Feedback-driven laser lipolysis robot treatment parameter self-optimization method and system

Abstract

This invention discloses a feedback-driven self-optimization method and system for laser fat reduction robot treatment parameters, belonging to the field of intelligent medical and health information processing technology. The system includes: an intelligent sensing and acquisition module, a data processing and storage module, an intelligent analysis engine, a health guidance and interaction module, and an adaptive optimization module. The method includes: collecting treatment-related data and corresponding device status parameters and treatment context parameters; performing standardized, structured, and preprocessing data; establishing an individual baseline based on historical data; calling the intelligent analysis engine to output health insights in conjunction with the individual baseline; mapping the health insights into personalized treatment parameter recommendations and outputting them; collecting post-treatment feedback and subsequent data to evaluate the parameter effects; updating the analysis strategy based on the parameter effects and forming a closed-loop iteration. This invention, using the above method and system, achieves continuous personalized optimization of treatment parameters through closed-loop iteration, improving the accuracy, safety, and long-term stability of treatment.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical and health information processing technology, and in particular to a feedback-driven method and system for self-optimization of laser fat reduction robot treatment parameters. Background Technology

[0002] With the increasing popularity of home and clinical beauty / rehabilitation energy therapy devices, users can perform soft tissue treatments and body shaping management using devices such as lasers, radiofrequency, and ultrasound. Among these, laser fat reduction robots can apply energy to target areas without contact and are typically combined with cooling and temperature monitoring to improve comfort and safety. However, the following shortcomings still exist in practical applications: First, treatment data is fragmented and lacks continuous trend management: Existing equipment mainly focuses on single treatments or single recordings, lacking the ability to uniformly organize, align, and analyze trends in historical treatment parameters, changes in body surface temperature, differences in treatment sites, and post-treatment reactions. Users find it difficult to form interpretable treatment management based on long-term changes, and doctors also find it difficult to quickly obtain structured, continuous treatment response information. Second, parameter settings rely on fixed protocols or manual experience, lacking individualization: Current laser fat reduction treatments often use fixed parameter combinations or manual adjustments based on experience, making it difficult to adapt to differences and long-term changes in skin sensitivity, fat thickness, and tissue thermal response among different individuals. Fixed strategies may lead to ineffective treatment when there are changes in weight, lifestyle habits, skin condition, or treatment sites. Issues such as instability, decreased comfort, or untimely risk warnings arise. Furthermore, the lack of a feedback loop prevents iterative optimization of parameter effects: many systems, even when providing parameter suggestions or default treatment plans, offer only one-way output, lacking a mechanism for continuous collection and quantitative evaluation of subjective feedback (such as pain, burning sensation, and comfort) and objective changes (such as surface temperature curves, skin reactions, and phased changes in body shape / circumference). This makes it impossible to determine the effectiveness of parameter combinations, let alone dynamically adjust subsequent treatment parameters, hindering the implementation of continuously improving personalized treatment strategies. Finally, long-term stability and controllability are insufficient: during long-term operation, changes in context such as treatment site, treatment distance and posture, cooling conditions, and ambient temperature, as well as data distribution drift, can affect the accuracy of analysis and recommendations. The lack of verification and safety constraint mechanisms for strategy updates may lead to risks such as overheating, persistent hot spots, or inappropriate parameter recommendations, reducing treatment safety and controllability.

[0003] Therefore, there is an urgent need for a technology solution that can be applied to laser fat reduction robots, which can perform standardized processing and intelligent analysis based on the acquisition of treatment-related data, and can adaptively optimize the analysis strategy and parameter recommendation strategy by combining post-treatment feedback and subsequent data changes, forming a continuously iterative closed-loop personalized recommendation and safety control technology solution for treatment parameters, so as to improve the personalization, accuracy, timeliness and safety of treatment parameter settings. Summary of the Invention

[0004] The purpose of this invention is to provide a feedback-driven method and system for self-optimizing laser fat reduction robot treatment parameters, which improves the continuity, individualization and timeliness of laser fat reduction treatment parameter settings, and maintains the controllability and stability of strategy updates during long-term use, thereby reducing the risk of adverse reactions.

