chemotherapy and radiotherapy. However, traditional electroporation ablation methods have the following technical problems:

[0003] (1) Inaccurate target identification: Traditional methods rely on doctors' experience and manual operation, which is not accurate enough in identifying and locating tumor areas, and can easily cause accidental damage to healthy tissues.

[0004] (2) Uneven electric field distribution: The lack of optimization in the selection of electrode layout and pulse parameters leads to uneven distribution of electric field in the target area, which affects the treatment effect.

[0005] (3) Uncontrollable treatment process: The lack of real-time monitoring and feedback mechanism makes it impossible to dynamically adjust pulse parameters according to treatment progress and tissue response, affecting dose control and treatment safety.

[0006] (4) Difficulty in adjusting treatment strategies: The lack of a data-based intelligent decision-making system makes it difficult to adjust strategies according to individual patient differences and treatment effects. Summary of the Invention

[0007] This invention provides a voltage control method and system for electroporation ablation based on pulse sequences, in order to solve the problem of how to accurately identify and apply pulse voltage to the tumor area during electroporation ablation treatment while minimizing the impact on surrounding healthy tissues, thereby achieving personalized treatment and precise dose control.

[0008] To address the above problems, this invention provides a voltage control method and system for electroporation ablation based on pulse sequences, comprising:

[0009] A voltage control method for electroporation ablation based on pulse sequences, comprising:

[0010] S100 integrates target recognition and electric field modeling, using AI technology to automatically identify target regions in medical images, such as tumors, and distinguish healthy tissues, and establish a three-dimensional electric field distribution model.

[0011] S200, combined with electric field optimization and dose control, forms a closed-loop control system that dynamically adjusts pulse parameters to optimize electric field distribution and ensure that the target area receives an accurate dose;

[0012] S300 integrates treatment effect assessment with sensor technology to monitor electric field strength, tissue state and dose distribution in real time, and provide feedback for dose control;

[0013] S400 integrates information from other modules, performs comprehensive analysis, guides adjustments to treatment strategies, and achieves personalized treatment.

[0014] As a preferred embodiment, step S100 includes:

[0015] S110. Utilize deep learning and computer vision technologies to identify target regions and ensure accurate identification of target regions;

[0016] S120. Based on the location, shape, and size of the target area, and combined with parameters such as tissue conductivity, establish an electric field distribution model.

[0017] S130. Generate optimal pulse sequence parameters to maximize the electroporation effect on tumor cells while minimizing the impact on healthy tissues.

[0018] As a preferred option, in step S200, the closed-loop control system automatically adjusts the pulse sequence by real-time monitoring of the electric field strength and tissue response in the target area to achieve precise dose control.

[0019] As a preferred option, step S300 integrates multiple sensors and imaging technologies to monitor key parameters during the treatment process in real time, providing a basis for dose control.

[0020] As a preferred option, step S400 integrates information for comprehensive analysis to guide adjustments to the treatment strategy, including changing pulse sequence parameters, electrode configuration, or adding adjuvant treatment methods.

[0021] As a preferred option, the following are included:

[0022] The target recognition module is used to identify target regions in medical images and distinguish healthy tissues;

[0023] The electric field modeling module is used to establish electric field distribution models;

[0024] The dose control module is used to form a closed-loop control system and dynamically adjust the pulse parameters;

[0025] The monitoring and feedback module is used to monitor the treatment process in real time and provide feedback.

[0026] The strategy adjustment module is used to integrate information and guide adjustments to treatment strategies.

[0027] As a preferred embodiment, the target recognition module utilizes deep learning and computer vision technologies for accurate recognition.

[0028] As a preferred embodiment, the dose control module automatically adjusts the pulse sequence parameters by monitoring the electric field strength and tissue response in the target area in real time.

[0029] As a preferred embodiment, the monitoring feedback module integrates multiple sensors and imaging technologies to monitor key parameters during the treatment process in real time.

[0030] As a preferred embodiment, the strategy adjustment module integrates information to guide the adjustment of personalized treatment strategies, thereby improving treatment effectiveness and safety.

[0031] The key innovations of this invention include:

[0032] 1. Utilizing deep learning and computer vision technologies, tumor areas are automatically identified and healthy tissues are distinguished, ensuring precise treatment targeting.

[0033] 2. Establish a three-dimensional electric field distribution model, optimize the electric field distribution by combining parameters such as tissue conductivity, and predict the electric field effect.

[0034] 3. Design a closed-loop control system to automatically adjust the pulse sequence by real-time monitoring of electric field strength and tissue response, thereby achieving precise dose control.

[0035] 4. Integrate multi-source data and use advanced data analysis and machine learning algorithms to evaluate treatment effectiveness, predict complications, and optimize treatment strategies.

[0036] 5. Achieve dynamic feedback loop, intelligently adjust subsequent treatment plans based on treatment progress and individual patient differences, improve treatment success rate and patient safety.

[0037] This invention, by introducing advanced technologies and algorithms, achieves precise control and personalized treatment in electroporation ablation therapy, solving problems such as inaccurate target identification, uneven electric field distribution, uncontrollable treatment process, and difficulty in adjusting treatment strategies in traditional methods. Specific beneficial effects are as follows:

[0038] 1. Improved target recognition accuracy: By utilizing deep learning and computer vision technology, tumor areas in medical images are automatically identified and healthy tissues are accurately distinguished, ensuring that the electroporation pulses are precisely applied to the target area, reducing damage to healthy tissues and improving the safety and effectiveness of treatment.

