Non-invasive full-automatic cancer diagnosis system based on microfluidic technology

Abstract

The invention discloses a non-invasive full-automatic cancer diagnosis system based on a micro-fluidic technology, and relates to the technical field of medical treatment, the system calculates and outputs a splitting decomposition response efficiency REI through a dynamic photo-thermal response adaptive splitting decomposition module, and comprehensively evaluates the splitting decomposition efficiency of a micro-capsule body. And further performing preliminary comparison on the cracking response efficiency REI through a cracking evaluation unit, and when the cracking response efficiency REI is between 3.0 and 5.0, automatically triggering a thermal-optical coordination auxiliary imaging module, and calculating and outputting a photothermal offset compensation value Gcoord for evaluating an imaging offset trend and an image response lag degree. And when the photo-thermal offset compensation value Gcoord is higher than a set threshold value, the system can automatically execute an adjustment strategy, prolong the heating area time of the electric heating element and finely adjust the LED irradiation eccentric angle, so that the dynamic correction of photo-thermal excitation paths in different sample states is realized, and the sufficiency of micro-capsule splitting decomposition and the stability of an imaging center are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and specifically to a non-invasive fully automatic cancer diagnosis system based on microfluidic technology. Background Art

[0002] Microfluidic technology has become a revolutionary technology in the field of molecular diagnosis, especially in the detection of cancer and infectious diseases. By integrating multiple biological technologies into a micro-scale detection platform, microfluidic devices can achieve highly efficient, pollution-free, high-precision, and high-accuracy detection with extremely small sample volumes. At the same time, in the diagnosis of cancer and infectious diseases, microfluidic technology has been widely used in nucleic acid detection and analysis. These developed microfluidic technology platforms have made significant progress in detection speed and convenience, providing a new option for traditional detection systems such as PCR that are complex and time-consuming. Many microfluidic chips do not rely on amplification technology, which can further accelerate the result generation speed.

[0003] At present, the mainstream cancer screening devices mostly adopt the blood sample collection + laboratory analysis process, which involves multiple complex steps such as specimen collection, centrifugal separation, marker fluorescence detection, and subsequent PCR. Although these systems have relatively high sensitivity, they generally have a series of problems such as long detection cycles, strong dependence on the detection environment, large device volumes, and frequent manual interventions. In addition, although some recently emerging portable POCT devices attempt to use microfluidic chips for detection, most of them still remain at the level of manual reading or fixed-condition detection, lacking an automatic photothermal regulation mechanism, being highly sensitive to sample individual differences such as viscosity and protein loading, and prone to technical bottlenecks such as false negatives or unreadable results due to insufficient release of microcapsules or image imaging deviation. The main reason for this current situation is that existing portable detection systems generally fail to establish a dynamic response coupling relationship between sample physical properties and reaction conditions, especially lacking a closed-loop control mechanism for the release behavior of microcapsules, image offset conditions, and imaging stability. When the sample in actual detection has characteristics such as high viscosity, high molecular background, or incomplete biotin release, the fixed-power LED irradiation or constant heating mechanism cannot adapt to the behavior patterns of different samples, resulting in insufficient lysis of microcapsules, insufficient biotin release, or severe offset of the imaging area, further affecting the readability of the test strip and the reliability of the diagnostic results. Such reaction mismatches not only reduce the detection sensitivity and repeatability but also easily lead to false positives, false negatives, or even completely no result output, severely restricting the feasibility and credibility of rapid cancer screening in non-professional environments. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a non-invasive fully automatic cancer diagnosis system based on microfluidic technology, which solves the problems mentioned in the background art.