[0005] To achieve the above objectives, this invention provides a feedback-driven self-optimization method and system for laser fat reduction robot treatment parameters, comprising the following steps: S1. Collect treatment-related data output by the laser fat reduction robot during and after treatment, and simultaneously collect the device status parameters and treatment context parameters corresponding to the treatment-related data; S2. Preprocess the treatment-related data, including data cleaning and outlier handling. S3. Standardize and structure the preprocessed data, including unit normalization, timestamp alignment and segmentation, to obtain structured data and store it in the database. S4. Establish individual baselines based on historical structured data within a preset time window. Individual baselines include at least one or more of the following: mean, fluctuation range, or stability indicators. S5. Call the intelligent analysis engine to perform multi-dimensional analysis on the newly added structured data, and output health insights in combination with individual baselines. Health insights include at least one or more of the following: abnormal events, risk levels, or trend conclusions. S6. Map health insights into structured personalized treatment parameter recommendation results and output them to the interactive interface. The personalized treatment parameter recommendation results shall include at least the parameter type and parameter value. The parameter type shall include at least one or more of the following: laser output related parameters, cooling related parameters, or scan / trajectory related parameters. S7. Collect user feedback on the treatment corresponding to the personalized treatment parameter recommendation results and subsequent newly added treatment-related data, and evaluate the effect of the parameters based on the post-treatment feedback and subsequent newly added treatment-related data. S8. Update the analysis strategy based on the parameter effect. The analysis strategy includes at least one or more of individualized thresholds, feature weights, or model parameters. Then, return to step S5 based on the updated analysis strategy to form a closed-loop iteration.

[0006] Preferably, the treatment-related data includes one or more of the following: treatment parameter data, body surface temperature data, image data, or user feedback data; Equipment status parameters include one or more of the following: equipment operating mode, laser output power or duty cycle status, cooling operating status, or self-test status. Treatment context parameters include one or more of the following: treatment site identifier, treatment distance, scan trajectory / velocity, or attitude parameters.

[0007] Preferably, the preprocessing includes one or more of the following: noise reduction, missing value marking, duplicate value removal, and outlier detection and correction; the database is a secure database, employing at least one of the following: data encryption storage and access control.

[0008] Preferably, multidimensional analysis includes: extracting features from structured data and comparing the extracted features with individual baselines or group reference intervals to output risk level or trend conclusions.

[0009] Preferably, the laser output-related parameters include at least power density, duty cycle, duration of action, or energy level; Cooling-related parameters should include at least the air cooling intensity, fan speed or flow rate, or cooling sequence. Scan / trajectory related parameters include at least scan speed, trajectory spacing, dwell time, or coverage mode.

[0010] Preferably, post-treatment feedback includes one or more of the following: user subjective feelings input, skin reaction records, or supplementary measurement results; The parameter effects are calculated based on the magnitude or trend of changes in post-treatment feedback and subsequent new treatment-related data.

[0011] Preferably, the health insights also include a summary of triggering causes, which includes at least one or more of the following: the triggering threshold, the comparison window, or the magnitude of change.

[0012] Preferably, the update in step S8 is an incremental update, and the candidate strategy is verified before the updated analysis strategy is enabled; when the verification result does not meet the preset security constraints or performance threshold, the candidate strategy is disabled and the current analysis strategy remains unchanged.

[0013] Preferably, the safety constraints include at least one or more of the following: upper limit constraint on body surface temperature, upper limit constraint on temperature rise rate, upper limit constraint on hot spot duration, upper limit constraint on laser output power / energy, or lower limit constraint on cooling intensity.