[0039] 2. Optimization of electric field distribution: Through a closed-loop system of electric field modeling and dose control, the electrode layout and pulse parameters are dynamically adjusted to ensure that the target area receives a precise dose of electroporation pulse. At the same time, a multi-objective optimization algorithm is used to protect healthy tissues, thereby improving the accuracy and efficiency of treatment.

[0040] 3. Provides real-time monitoring and feedback: Integrates sensor and imaging technologies to monitor key parameters during treatment, such as electric field distribution, tissue temperature, and degree of cell damage, providing a basis for dose control and ensuring the controllability and safety of the treatment process.

[0041] 4. Personalized treatment strategies were implemented: Through the comprehensive analysis module, subsequent treatment plans were intelligently adjusted, including changing pulse sequence parameters, electrode configuration, or adding auxiliary treatment methods, which improved the success rate of treatment and patient safety, while reducing the risk of complications.

[0042] 5. Promotes intelligent treatment decision-making: Through machine learning algorithms, learning from historical treatment cases, continuously optimizing feedback control algorithms and multi-objective optimization strategies, improving the accuracy and efficiency of future treatments, and forming a feedback loop of continuous improvement.

[0043] 6. Reduced uncertainty in the treatment process: Through real-time monitoring and feedback mechanisms, as well as model predictive control (MPC) strategies, the precise adjustment of electrode voltage output is ensured, avoiding dose deviations caused by individual differences and treatment progress, and ensuring the consistency of treatment effects. Attached Figure Description

[0044] Figure 1 This is a schematic flowchart of a voltage control method and system for electroporation ablation based on pulse sequences provided in an embodiment of the present invention. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of this application. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0047] Reference Figure 1This is a flowchart of a voltage control method and system for electroporation ablation based on pulse sequences, according to an embodiment of the present invention. The voltage control method and system for electroporation ablation based on pulse sequences may include at least steps S100-S400:

[0048] The S100 integrates target recognition and electric field modeling, using AI for intelligent identification and treatment planning.

[0049] S200, combined with electric field optimization and dose control, forms a closed-loop control system that dynamically adjusts pulse parameters.

[0050] The S300 integrates treatment effectiveness assessment with sensor technology to provide real-time monitoring and feedback.

[0051] S400 integrates information from other modules, performs comprehensive analysis, and guides adjustments to treatment strategies.

[0052] The voltage control method and system for electroporation ablation based on pulse sequence provided in this invention are mainly applied to solve the technical problem of "how to accurately identify and apply pulse voltage to the tumor area in electroporation ablation treatment while minimizing the impact on surrounding healthy tissues, so as to achieve personalized treatment and precise dose control".

[0053] This invention discloses a pulse sequence-based method and system for controlling the voltage of electroporation ablation, involving key technologies such as intelligent target region identification, electric field modeling, dose control, real-time monitoring, and feedback. AI technology is used to accurately identify tumors and plan treatment, optimizing the electric field distribution to ensure precise application of the electroporation pulses. Closed-loop control dynamically adjusts pulse parameters, and sensor fusion technology provides treatment efficacy evaluation. The system integrates data analysis and machine learning to intelligently adjust treatment strategies, enabling personalized treatment and complication prediction, thereby improving treatment effectiveness and patient safety. This invention is applicable to precision tumor treatment, ensuring efficacy while protecting healthy tissues.

[0054] S100: Integrating target recognition and electric field modeling, AI is used for intelligent identification and treatment planning. Step S100 includes steps S110-S130:

[0055] S110. Utilizing deep learning and computer vision technologies, automatically identify target regions (such as tumors) in medical images and distinguish them from healthy tissue. This step involves image preprocessing, feature extraction, classification, and localization to ensure accurate identification of the target region. Specifically:

[0056] ① Data preprocessing and feature encoding: Let the medical image data be I, where I∈R H×WH and W represent the height and width of the image, respectively. First, the image is preprocessed, including normalization and resizing, to optimize subsequent calculations. We introduce a preprocessing function Φ, which takes the original image I as input and outputs the preprocessed image I':

[0057] I′=Φ(I)

[0058] Next, we use a concept based on abstract algebra—feature encoding—to transform the preprocessed image I' into a feature vector F. Here, we define a feature extractor Ψ, which converts the image into points in the feature space:

[0059] F = Ψ(I′)

[0060] ② The feature vector F contains features from different regions of the image, including texture, shape, and color. We use a multiple linear regression model M. LR To predict the probability p that each pixel in the image belongs to the target region. Assuming F is a vector composed of n features, the multivariate regression model can be expressed as:

[0061]

[0062] Where, β i These are the model's weight coefficients, F. i It is the i-th element in the feature vector.

[0063] ③ To further improve the accuracy of target region identification, we introduce differential equation modeling to simulate the boundary diffusion process of the tumor region. Let u(x,y,t) be the probability distribution of the tumor boundary at spatial location (x,y) at time t. We can establish a model based on the diffusion equation:

[0064]

[0065] Where D is the diffusion coefficient. It is the Laplace operator, and f(u) is a nonlinear function used to describe the interaction between the tumor boundary and healthy tissue. The initial condition is u(x,y,0)=p(x,y), that is, the result of multivariate regression analysis is used as the initial probability distribution.