[0005] To achieve the above object, the present invention is realized through the following technical solutions: A non-invasive fully automatic cancer diagnosis system based on microfluidic technology, including a sample loading module, a sample data processing module, a dynamic photothermal response adaptive lysis module, a thermo-optical coordinated auxiliary imaging module, and an imaging stability analysis module; The sample loading module sets a microfluidic chip and sets data acquisition nodes in the microfluidic chip to collect sample data in real time, and sets a central control processor in the microfluidic chip to wirelessly transmit the sample data to the central control processor; The sample data processing module preprocesses the sample data in the central control processor to obtain a standardized digital set, constructs a sample database, and stores the standardized digital set; The dynamic photothermal response adaptive lysis module extracts the standardized digital set, calculates and outputs the lysis response efficiency REI, and based on the output result of the lysis response efficiency REI, conducts a preliminary comparative evaluation to judge the lysis situation; When the thermo-optical coordinated auxiliary imaging module preliminarily compares and evaluates that the lysis is abnormal, it calculates and outputs the thermo-optical offset compensation value number Gcoord based on the lysis response efficiency REI, and conducts a secondary comparative evaluation based on the output result of the thermo-optical offset compensation value number Gcoord to judge the imaging conditions; The imaging stability analysis module calculates and outputs the imaging stability evaluation index FSI through the thermo-optical offset compensation value number Gcoord, and conducts a comprehensive evaluation based on the output result of the imaging stability evaluation index FSI to judge the stability of the imaging.

[0006] Preferably, the sample loading module includes a sample loading unit and a sample transmission unit; The sample loading unit loads a saliva sample into a disposable microfluidic chip, and sets data acquisition nodes in the microfluidic chip to collect sample data in real time; The microfluidic chip is embedded with electric heating elements and LED irradiation windows, and microcapsules are set in the detection channels of the microfluidic chip; The data acquisition nodes include point A, point B, point C, and point D; Among them, point A represents the sample pretreatment area, point B represents the SPR photothermal cavity, point C represents the release channel, and point D represents the front end of the test strip reaction; The sample data includes flow resistance, pre-irradiation reflectivity, post-irradiation reflectivity, micro-resistance fluctuation, channel resistance, and the imaging area center offset angle Pdef; Point A integrates a high-frequency oscillating viscosity sensing array at the inlet of the microfluidic chip. When the saliva sample flows through, the saliva sample is slightly heated by the electrothermal element to a temperature increase of 2 degrees Celsius. At the same time, a frequency oscillation is applied through the high-frequency oscillating viscosity sensing array to detect the damping change of the oscillation signal by the sample and obtain the flow resistance; Point B sets a micro-optical detection probe group in the area where gold nanoparticles are distributed in the microfluidic chip. After the saliva sample is loaded, the electrothermal element and the LED irradiation window are activated to irradiate the gold nanoparticles, convert the light energy into local heat energy, heat the microcapsules, release biotin by cracking, and stimulate the surface plasmon resonance (SPR) effect. Through the micro-optical detection probe group, the position and intensity of the PR reflection peak are detected in real time as they change with the interaction between the gold nanoparticles and the light irradiation, and the reflectance before irradiation and the reflectance after irradiation are recorded; Point C embeds an embedded microelectrode pair inside the release channel in the microfluidic chip. After the microcapsules are lysed and the released contents are measured, the micro-resistance fluctuation is measured by the embedded microelectrode pair at a sampling frequency of 10 kHz; Point D sets an embedded CMOS image sensor upstream of the test strip in the microfluidic chip to track in real time the center position of the stripe generated by the development reaction of the contents released by the microcapsules, and compare the deviation angle between the ideal central axis and the actual imaging position to obtain the central offset angle Pdef of the imaging area; The sample transmission unit sets a central control processor in the microfluidic chip and transmits the sample data obtained in real time to the central control processor through Bluetooth communication in the microfluidic chip.