[0014] This invention also provides a feedback-driven self-optimization system for laser fat reduction robot treatment parameters, comprising: The intelligent sensing and acquisition module is used to collect treatment-related data output by the laser fat reduction robot during and after treatment, and simultaneously collect the device status parameters and treatment context parameters corresponding to the treatment-related data. The data processing and storage module is used to preprocess and standardize the structured data related to treatment before storing it in the database; The intelligent analysis engine is used to establish an individual baseline based on historical structured data within a preset time window, and to perform multi-dimensional analysis on newly added structured data, and output health insights in combination with the individual baseline. The health guidance interaction module is used to map health insights into structured personalized treatment parameter recommendations and output them, while collecting user feedback on the treatment corresponding to the personalized treatment parameter recommendations after treatment. The adaptive optimization module is used to evaluate the effectiveness of parameters based on post-treatment feedback and subsequent new treatment-related data, and to update the analysis strategy based on the parameter effectiveness. The analysis strategy includes at least one or more of individualized thresholds, feature weights, or model parameters. The updated analysis strategy is fed back to the intelligent analysis engine for the next round of health insight output, forming a closed-loop iteration.

[0015] Therefore, the present invention employs the above-mentioned feedback-driven self-optimization method and system for laser fat reduction robot treatment parameters, which has the following advantages: 1) Unified structuring and continuous management: Standardize and structure data related to laser fat reduction robot treatment and align it with time to facilitate long-term storage, comparison and trend analysis; 2) Personalized analysis based on individual baselines: By using individual baselines and individualized safety boundaries / reference intervals, the individual adaptability of anomaly detection, risk assessment and trend identification is improved; 3) Recommendations are traceable and assessable: Structured parameter recommendation results are linked with the summary of triggering reasons for easy understanding, execution, and tracking of effects; 4) Feedback-driven self-optimization iteration: The effect of parameters is evaluated by post-treatment feedback and subsequent data changes, and the thresholds / weights / parameters are updated to achieve continuous optimization that "becomes more and more suitable with use"; 5) Controllable updates: Candidate strategies are validated and subject to safety constraints before implementation, reducing the risk of inappropriate recommendations during long-term operation and improving long-term operational stability and treatment safety.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] Figure 1 This is a system architecture diagram of an embodiment of the present invention; Figure 2This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0019] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0020] Example like Figure 1 As shown, the system of the present invention includes at least: an intelligent sensing and acquisition module, a data processing and storage module, an intelligent analysis engine, a health guidance and interaction module, and an adaptive optimization module.

[0021] Among them, the intelligent sensing and acquisition module is used to collect treatment-related data output by the laser fat reduction robot during and after treatment, and simultaneously collect the device status parameters and treatment context parameters corresponding to the treatment-related data; The data processing and storage module is used to preprocess and standardize the structured data related to treatment before storing it in the database; The intelligent analysis engine is used to establish an individual baseline based on historical structured data within a preset time window, and to perform multi-dimensional analysis on newly added structured data, and output health insights in combination with the individual baseline; The health guidance interaction module is used to map health insights into structured personalized treatment parameter recommendations and output them, while collecting user feedback on the treatment corresponding to the personalized treatment parameter recommendations after treatment. The adaptive optimization module is used to evaluate the effectiveness of parameters based on post-treatment feedback and subsequent new treatment-related data, and to update the analysis strategy based on the effect of the parameters. The analysis strategy includes at least one or more of individualized thresholds, feature weights, or model parameters. The updated analysis strategy is fed back to the intelligent analysis engine for the next round of health insight output, forming a closed-loop iteration.

[0022] like Figure 2As shown, the method of the present invention includes at least the following steps: S1. Data Acquisition: The intelligent sensing and acquisition module communicates with the laser fat reduction robot to obtain treatment-related data D, and simultaneously acquires the device status parameters S and treatment context parameters X corresponding to the treatment-related data.

[0023] Among them, the treatment-related data D includes at least one or more of the following: treatment parameter data, body surface temperature data, cooling operation data, or image data; the device status parameters S include at least one or more of the following: device operating mode, laser output status, cooling operation status, or self-test status; and the treatment context parameters X include at least one or more of the following: treatment time, treatment duration, treatment site identification, treatment distance, robotic arm posture, or trajectory parameters.