[0066] ④ Finally, by solving the above differential equation, we obtain the state of the probability distribution u(x,y,T) of the tumor region at the final time T. Then, we apply a threshold θ to determine the binary mask M of the target region:

[0067] M(x,y) = begin{cases}

[0068] 1&if u(x,y,T)>θ

[0069] 0 & otherwise

[0070] end{cases}

[0071] In this way, the mask M will clearly distinguish the target area (such as a tumor) from healthy tissue. This process fully utilizes the data source and input quantity indicators in our patented invention to ensure that the electroporation pulses can be precisely applied to the target area without affecting the surrounding healthy tissue.

[0072] S120. Based on the location, shape, and size of the target area, and combined with biophysical parameters such as tissue conductivity, establish a three-dimensional electric field distribution model. The model needs to consider factors such as electrode layout and pulse parameters to predict the distribution and effect of the electric field within the target area. Specifically:

[0073] ① Input data

[0074] G: Location, shape, and size information of the target area.

[0075] σ g σ g Electrical conductivity of the target region.

[0076] σ h σ h Electrical conductivity of healthy tissue.

[0077] E: Electrode layout, including the position, orientation, and spacing of the electrodes.

[0078] V: Electrode voltage.

[0079] P: Pulse parameters, including frequency, intensity, and duration.

[0080] ② Let's model the electric field distribution. Let the electric field strength be E and the electric potential be φ. Then the relationship between the electric field strength E and the electric potential φ is: The electric potential φ satisfies the following equation:

[0081]

[0082] Where σ(x) represents the conductivity at position x, and Ω is the entire modeling region.

[0083] At the electrode location, the potential satisfies the Dirichlet boundary condition:

[0084] φ(x)=V for x∈Γ e

[0085] At the boundary Γ b Above, the electric potential satisfies the Neumann boundary condition:

[0086]

[0087] Where n is the boundary Γ b The external normal vector.

[0088] ③ Solve the above partial differential equations using the finite element method to obtain the numerical solution of the electric potential φ(x), thereby obtaining the electric field strength E(x). To optimize the electric field distribution, we define an objective function J, which comprehensively considers the desired distribution of the electric field strength in the target region and the minimization of the electric field strength in the healthy tissue region:

[0089]

[0090] Among them, E d λ is the desired electric field intensity distribution in the target region, and λ is the Lagrange multiplier used to balance the importance of the two objectives.

[0091] ④ By solving the above optimization problem, we can adjust the electrode layout E and the pulse parameter P to optimize the electric field distribution. Using optimization algorithms such as gradient descent or Newton's method, we iteratively update E and P until the objective function J reaches its minimum.

[0092] ⑤ Output the optimized electric field distribution model, including the distribution map of electric field intensity E(x) and the prediction of electric field effects. Perform numerical simulation to verify that the electroporation pulse is accurately applied to the target area while protecting the surrounding healthy tissue from damage.

[0093] Where: G: geometric information of the target region, σ g It is the conductivity of the target region, σ h Here, E is the electrical conductivity of healthy tissue, E is the electrode layout, V is the electrode voltage, P is the pulse parameter, φ is the potential, Ω is the modeling region, and Γ is the electrical conductivity of healthy tissue. e It is the electrode position, Γ b It is the boundary, n is the normal vector outside the boundary, J is the objective function, and E is the objective function. d λ represents the desired electric field intensity distribution, and λ is the Lagrange multiplier.

[0094] S130. Based on the electric field model and target recognition results, the intelligent algorithm will generate optimal pulse sequence parameters (such as frequency, intensity, and duration) to maximize the electroporation effect on tumor cells while minimizing the impact on surrounding healthy tissues. The planning process needs to consider multi-objective optimization problems to ensure the safety and effectiveness of the treatment. Specifically:

[0095] ① Data Input and Preprocessing. First, we define the input dataset, which includes:

[0096] X: Spatial coordinates of the tumor region; C: Electrode configuration matrix; S: Tissue state matrix, including conductivity, resistance, and dielectric constant; D: Preliminary parameters of the pulse sequence, including frequency, intensity, and duration; H: Healthy tissue protection threshold matrix.

[0097] ② Establish a nonlinear model of the electric field distribution. Assuming the boundary between the tumor region and healthy tissue can be characterized on the manifold, we use the Laplace-Beltramie operator ΔM on the Riemannian manifold to describe the change in potential V, where M represents the complex manifold composed of tumor and healthy tissue:

[0098]

[0099] λ(x) is the eigenvalue at position x, which is related to the tissue state S; F(x) is the external electric field source introduced by the electrode configuration C.

[0100] ③ To maximize the effect of electroporation on tumor cells and minimize the impact on healthy tissue, we construct an objective function J based on functional analysis, which combines the tumor electroporation efficiency Φ and the healthy tissue protection coefficient Ψ:

[0101]

[0102] Among them, Ω t and Ω h These represent the tumor region and the healthy tissue region, respectively; α is a weighting factor that regulates the balance between the two targets.

[0103] ④ We use the variational method to solve for the extreme points of the objective function J and find the optimal pulse sequence parameters D. * and electrode configuration C * :

[0104]

[0105] ⑤ Apply algorithms to check whether the model's iterative process converges, ensuring the stability of the treatment plan.

[0106] If the convergence condition is not met, fine-tune α or reinitialize D and C, and repeat steps ③ and ④.

[0107] ⑥ Once the model converges, output the optimal pulse sequence parameters D. * and electrode configuration C * , as well as the corresponding electric field distribution diagram and treatment effect evaluation report.