[0007] Preferably, the sample data processing module includes a preprocessing unit and a data storage unit; The preprocessing unit receives the sample data in real time in the central control processor and preprocesses the sample data to obtain a standardized digital set; The preprocessing includes feature extraction and standardization processing; The standardized digital set includes the relative viscosity change rate Urel, the local SPR response change rate Espr, the transient lysis resistance signal Srupt, and the central offset angle Pdef of the imaging area; The feature extraction combines and calculates the sample data set in the central control processor to obtain the relative viscosity change rate Urel, the local SPR response change rate Espr, and the transient lysis resistance signal Srupt; The above-mentioned standardization process uses the Z-Score standardization method to standardize the rate of change of viscosity Urel, the rate of change of local SPR response Espr, and the transient cracking resistance signal Srupt obtained by feature extraction, combined with the central offset angle Pdef of the imaging area, to remove the dimensional influence between parameters, convert them into computer numbers, and then summarize the standardized parameters to obtain a standardized digital set. The data storage unit constructs a sample database in the microfluidic chip, connects the sample database to the central control processor, and sets an automatic write port and an automatic read port for the sample database. After preprocessing, the real-time obtained standardized digital set is automatically stored in the sample database.

[0008] Preferably, the dynamic optothermal response adaptive cracking module includes a cracking analysis unit and a cracking evaluation unit. The cracking analysis unit constructs a non-linear coupling mapping algorithm model. The non-linear coupling mapping algorithm model performs comprehensive modeling by inputting energy, sample characteristics, release difficulty, and heating time. Then, through the automatic read port, the real-time obtained standardized digital set is extracted and input into the non-linear coupling mapping algorithm model for calculation to output the cracking response efficiency REI and analyze the effectiveness of microcapsule cracking.

[0009] Preferably, the cracking evaluation unit performs preliminary comparative evaluation based on the output result of the cracking response efficiency REI, judges the cracking situation of the microcapsules in the microfluidic chip, and triggers the optothermal coordinated auxiliary imaging module based on the evaluation result. The specific evaluation content is as follows: When the cracking response efficiency REI ≥ 5.0, it indicates normal cracking. At this time, direct imaging is performed and the test strip enters the color development stage. When 3.0 ≤ cracking response efficiency REI < 5.0, it indicates abnormal cracking. At this time, the optothermal coordinated auxiliary imaging module is triggered. When the cracking response efficiency REI < 3.0, it indicates cracking failure. At this time, the detection is stopped and re-sampling is prompted.

[0010] Preferably, the optothermal coordinated auxiliary imaging module includes an optothermal offset compensation analysis unit and an execution unit. After preliminary comparative evaluation shows abnormal cracking, the optothermal offset compensation analysis unit calculates and outputs the optothermal offset compensation value Gcoord based on the currently obtained cracking response efficiency REI, combined with the central offset angle Pdef of the imaging area, to adaptively compensate for signal imaging deviation and intensity loss.

[0011] Preferably, the execution unit performs secondary comparative evaluation based on the output result of the optothermal offset compensation value Gcoord, and executes an adjustment strategy based on the secondary comparative evaluation result. The specific evaluation content is as follows: When the photothermal offset compensation value number Gcoord ≥ 4.0, the adjustment strategy is triggered and executed at this time; When the photothermal offset compensation value number Gcoord < 4.0, there is no need to adjust at this time, and direct development is carried out; The adjustment strategy sends a control instruction to the electrothermal element and the LED irradiation window through the central control processor, automatically extends the heating area time of the electrothermal element by +5s, and at the same time finely adjusts the eccentric angle of the irradiation window by 5°. After the adjustment, the iterative execution system is carried out until the development ends the iteration.

[0012] Preferably, the imaging stability analysis module includes an imaging data extraction unit, an imaging analysis unit, and an imaging stability evaluation unit; The imaging data extraction unit performs image intensity analysis and colorimetric rate fitting on the test strip area after development; The image intensity analysis of the test strip area is performed by using an embedded CMOS image sensor to perform gray intensity integration processing on the developed strip image to obtain the average pixel value Iavg of the central area and the background area value Ibg; The colorimetric rate fitting is performed by recording the reaction development time tpeak; And data standardization processing is performed on the average pixel value Iavg, the background area value Ibg, and the reaction development time tpeak to eliminate the parameter dimension.

[0013] Preferably, the imaging analysis unit calculates and outputs the imaging stability evaluation index FSI based on the photothermal offset compensation value number Gcoord output after development, in combination with the obtained average pixel value Iavg, the background area value Ibg, and the reaction development time tpeak, and analyzes the readability and model uniformity of the image strip.