[0024] S2, Preprocessing: The treatment-related data D is preprocessed, including at least one of the following: missing value labeling, duplicate value removal, outlier detection and correction; outlier detection can be implemented using the rule of "labeling if it is outside the physical feasible range" or "labeling if the deviation from the nearest window exceeds a threshold".

[0025] S3, Standardized Structure, Alignment, and Storage: The preprocessed data is standardized and structured, including unit normalization and timestamp alignment. The processed data is then segmented into structured data R according to a preset window (e.g., a treatment process or a continuous segment of N seconds / minutes) and stored in the database.

[0026] Structured data R must contain at least the following: user identifier, device identifier, timestamp, data type, numeric field, device status parameter, treatment context parameter, and data quality marker field.

[0027] S4. Individual baseline establishment: An individual baseline B is established based on historical structured data within a preset time window (e.g., the last 7 days, the last M effective treatments, or the last N effective records). The individual baseline B includes at least one or more of the mean and fluctuation range of key indicators, and forms an individualized reference interval or individualized safety boundary for subsequent risk assessment and parameter recommendation constraints.

[0028] S5, Intelligent Analysis and Health Insight Generation: The intelligent analysis engine is invoked to perform multi-dimensional analysis on the newly added structured data R, and combined with the individual baseline B to output health insights I; health insights I include at least one or more of the following: abnormal events, risk levels, or trend conclusions.

[0029] in: Abnormal events can be triggered by “body surface temperature exceeding individualized safety boundaries”, “abnormally increased rate of temperature rise”, “hot spot duration exceeding threshold”, or “abnormal combination of parameter states”. Risk levels can be generated by a combination of rules: the extent of exceeding the limit + the number of times or the duration + the treatment context (location / distance / cooldown status); Trend conclusions can be generated from moving average change rates or window slopes, for example, to determine whether the temperature response to multiple treatments at the same site gradually increases or gradually stabilizes.

[0030] For ease of interpretation, Health Insight I can simultaneously output a summary of the triggering cause (e.g., comparison window, exceeding the limit, duration, cooling level and distance conditions, etc.).

[0031] S6. Generation and output of personalized treatment parameter recommendations: The health insight I is mapped to a structured personalized treatment parameter recommendation result G and output to the interactive interface; the personalized treatment parameter recommendation result G includes at least the parameter type and parameter value, and may include a summary of the triggering reason and safety prompts.

[0032] The parameter types include at least one or more of the following: laser output related parameters (power density / duty cycle / duty time / energy level) and cooling related parameters (air cooling intensity / wind speed or flow rate level / cooling sequence), and may further include scanning or trajectory related parameters (scanning speed / trajectory spacing / dwell time / coverage mode).

[0033] S7. Feedback Collection and Effect Evaluation: Collect user feedback F on the treatment corresponding to the personalized treatment parameter recommendation results and subsequent newly added treatment-related data D'; the post-treatment feedback F includes at least one or more of the following: subjective feeling input (comfort, pain, burning sensation), skin reaction record (erythema, burning duration), or supplementary measurement results; The effect of parameter E is evaluated based on post-treatment feedback F and subsequent newly added treatment-related data D'. The effect of parameter E includes at least one or more of the following: whether the risk level has decreased, whether the abnormal event has been resolved, whether the body surface temperature curve has returned to the individualized safety boundary, and whether the comfort level has improved or the trend has improved.

[0034] S8, Adaptive Optimization and Closed-Loop Iteration: The analysis strategy P is updated based on the parameter effect E. The analysis strategy P includes at least one or more of individualized thresholds, feature weights, or model parameters. The updated analysis strategy is used in the next round of steps S5 to S7 to form a closed loop iteration.

[0035] Before updating, the candidate strategy is verified. If the verification result does not meet the preset security constraints or performance threshold, the candidate strategy is not enabled and the current analysis strategy remains unchanged.