[0108] S200, combined with electric field optimization and dose control, forms a closed-loop control system that dynamically adjusts pulse parameters.

[0109] Step S200 includes steps S210-S220:

[0110] S210. During treatment, the voltage output of the electrodes is dynamically adjusted based on real-time feedback information to optimize the electric field distribution and ensure that the target area receives a precise dose. Specifically:

[0111] ① Real-time monitoring and data collection. Electric field intensity monitoring: A sensor network is used to measure the electric field intensity E(x,t) in the tumor region and surrounding healthy tissue in real time, where x represents spatial location and t represents time. Tissue state feedback: The tissue state matrix S(t) is collected, including changes in conductivity, resistance, and dielectric constant over time. These parameters affect the electric field distribution and penetration depth. Dose distribution assessment: Based on the current electric field distribution and tissue state, the dose distribution D(x,t) in the tumor region is calculated.

[0112] ② Deviation calculation: Compare the current dose distribution D(x,t) with the target dose distribution D target (x), calculate the deviation ΔD(x,t)=D(x,t)-D target (x). Electric field distribution assessment: The effectiveness and safety of the electric field distribution are assessed by using real-time electric field intensity data and tissue status feedback, and whether it deviates from the ideal state.

[0113] ③ Dynamic dose control strategy. Feedback control algorithm: A model predictive control (MPC)-based strategy is adopted, combining feedforward control and feedback control to adjust the electrode voltage output V(t) in real time to correct the deviation ΔD(x,t). Multi-objective optimization: While adjusting the voltage output, the healthy tissue protection threshold matrix H is considered. A multi-objective optimization algorithm is used to ensure that the dose to the tumor region is increased while avoiding damage to healthy tissue.

[0114] ④ Dynamic adjustment of electrode voltage. Voltage adjustment formula: Based on the deviation ΔD(x,t) and the tissue state matrix S(t), the electrode voltage output V(t) is dynamically adjusted. Let k v Let be the voltage adjustment coefficient, and ΔV(t) be the voltage increment. Then we have:

[0115]

[0116] in This indicates the sensitivity of electric field strength to voltage; the integral term reflects the overall response of the electric field strength in the entire tumor region to voltage changes.

[0117] Safety threshold check: After each adjustment, check whether the new electrode voltage output V(t) + ΔV(t) exceeds the healthy tissue protection threshold H. If it does, limit the voltage adjustment range to ensure safety.

[0118] ⑤ Implementation and Monitoring. Adjustment: Update the electrode voltage output V(t) based on the calculated voltage increment ΔV(t). Continuous Monitoring: Continuously monitor the electric field distribution, dose distribution, and tissue status throughout the treatment process to ensure treatment efficacy and safety.

[0119] ⑥ Results Feedback and System Learning. Results Evaluation: After treatment, assess the consistency between the final dose distribution and the target distribution, record deviations and adjustments. System Learning: Utilize machine learning algorithms to learn from historical treatment cases, continuously optimizing the feedback control algorithm and multi-objective optimization strategy to improve the accuracy and efficiency of future treatments.

[0120] The S220 closed-loop control system automatically adjusts the pulse sequence to achieve precise dose control by monitoring the electric field strength and tissue response in the target area in real time. This includes fine-tuning the frequency, intensity, and duration of the pulses based on real-time data to adapt to individual differences and treatment progress. Specifically:

[0121] ① The electric field strength E(x,t) of the target area is monitored in real time using a sensor array, where x represents the spatial location and t represents time. Tissue response monitoring: Real-time response data of the target area and surrounding tissues are collected, including but not limited to temperature changes, electrical impedance changes, and metabolic activity indicators. These data reflect the immediate response of the tissue to the electroporation pulse.

[0122] Electrophysiological parameters updated: Tissue conductivity σ(x,t) and dielectric constant ε(x,t) are updated, as these parameters change with time and tissue state.

[0123] ② Data Processing and Analysis. Electric Field Distribution Reconstruction: Using real-time monitored electric field intensity data and tissue electrophysiological parameters, the three-dimensional electric field distribution E3D(x,t) of the target area is reconstructed. Tissue Response Analysis: Tissue response data is analyzed to assess the effects of the pulse sequence on the target area and surrounding tissues, and to identify potential areas of excessive or insufficient dose.

[0124] ③ Model Prediction and Dose Assessment. Dose Distribution Prediction: Based on the reconstructed electric field distribution E3D(x,t) and tissue response analysis results, the dose distribution D(x,t) in the target area is predicted. Dose Deviation Identification: The predicted dose distribution D(x,t) is compared with the ideal dose distribution D. ideal x, calculate the dose deviation ΔD(x,t)=D(x,t)-D ideal x.

[0125] ④ Pulse sequence parameter adjustment. Parameter adjustment algorithm: Design a parameter adjustment algorithm to dynamically adjust the pulse sequence parameters, including frequency f, intensity I, and duration t, based on the dose deviation ΔD(x,t). pulse Individual Difference Adaptation: The algorithm must consider individual differences, including but not limited to tissue type, thickness, and metabolic rate, to ensure that the adjusted pulse sequence parameters are suitable for the current patient condition. Treatment Progress Adaptation: Based on the treatment stage and tissue response, the pulse sequence parameters are gradually adjusted to adapt to treatment progress and optimize overall treatment efficacy.