[0014] Preferably, the imaging stability evaluation unit performs a comprehensive evaluation based on the output result of the imaging stability evaluation index FSI to judge the stability of the imaging. The specific evaluation content is as follows; When the imaging stability evaluation index FSI ≥ 2.5, it indicates that the imaging is stable, and the automatic result is output, and the current imaging content is remotely transmitted to the diagnostic user terminal; When the imaging stability evaluation index FSI < 2.5, it indicates that the imaging is abnormal. At this time, the transmission is blocked, and the microfluidic chip is restarted for re-imaging.

[0015] The present invention provides a non-invasive fully automatic cancer diagnosis system based on microfluidic technology. It has the following beneficial effects: (1) The system sets four highly sensitive data acquisition nodes, namely point A, point B, point C, and point D, in the microfluidic chip to collect sample data in real time respectively. And the sample data is subjected to feature extraction and Z-score normalization processing in the central control processor to construct a normalized digital set. Without changing the hardware conditions, this structure significantly improves the adaptability of the system to characteristics such as the viscosity change of complex samples, the difference in photothermal response, and the microcapsule lysis behavior, provides a unified data basis for subsequent photothermal regulation and imaging quality evaluation, and effectively enhances the detection stability and spectral adaptability of the system.

[0016] (2) The system constructs a non-linear coupling mapping model of the lysis response efficiency REI through the dynamic photothermal response adaptation lysis module, and comprehensively evaluates the lysis efficiency of microcapsules based on factors such as energy input, sample viscosity, release resistance, and LED irradiation time. Further, the lysis evaluation unit makes a preliminary comparison of the lysis response efficiency REI. When the lysis response efficiency REI is between 3.0 and 5.0, the photothermal coordination assisted imaging module is automatically triggered to calculate and output the photothermal offset compensation value Gcoord, which is used to evaluate the imaging offset trend and the image response lag degree. When the photothermal offset compensation value Gcoord is higher than the set threshold, the system will automatically execute the adjustment strategy, extend the heating area time of the electrothermal element and fine-tune the LED irradiation eccentric angle, realizing the dynamic correction of the photothermal excitation path under different sample states, thereby significantly improving the sufficiency of microcapsule lysis and the stability of the imaging center.

[0017] (3) The system extracts three types of image feature parameters, namely the average pixel value Iavg, the background area value Ibg, and the reaction development time tpeak, through the imaging stability analysis module after the test strip development is completed, and calculates the imaging stability evaluation index FSI in combination with the compensation value Gcoord. The system makes a hierarchical judgment based on the FSI value. When the imaging stability evaluation index FSI ≥ 2.5, it is considered that the imaging is clear and interpretable, and the detection result is automatically output and the image is remotely transmitted to the user terminal through the wireless module; when the imaging stability evaluation index FSI < 2.5, the system determines that there are problems such as image blurring or imaging offset, blocks the output of diagnostic data and restarts the microfluidic chip for re-detection. This intelligent judgment and output mechanism ensures that each detection result has imaging integrity and diagnostic credibility, effectively reduces the risk of false negatives and image misjudgment, and improves the clinical application safety and on-site detection reliability of the system. Brief Description of the Drawings

[0018] Figure 1 It is a schematic diagram of the steps of the non-invasive fully automatic cancer diagnosis system based on microfluidic technology of the present invention; Figure 2 It is a schematic diagram of the microfluidic chip; Figure 3Schematic diagram of the central processing unit of the microfluidic chip. Specific implementation mode