[0036] Example 1 Taking a laser fat reduction robot as an example, this device can output measurement data related to the treatment process as treatment-related data D during the treatment, and simultaneously acquire treatment time, treatment duration, treatment site identification, etc. as treatment context parameters X, and device working mode, self-test status, etc. as device status parameters S. The system follows... Figure 2 Process execution: First, the intelligent sensing and acquisition module acquires the aforementioned data R; The data processing and storage module performs standardized and structured processing and preprocessing on data R before storing it in the database; During the individual baseline establishment phase, the system establishes individual baseline B based on historical structured data within a preset time window and generates an individual reference interval; The intelligent analysis engine compares the newly added structured data with the individual baseline and outputs Health Insight I, which includes abnormal events, risk levels, or trend conclusions. The health guidance interaction module maps health insights I into structured personalized treatment parameter recommendation results G and outputs them. The personalized treatment parameter recommendation results include risk warning prompts, retest / follow-up prompts, or treatment process management prompts. The system then collects the user's post-treatment feedback F on the personalized treatment parameter recommendation results and the newly added treatment-related data D', and evaluates the parameter effect E based on this; The adaptive optimization module incrementally updates the analysis strategy P based on the parameter effect E and verifies it before activation to achieve long-term stable closed-loop iteration, so that the personalized treatment parameter recommendation results gradually conform to individual characteristics as the user's long-term data becomes available.

[0037] Therefore, the present invention employs the above-mentioned feedback-driven self-optimization method and system for laser fat reduction robot treatment parameters, which has the following advantages: 1) Unified structuring and continuous management: Standardize and structure data related to laser fat reduction robot treatment and align it with time to facilitate long-term storage, comparison and trend analysis; 2) Personalized analysis based on individual baselines: By using individual baselines and individualized safety boundaries / reference intervals, the individual adaptability of anomaly detection, risk assessment and trend identification is improved; 3) Recommendations are traceable and assessable: Structured parameter recommendation results are linked with the summary of triggering reasons for easy understanding, execution, and tracking of effects; 4) Feedback-driven self-optimization iteration: The effect of parameters is evaluated by post-treatment feedback and subsequent data changes, and the thresholds / weights / parameters are updated to achieve continuous optimization that "becomes more and more suitable with use"; 5) Controllable updates: Candidate strategies are validated and subject to safety constraints before implementation, reducing the risk of inappropriate recommendations during long-term operation and improving long-term operational stability and treatment safety.

[0038] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A feedback-driven self-optimization method for laser fat reduction robot treatment parameters, characterized in that, Includes the following steps: S1. Collect treatment-related data output by the laser fat reduction robot during and after treatment, and simultaneously collect the device status parameters and treatment context parameters corresponding to the treatment-related data; S2. Preprocess the treatment-related data, including data cleaning and outlier handling. S3. Standardize and structure the preprocessed data, including unit normalization, timestamp alignment and segmentation, to obtain structured data and store it in the database. S4. Establish individual baselines based on historical structured data within a preset time window. Individual baselines include at least one or more of the following: mean, fluctuation range, or stability indicators. S5. Call the intelligent analysis engine to perform multi-dimensional analysis on the newly added structured data, and output health insights in combination with individual baselines. Health insights include at least one or more of the following: abnormal events, risk levels, or trend conclusions. S6. Map health insights into structured personalized treatment parameter recommendation results and output them to the interactive interface. The personalized treatment parameter recommendation results shall include at least the parameter type and parameter value. The parameter type shall include at least one or more of the following: laser output related parameters, cooling related parameters, or scan / trajectory related parameters. S7. Collect user feedback on the treatment corresponding to the personalized treatment parameter recommendation results and subsequent newly added treatment-related data, and evaluate the effect of the parameters based on the post-treatment feedback and subsequent newly added treatment-related data. S8. Update the analysis strategy based on the parameter effect. The analysis strategy includes at least one or more of individualized thresholds, feature weights, or model parameters. Then, return to step S5 based on the updated analysis strategy to form a closed-loop iteration.