[0126] ⑤ Closed-loop control and feedback. Dosage control closed loop: The dosage control function forms a closed-loop control system, monitoring the dose deviation ΔD(x,t) in real time and adjusting the pulse sequence parameters according to the deviation until the deviation is reduced to an acceptable range. Safety threshold monitoring: Throughout the treatment process, the pulse sequence parameters are monitored to ensure they do not exceed the safety threshold, avoiding damage to surrounding healthy tissues.

[0127] ⑥ Results Feedback and Iterative Optimization. Results Evaluation and Recording: Record the pulse sequence parameters and dose distribution during each dose control process to evaluate treatment efficacy and safety. Iterative Optimization: Utilize historical data and machine learning algorithms to continuously optimize the dose control algorithm, improving dose control accuracy and efficiency.

[0128] Through the above process, the dose control function achieves precise dose control based on real-time data, ensuring that the electroporation pulse is accurately applied to the target area, while adapting to individual differences and treatment progress, demonstrating the innovation and practice of this invention patent in precise dose control and prediction.

[0129] S300 integrates treatment efficacy assessment with sensor technology to provide real-time monitoring and feedback. Step S300 includes steps S310-S320:

[0130] The S310 integrates multiple sensors and imaging technologies (such as ultrasound, MRI, and CT) to monitor key parameters such as electric field distribution, tissue temperature, and cell damage level during treatment in real time, providing a basis for dose control. Specifically:

[0131] ① Data Input and Acquisition. Electric Field Intensity Data: Electric field intensity E(x,t) is collected in real-time using an array of electric field sensors deployed around the target area, where x represents spatial location and t represents time. Tissue Temperature Data: Temperature changes in the target area and surrounding tissues are monitored using temperature sensors, and temperature data T(x,t) is collected. Cell Damage Indicators: The degree of cell damage is monitored using biosensors or markers, and cell damage indicators C(x,t) are collected. Imaging Data: Imaging technologies such as ultrasound, MRI, and CT are integrated to acquire real-time structural and functional changes in the target area and surrounding tissues, and imaging data I(x,t) is collected.

[0132] ② Data Processing and Analysis. Electric Field Distribution Reconstruction: Based on the electric field intensity data E(x,t), the three-dimensional electric field distribution of the target area is reconstructed in real time using an inverse problem solution method. Temperature Distribution Analysis: Combining temperature data T(x,t) and a tissue heat conduction model, the temperature distribution is analyzed to identify potential thermal damage areas. Cell Damage Assessment: Based on the cell damage index C(x,t) and imaging data I(x,t), the degree of cell damage and treatment efficacy are assessed.

[0133] ③ Real-time feedback and dosage control basis. Dosage control feedback: Real-time monitoring of key parameters (electric field distribution, tissue temperature, degree of cell damage) is compared with treatment goals and safety thresholds to generate real-time feedback signals. Parameter adjustment basis: Based on the real-time feedback signals, the dosage control function is provided with a basis for adjusting pulse sequence parameters (such as frequency, intensity, and duration) to optimize treatment efficacy and safety.

[0134] ④ Data Fusion and Intelligent Analysis. Multimodal Data Fusion: Data fusion algorithms are used to integrate data from different sensors and imaging technologies, improving the accuracy and reliability of monitoring results. Intelligent Analysis Model: Machine learning or deep learning models are used to intelligently analyze the fused data, predict treatment progress and potential complications, provide early warnings, and guide dosage control.

[0135] ⑤ Results Output and Recording. Monitoring Results Display: Real-time display of dynamic changes in key parameters such as electric field distribution, tissue temperature, and cell damage level, facilitating a direct understanding of the treatment process by clinicians. Historical Data Recording: Saving all monitoring data and analysis results for subsequent efficacy evaluation, research analysis, and iterative optimization of dosage control algorithms.

[0136] S320. Compare the monitoring data with the expected results to generate feedback signals for dynamic adjustment of the treatment strategy. This includes identifying undertreatment or overtreatment, as well as any unexpected tissue reactions, so as to take corrective measures in a timely manner. Specifically:

[0137] ① Data Input and Expected Effect Setting. Electric Field Intensity and Distribution: Real-time electric field intensity data is collected from the sensor array and compared with the preset ideal electric field distribution to check for sufficient coverage and uniformity. Tissue Temperature Data: Temperature sensors are used to monitor temperature changes in the target area, ensuring the temperature remains within the effective treatment range without causing thermal damage. Cell Damage Indicators: Data on the degree of cell damage is collected through biomarkers or imaging techniques and compared with the target damage level defined in the treatment plan. Imaging Feedback: Ultrasound, MRI, or CT imaging technologies are integrated to acquire real-time images of tissue structural changes and assess changes in the treatment area.

[0138] ② Data Processing and Analysis. Real-time Comparative Analysis: The monitored data is compared in real time with pre-set treatment goals and safety thresholds to identify deviations in the treatment process. Anomaly Detection: Statistical methods or machine learning models are used to identify insufficient, excessive, or unexpected tissue responses, such as excessively high local temperatures or uneven electric field distribution. Trend Prediction: Based on historical data and the current state, the trend of treatment effectiveness is predicted to determine whether it meets expectations.

[0139] ③ Feedback Signal Generation and Strategy Adjustment. Feedback Signal Generation: When monitoring data deviates from the expected effect, a feedback signal is generated, including suggestions for adjusting treatment parameters, such as fine-tuning the electric field strength, pulse frequency, or duration. Strategy Adjustment Instructions: The feedback signal is translated into specific strategy adjustment instructions, such as increasing or decreasing the intensity of the electroporation pulse, or extending or shortening the treatment cycle. Immediate Intervention: For detected emergencies, such as potential thermal damage risks, immediate warnings are issued and corrective measures are proposed to avoid unnecessary tissue damage.