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] Embodiment 1: The present invention provides a non-invasive fully automatic cancer diagnosis system based on microfluidic technology. Please refer to Figure 1 , including a sample loading module, a sample data processing module, a dynamic optothermal response adaptive lysis module, a thermo-optical coordination assisted imaging module, and an imaging stability analysis module; The sample loading module sets a microfluidic chip and sets data acquisition nodes in the microfluidic chip to collect sample data in real time. A central control processor is set in the microfluidic chip to wirelessly transmit the sample data to the central control processor; The sample data processing module preprocesses the sample data in the central control processor to obtain a standardized digital set, constructs a sample database, and stores the standardized digital set; The dynamic optothermal response adaptive lysis module extracts the standardized digital set, calculates and outputs the lysis response efficiency REI, and based on the output result of the lysis response efficiency REI, conducts a preliminary comparison and evaluation to judge the lysis situation; When the thermo-optical coordination assisted imaging module preliminarily compares and evaluates that the lysis is abnormal, it calculates and outputs the optothermal offset compensation value number Gcoord based on the lysis response efficiency REI, and conducts a secondary comparison and evaluation based on the output result of the optothermal offset compensation value number Gcoord to judge the imaging conditions; The imaging stability analysis module calculates and outputs the imaging stability evaluation index FSI based on the optothermal offset compensation value number Gcoord, and conducts a comprehensive evaluation based on the output result of the imaging stability evaluation index FSI to judge the imaging stability.

[0021] In this embodiment, the sample loading module of the system realizes non-invasive loading of non-blood samples such as saliva through a disposable microfluidic chip, and sets four highly sensitive acquisition nodes A, B, C, and D inside the chip to obtain multi-source data such as the flow resistance of the sample, the change in SPR reflectivity, the micro-resistance fluctuation, and the imaging deflection angle in real time. Then, the embedded central control processor transmits it to the external processing unit through wireless communication. The sample data processing module extracts features and performs Z-score normalization processing on the collected multi-dimensional sample data in the central control processor, forms a standardized digital set and stores it in the sample database to construct a dynamic sample portrait; the dynamic photothermal response adaptive lysis module constructs a non-linear coupling mapping model based on the extracted digital set, outputs the lysis response efficiency REI, and realizes the evaluation and judgment of the microcapsule release behavior; if the lysis response efficiency REI result is lower than the stable interval, the photothermal coordination assisted imaging module is automatically triggered, and the photothermal offset compensation value number Gcoord is calculated and output based on the REI value and the imaging deflection angle, and the photothermal path is corrected by means of eccentric adjustment of LED irradiation and heating delay; the imaging stability analysis module further outputs the imaging stability evaluation index FSI based on parameters such as the photothermal offset compensation value number Gcoord value, image intensity, and colorimetric time, judges the visibility and readability of the imaging strip, and decides whether to directly output the detection result or start the image re-acquisition mechanism accordingly. This system does not need to change the hardware structure of the microfluidic chip, and only through the collaborative linkage of acquisition parameter optimization, model guidance, automatic control, and image decision-making, it realizes intelligent adaptation and high-stability detection under different sample states. By introducing three core calculation indicators, namely, the lysis response efficiency REI, the thermal offset compensation index Gcoord, and the imaging stability evaluation index FSI, an algorithm-driven closed-loop control for the whole process from "sample loading, lysis evaluation, image imaging, and imaging interpretation" is realized, greatly improving the system's fault tolerance for sample differences, self-repair ability for abnormal states, as well as the interpretability and credibility of the overall detection results, providing a more efficient, accurate, and portable non-invasive diagnostic solution for early cancer screening.