2. The feedback-driven self-optimization method for laser fat reduction robot treatment parameters according to claim 1, characterized in that, Treatment-related data includes one or more of the following: treatment parameter data, body surface temperature data, image data, or user feedback data; Equipment status parameters include one or more of the following: equipment operating mode, laser output power or duty cycle status, cooling operating status, or self-test status. Treatment context parameters include one or more of the following: treatment site identifier, treatment distance, scan trajectory / velocity, or attitude parameters.

3. The feedback-driven self-optimization method for laser fat reduction robot treatment parameters according to claim 1, characterized in that, Preprocessing includes one or more of the following: noise reduction, missing value labeling, duplicate value removal, and outlier detection and correction. The database is a secure database, employing at least one of the following: data encryption storage and access control.

4. The feedback-driven self-optimization method for laser fat reduction robot treatment parameters according to claim 1, characterized in that, Multidimensional analysis includes: extracting features from structured data and comparing the extracted features with individual baselines or group reference intervals to output risk levels or trend conclusions.

5. The feedback-driven self-optimization method for laser fat reduction robot treatment parameters according to claim 1, characterized in that, Laser output parameters include at least power density, duty cycle, duration of action, or energy level. Cooling-related parameters should include at least air cooling intensity, fan speed / flow rate setting, or cooling sequence. Scan / trajectory related parameters include at least scan speed, trajectory spacing, dwell time, or coverage mode.

6. The feedback-driven self-optimization method for laser fat reduction robot treatment parameters according to claim 1, characterized in that, Post-treatment feedback includes one or more of the following: user subjective feelings input, skin reaction records, or supplementary measurement results; The parameter effects are calculated based on the magnitude or trend of changes in post-treatment feedback and subsequent new treatment-related data.

7. The feedback-driven self-optimization method for laser fat reduction robot treatment parameters according to claim 1, characterized in that, Health insights also include a summary of trigger reasons, which includes at least one or more of the following: the trigger threshold, the comparison window, or the magnitude of change.

8. The feedback-driven self-optimization method for laser fat reduction robot treatment parameters according to claim 1, characterized in that, The update in step S8 is an incremental update, and the candidate strategy is verified before the updated analysis strategy is enabled. When the verification result does not meet the preset security constraints or performance threshold, the candidate strategy is disabled and the current analysis strategy remains unchanged.

9. The feedback-driven self-optimization method for laser fat reduction robot treatment parameters according to claim 8, characterized in that, Safety constraints include at least one or more of the following: upper limit constraint on body surface temperature, upper limit constraint on temperature rise rate, upper limit constraint on hot spot duration, upper limit constraint on laser output power / energy, or lower limit constraint on cooling intensity.

10. A feedback-driven self-optimization system for laser fat reduction robot treatment parameters, used to execute the feedback-driven self-optimization method for laser fat reduction robot treatment parameters as described in any one of claims 1-9, characterized in that, include: The intelligent sensing and acquisition module is used to collect treatment-related data output by the laser fat reduction robot during and after treatment, and simultaneously collect the device status parameters and treatment context parameters corresponding to the treatment-related data. The data processing and storage module is used to preprocess and standardize the structured data related to treatment before storing it in the database; The intelligent analysis engine is used to establish an individual baseline based on historical structured data within a preset time window, and to perform multi-dimensional analysis on newly added structured data, and output health insights in combination with the individual baseline. The health guidance interaction module is used to map health insights into structured personalized treatment parameter recommendations and output them, while collecting user feedback on the treatment corresponding to the personalized treatment parameter recommendations after treatment. The adaptive optimization module is used to evaluate the effectiveness of parameters based on post-treatment feedback and subsequent new treatment-related data, and to update the analysis strategy based on the parameter effectiveness. The analysis strategy includes at least one or more of individualized thresholds, feature weights, or model parameters. The updated analysis strategy is fed back to the intelligent analysis engine for the next round of health insight output, forming a closed-loop iteration.

Images