[0140] ④ Dynamic adjustment of treatment strategies. Real-time adjustment: Treatment parameters are dynamically adjusted based on feedback signals and strategy adjustment instructions to ensure the treatment process is always in an optimal state. Adaptive control: The system has self-learning capabilities and can adaptively adjust the treatment plan according to individual patient differences and changes during the treatment process. Efficacy optimization: Through continuous feedback loops, dosage control is continuously optimized to achieve the most ideal treatment effect while minimizing side effects.

[0141] ⑤ Results Recording and Analysis. Feedback Loop Recording: Each generated feedback signal and its resulting strategy adjustments are recorded, serving as the basis for subsequent data analysis and treatment strategy improvement. Long-Term Efficacy Assessment: Feedback data is collected and analyzed to assess the long-term effectiveness and safety of the treatment, providing a reference for future treatment planning.

[0142] S400: Integrate information from other modules, perform comprehensive analysis, and guide adjustments to the treatment strategy. Step S400 includes steps S410-S430:

[0143] S410 collects and analyzes all data from the intelligent identification and planning, dynamic dose control, and multimodal monitoring modules, including target area information, treatment parameters, and real-time feedback. Specifically:

[0144] ① Data Collection. Target Area Information: The intelligent identification and planning module acquires the specific location, shape, size, and relationship with surrounding important structures of the tumor or other lesions. This data comes from high-precision imaging technologies such as MRI, CT, or PET scans. Treatment Parameters: Electroporation treatment parameters set by the dynamic dose control module are collected, including but not limited to electric field strength, pulse width, pulse interval, and total number of pulses. These parameters are carefully designed based on the characteristics of the target area and the expected degree of cell damage. Real-time Feedback: Real-time monitoring data from the multimodal monitoring module is received, including electric field distribution, tissue temperature, cell damage indicators, and imaging feedback, to verify treatment effectiveness and promptly identify potential problems.

[0145] ② Data Analysis and Integration. Cross-validation: Cross-validate the collected data to ensure consistency and accuracy. For example, compare imaging data from different time points to confirm whether changes in the treatment area meet expectations. Pattern Recognition: Utilize machine learning algorithms to analyze data patterns and identify factors that may affect treatment outcomes, such as tumor heterogeneity and hemodynamic changes. Deviation Detection: Compare actual monitoring data with expected values ​​in the treatment plan to identify potential deviations or abnormalities during treatment, such as uneven electric field distribution or tissue responses exceeding expectations.

[0146] ③ Data Fusion for Comprehensive Processing and Decision Making: Data from all sources is integrated and processed to form a comprehensive view reflecting the current treatment status. Intelligent Decision Making: Based on the integrated data, the intelligent decision-making system can automatically identify the optimal treatment path, including whether treatment parameters need to be adjusted, dosage changed, or treatment area replanned. Predictive Analysis: Using historical data and current treatment status, the system predicts possible future treatment outcomes, helping doctors make forward-looking treatment decisions.

[0147] ④ Output Guidance. Optimization Suggestions: Sends optimized treatment parameters and strategies to the treatment execution unit to ensure more precise and effective subsequent treatment operations. Alarms and Warnings: If a signal that may adversely affect the patient is detected, the system will immediately issue an alarm, prompting the medical team to take necessary intervention measures. Report Generation: Provides the medical team with detailed analysis reports, including treatment progress, effect evaluation, and potential risks, facilitating doctors' review and adjustment of treatment plans.

[0148] S420: Based on the collected data, advanced data analysis and machine learning algorithms are used to evaluate treatment effectiveness, predict potential complications, and optimize treatment strategies. Specifically:

[0149] ① Data Input and Preprocessing. Target Area Information: This includes the precise location, shape, and volume of the tumor or lesion, as well as its boundary with surrounding healthy tissue. This information is derived from high-resolution imaging data, such as MRI, CT, or PET scans. Treatment Parameters: Details of the electroporation pulses, such as pulse frequency, duration, and intensity. These are crucial factors ensuring precise targeting of the electroporation area. Real-time Monitoring Data: Feedback from a multimodal monitoring system, including electric field distribution, tissue temperature, and cell damage indicators, used to monitor treatment progress and effectiveness. Patient Physiological Data: This includes hemodynamic parameters, heart rate, and blood pressure, used to assess the patient's overall health and predict potential complications.

[0150] ② Data Analysis and Machine Learning Applications. Treatment Efficacy Assessment Model: Utilizing machine learning algorithms, such as random forests, neural networks, or support vector machines, to analyze pre- and post-treatment imaging changes, quantify tumor shrinkage, apoptosis rates, etc., to assess treatment efficacy. Complication Prediction Model: Based on historical datasets, models are trained to identify which treatment parameters or patient characteristics may lead to specific complications, such as cardiac arrhythmias and skin burns. Dosage Optimization Model: Using algorithms such as reinforcement learning, electroporation pulse parameters are adjusted based on real-time feedback to minimize the impact on healthy tissue while maximizing treatment efficacy.