[0022] Embodiment 2: Please refer to Figure 1 、 Figure 2 and Figure 3 , specifically: The sample loading module includes a sample loading unit and a sample transmission unit; The sample loading unit loads the saliva sample into a disposable microfluidic chip, and sets data acquisition nodes in the microfluidic chip to collect sample data in real time; An electrothermal element and an LED irradiation window are embedded in the microfluidic chip, and microcapsules are set in the detection channel of the microfluidic chip; The data acquisition nodes include point A, point B, point C, and point D; Among them, point A represents the sample pretreatment area, point B represents the SPR photothermal cavity, point C represents the release channel, and point D represents the front end of the test strip reaction; The sample data includes flow resistance, reflectivity before irradiation, reflectivity after irradiation, micro-resistance fluctuation, channel resistance, and the central offset angle Pdef of the imaging area; At point A, a high-frequency oscillation viscosity sensing array is integrated at the entrance of the microfluidic chip. When the saliva sample flows through, the saliva sample is slightly heated by the electrothermal element to a temperature of 2 degrees Celsius. At the same time, a frequency oscillation is applied through the high-frequency oscillation viscosity sensing array to detect the damping change of the oscillation signal by the sample, and the flow resistance is obtained; At point B, a micro-optical detection probe group is set in the area where gold nanoparticles are distributed in the microfluidic chip. After the saliva sample is loaded, the electrothermal element and the LED irradiation window are activated to irradiate the gold nanoparticles, convert the light energy into local heat energy, heat the microcapsules, and release biotin by cracking, thereby exciting the surface plasmon resonance (SPR) effect. Through the micro-optical detection probe group, the position and intensity of the PR reflection peak are detected in real time as they change with the interaction between the gold nanoparticles and the light irradiation, and the reflectivity before irradiation and the reflectivity after irradiation are recorded; At point C, an embedded microelectrode pair is embedded inside the release channel in the microfluidic chip. After the microcapsules are lysed and the released contents, such as biotin and ionic solution, are detected, the instantaneous change in the channel resistance is measured. The micro-resistance fluctuation is measured by the embedded microelectrode pair at a sampling frequency of 10 kHz, which is used to determine whether the microcapsules are lysed sufficiently and whether multiple points are lysed simultaneously, and to screen out false positive backgrounds or non-specific interference of the solution that have not reacted; At point D, an embedded CMOS image sensor is set upstream of the test strip in the microfluidic chip to track in real time the center position of the stripe generated by the development reaction of the contents released by the microcapsules, and to obtain the central offset angle Pdef of the imaging area by comparing the deviation angle between the ideal central axis and the actual imaging position; The sample transmission unit sets a central control processor in the microfluidic chip and transmits the sample data obtained in real time to the central control processor through Bluetooth communication in the microfluidic chip.

[0023] In this embodiment, the system introduces the saliva sample into the disposable microfluidic chip through the sample loading unit, and sets multiple data acquisition nodes inside the chip, including the sample pretreatment area at point A, the SPR photothermal cavity at point B, the release channel at point C, and the test strip reaction front end at point D, to achieve real-time dynamic acquisition of key parameters of the sample in the entire diagnostic process. Among them, point A integrates a high-frequency oscillation viscosity sensor array and an electric heating element to work in conjunction, micro-heating and oscillating the sample when it is loaded, detecting the damping effect of saliva on the frequency-oscillation signal in real time, and extracting the sample flow resistance and viscosity change trend; point B sets a micro-optical detection probe group to capture the dynamic changes of the SPR reflection peak of the gold nanoparticle area before and after LED irradiation, record the reflectivity before and after irradiation, and realize accurate monitoring of the local photothermal response; point C uses an embedded microelectrode pair to continuously track the resistance fluctuation during the lysis of the microcapsule at a sampling frequency of 10kHz, judge whether the lysis behavior occurs and its spatial uniformity, and effectively eliminate the false signal interference caused by non-specific release; point D sets a CMOS image sensor module to capture the offset angle Pdef between the center of the imaging strip of the test strip and the ideal alignment center in real time, providing basic data for subsequent imaging correction and quality assessment. The sample transmission unit completes the real-time reception and caching of all collected data through the built-in central control processor of the microfluidic chip, and stably transmits the multi-dimensional sample data to the external analysis unit or mobile terminal through the Bluetooth communication module. The overall implementation of this module not only realizes the accurate perception of the sample status, lysis behavior and imaging quality in the microchip, but also enables the system to complete real-time adaptation and feedback control of complex sample behavior without adding additional hardware structure.