[0151] ③ Result Interpretation and Strategy Adjustment. Intelligent Decision-Making: By comprehensively analyzing the model's output, the intelligent system can recommend adjustments to the treatment strategy, such as changing the pulse sequence, increasing or decreasing pulse energy, to achieve more ideal treatment outcomes. Personalized Treatment Plans: Based on each patient's unique situation, the system can generate customized treatment plans, ensuring precise dosage control while reducing the risk of complications. Dynamic Feedback Loop: The treatment decision-making and optimization center continuously collects new data from the treatment process, updates the analysis model, and forms a closed-loop system that continuously learns and improves.

[0152] ④ Outputs and Reports. Treatment Efficacy Report: Provided to doctors and patients, showcasing treatment progress, efficacy assessment, and potential risks. Strategy Optimization Suggestions: Sends optimized treatment parameters to the treatment execution unit to ensure more precise subsequent operations. Long-Term Prediction and Monitoring: Provides patients with long-term follow-up suggestions, including possible subsequent treatments or monitoring indicators.

[0153] Through the aforementioned process, the comprehensive analysis function ensures that the Treatment Decision and Optimization Center can efficiently utilize all available data, improve the personalization and safety of treatment through intelligent means, and provide strong support to clinicians to address the challenges of precise dose control and prediction. This mechanism maximizes the effectiveness of electroporation ablation therapy while protecting healthy tissue.

[0154] S430. Based on the comprehensive analysis results, the subsequent treatment plan is intelligently adjusted, including changing pulse sequence parameters, electrode configuration, or adding auxiliary treatment methods to improve the treatment success rate and patient safety. Specifically:

[0155] ① Data is received from sensors during treatment, including electrode configuration status, pulse sequence parameters (frequency, amplitude, duration), tissue responses (such as changes in conductivity and thermal effects), and patient physiological responses (such as heart rate and blood pressure). Treatment efficacy is assessed using the previously discussed machine learning models to analyze the impact of the current treatment on the target area, including tumor reduction and apoptosis rate, while simultaneously monitoring potential damage to healthy tissue. Complication prediction is based on historical data and patient-specific conditions to predict the risk of potential complications, such as local inflammatory responses, skin burns, or other organ dysfunction.

[0156] ② Based on the gap between current treatment parameters and expected treatment goals, the intelligent system evaluates the effectiveness of existing strategies. Treatment optimization: Using techniques such as reinforcement learning or genetic algorithms, the system simulates the effects of different combinations of treatment parameters to find the most likely approach to improve treatment success rate and patient safety. Integration of auxiliary treatments: Consideration is given to whether to introduce chemical drugs, photodynamic therapy, or other auxiliary treatments to enhance the effects of electroporation or reduce side effects.

[0157] ③ Pulse sequence parameter adjustment: Based on intelligent decision-making, the pulse frequency, intensity, and duration are fine-tuned to ensure that the electroporation pulses can act more accurately on the target area while avoiding damage to surrounding healthy tissues. Electrode configuration optimization: The optimal placement and orientation of the electrodes are recalculated to optimize the electric field distribution and improve electroporation efficiency.

[0158] ④ The intelligent system generates an updated treatment plan, including new pulse sequence parameters, electrode layout, and any additional adjunctive therapy suggestions. Execution and Supervision: The adjusted treatment plan is transmitted to the treatment execution unit to implement the modified treatment procedure, and treatment progress and patient response are continuously monitored in real time.

[0159] ⑤ Feedback Loop and Continuous Optimization. After treatment, the treatment effect and patient condition are reassessed, and data is collected as input for future intelligent decision-making. Learning and Iteration: The intelligent system learns from each treatment, gradually improving its predictive capabilities and strategy adjustment algorithms, forming a continuously improving feedback loop.

[0160] The key innovations of this invention include:

[0161] 1. Utilizing deep learning and computer vision technologies, tumor areas are automatically identified and healthy tissues are distinguished, ensuring precise treatment targeting.

[0162] 2. Establish a three-dimensional electric field distribution model, optimize the electric field distribution by combining parameters such as tissue conductivity, and predict the electric field effect.

[0163] 3. Design a closed-loop control system to automatically adjust the pulse sequence by real-time monitoring of electric field strength and tissue response, thereby achieving precise dose control.

[0164] 4. Integrate multi-source data and use advanced data analysis and machine learning algorithms to evaluate treatment effectiveness, predict complications, and optimize treatment strategies.

[0165] 5. Achieve dynamic feedback loop, intelligently adjust subsequent treatment plans based on treatment progress and individual patient differences, improve treatment success rate and patient safety.

[0166] This invention, "A Voltage Control Method and System for Electroporation Ablation Based on Pulse Sequence," introduces advanced technologies and algorithms to achieve precise control and personalized treatment in electroporation ablation therapy. It solves problems such as inaccurate target identification, uneven electric field distribution, uncontrollable treatment process, and difficulty in adjusting treatment strategies in traditional methods. Specific beneficial effects are as follows:

[0167] 1. Improved target recognition accuracy: By utilizing deep learning and computer vision technology, tumor areas in medical images are automatically identified and healthy tissues are accurately distinguished, ensuring that the electroporation pulses are precisely applied to the target area, reducing damage to healthy tissues and improving the safety and effectiveness of treatment.

[0168] 2. Optimization of electric field distribution: Through a closed-loop system of electric field modeling and dose control, the electrode layout and pulse parameters are dynamically adjusted to ensure that the target area receives a precise dose of electroporation pulse. At the same time, a multi-objective optimization algorithm is used to protect healthy tissues, thereby improving the accuracy and efficiency of treatment.

[0169] 3. Provides real-time monitoring and feedback: Integrates sensor and imaging technologies to monitor key parameters during treatment, such as electric field distribution, tissue temperature, and degree of cell damage, providing a basis for dose control and ensuring the controllability and safety of the treatment process.

[0170] 4. Personalized treatment strategies were implemented: Through the comprehensive analysis module, subsequent treatment plans were intelligently adjusted, including changing pulse sequence parameters, electrode configuration, or adding auxiliary treatment methods, which improved the success rate of treatment and patient safety, while reducing the risk of complications.

[0171] 5. Promotes intelligent treatment decision-making: Through machine learning algorithms, learning from historical treatment cases, continuously optimizing feedback control algorithms and multi-objective optimization strategies, improving the accuracy and efficiency of future treatments, and forming a feedback loop of continuous improvement.

[0172] 6. Reduced uncertainty in the treatment process: Through real-time monitoring and feedback mechanisms, as well as model predictive control (MPC) strategies, the precise adjustment of electrode voltage output is ensured, avoiding dose deviations caused by individual differences and treatment progress, and ensuring the consistency of treatment effects.

[0173] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0174] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0175] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process. Furthermore, the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functionality involved.

[0176] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus or device (such as a computer-based system, a processor-included system or other system that can fetch and execute instructions from, an instruction execution system, apparatus or device).

[0177] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. All or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware, the program being stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiments.

[0178] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. This storage medium can be a read-only memory, a disk, or an optical disk, etc.

[0179] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A voltage control system for electroporation ablation based on pulse sequences, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When executed, the program performs the following steps: S100 integrates target recognition and electric field modeling, using AI technology to automatically identify target regions in medical images, distinguish healthy tissues, and establish a three-dimensional electric field distribution model; S200, combining electric field optimization and dose control, forms a closed-loop control system that dynamically adjusts pulse parameters to optimize the electric field distribution and ensure the target area receives a precise dose. Specifically, by real-time monitoring of the electric field intensity and tissue response in the target area, the dose distribution D(x, t) of the target area is determined, and the current dose distribution D(x, t) is compared with the target dose distribution. Calculate the deviation Based on the deviation ΔD(x,t) and the tissue state matrix S(t), the electrode voltage output V(t) is dynamically adjusted. Let be the voltage adjustment coefficient, and ΔV(t) be the voltage increment. Then we have: ； in The integral term in the above formula represents the sensitivity of the electric field intensity to the voltage. It reflects the comprehensive response of the electric field intensity of the entire tumor region to voltage changes. The electrode voltage output V(t) is updated based on the calculated voltage increment ΔV(t). S300 integrates treatment effect assessment with sensor technology to monitor electric field strength, tissue state and dose distribution in real time, and provide feedback for dose control; The S400 integrates information from the target area, treatment parameters, and real-time feedback, performs comprehensive analysis, guides subsequent treatment strategy adjustments, and achieves personalized treatment.

2. The system according to claim 1, characterized in that, Step S100 includes: S110. Utilize deep learning and computer vision technologies to identify target regions and ensure accurate identification of target regions; S120. Based on the location, shape, and size of the target area, and combined with the tissue conductivity parameters, establish an electric field distribution model; S130. Generate optimal pulse sequence parameters to maximize the electroporation effect on tumor cells while minimizing the impact on healthy tissues.

3. The system according to claim 1, characterized in that, Step S300 integrates multiple sensors and imaging technologies to monitor key parameters during the treatment process in real time, providing a basis for dose control.

4. The system according to claim 1, characterized in that, In step S400, the integrated information is analyzed to guide the adjustment of the treatment strategy, including changing the pulse sequence parameters, electrode configuration, or adding auxiliary treatment methods.

5. An implementation device for a voltage control system of electroporation ablation based on pulse sequence, characterized in that, include: The target recognition module is used to identify target regions in medical images and distinguish healthy tissues; The electric field modeling module is used to establish electric field distribution models; The dose control module is used to form a closed-loop control system and dynamically adjust pulse parameters. Specifically, it determines the dose distribution D(x, t) of the target area by real-time monitoring of the electric field intensity and tissue state response, and compares the current dose distribution D(x, t) with the target dose distribution. Calculate the deviation Based on the deviation ΔD(x,t) and the tissue state matrix S(t), the electrode voltage output V(t) is dynamically adjusted. Let be the voltage adjustment coefficient, and ΔV(t) be the voltage increment. Then we have: ； in The integral term in the above formula represents the sensitivity of the electric field intensity to the voltage. It reflects the comprehensive response of the electric field intensity of the entire tumor region to voltage changes. The electrode voltage output V(t) is updated based on the calculated voltage increment ΔV(t). The monitoring and feedback module is used to monitor the treatment process in real time and provide feedback. The strategy adjustment module integrates target area, treatment parameters, and real-time feedback information to guide subsequent adjustments to the treatment strategy.

6. The apparatus according to claim 5, characterized in that, The target recognition module uses deep learning and computer vision technologies for accurate identification.

7. The apparatus according to claim 5, characterized in that, The monitoring and feedback module integrates multiple sensors and imaging technologies to monitor key parameters during the treatment process in real time.

8. The apparatus according to claim 5, characterized in that, The strategy adjustment module integrates information to guide personalized treatment strategy adjustments, thereby improving treatment effectiveness and safety.

Images