[0024] Example 3: Please refer to Figure 1 ,Specifically: the sample data processing module includes a pre-processing unit and a data storage unit; The preprocessing unit receives the sample data in real time in the central control processor and preprocesses the sample data to obtain a standardized digital set; Preprocessing includes feature extraction and standardization; The standardized digital set includes the relative viscosity change rate Urel, the local SPR response change rate Espr, the transient cleavage resistance signal Srupt, and the imaging area center offset angle Pdef; Feature extraction is performed by combining and calculating the sample data set in the central control processor to obtain the relative viscosity change rate Urel, the local SPR response change rate Espr and the transient cleavage resistance signal Srupt; The relative viscosity change rate Urel is obtained by extracting the flow resistance before and after micro-heating and calculating the change rate; The local SPR response change rate Espr is obtained by calculating the change rate of the reflectivity before irradiation and the reflectivity after irradiation; The transient lysis resistance signal Srupt is obtained by extracting the micro-resistance fluctuations after instantaneous microcapsule lysis; The normalization process uses the Z-Score normalization method to normalize the relative viscosity change rate Urel, the local SPR response change rate Espr, and the transient lysis resistance signal Srupt obtained by feature extraction, combined with the central offset angle Pdef of the imaging area, to remove the dimensional influence between parameters, convert them into computer numbers, and then summarize the normalized parameters to obtain a normalized digital set; The data storage unit constructs a sample database in the microfluidic chip, connects the sample database to the central control processor, and sets an automatic write port and an automatic read port for the sample database. After preprocessing, the real-time obtained normalized digital set is automatically stored in the sample database.

[0025] In this embodiment, the preprocessing unit of the system is set in the central control processor, which is used to receive the data streams from four acquisition points A, B, C, and D in the microfluidic chip in real time, and perform feature extraction and normalization processing on the data streams, and finally generate a normalized digital set. Specifically, the system calculates the change rate of the flow resistance difference before and after micro-heating to obtain the relative viscosity change rate Urel; extracts the local SPR response change rate Espr by analyzing the change in SPR reflectivity before and after irradiation; and obtains the transient micro-resistance fluctuations caused by microcapsule lysis by high-frequency sampling of the microelectrode signal to obtain the transient lysis resistance signal Srupt. The above three characteristic parameters, combined with the central offset angle Pdef of the imaging area extracted from the image recognition module, jointly constitute a multi-dimensional sample parameter set. To eliminate the dimensional differences between various physical parameters and improve the unity and comparability of data processing, the system uses the Z-Score normalization method to normalize the above characteristic parameters, and outputs a dimensionless normalized digital set, so that the subsequent algorithm model has better calculation stability and generalization ability. The processed normalized digital set is stored in real time in the sample database constructed in the microfluidic chip through the automatic write port, and supports data exchange with other modules through the automatic read port, forming a closed-loop information management mechanism.

[0026] Example 4: Please refer to Figure 1 , specifically: The dynamic optothermal response adaptive lysis module includes a lysis analysis unit and a lysis evaluation unit; The lysis analysis unit constructs a non-linear coupling mapping algorithm model. The non-linear coupling mapping algorithm model performs comprehensive modeling by inputting energy, sample characteristics, release difficulty, and heating time. Then, through the automatic read port, it extracts the real-time obtained normalized digital set, inputs it into the non-linear coupling mapping algorithm model, and calculates and outputs the lysis response efficiency REI to analyze the effectiveness of microcapsule lysis; The cracking response efficiency REI is calculated and output through the following non - linear coupling mapping algorithm model; ; In the formula, ln represents the natural logarithm function, k1 represents the photothermal transfer rate constant, a dimensionless value calibrated for the device, t led represents the excitation duration of the LED light source, which is precisely timed and counted by the LED driver of the microfluidic chip, and e represents the exponential function; Among them, represents the energy input factor; represents the cracking resistance factor; represents the time - response excitation factor.

[0027] The cracking evaluation unit conducts a preliminary comparative evaluation based on the output result of the cracking response efficiency REI, judges the cracking situation of the microcapsules in the microfluidic chip, and triggers the thermo - optical coordination - assisted imaging module based on the evaluation result. The specific evaluation content is as follows; When the cracking response efficiency REI ≥ 5.0, it indicates normal cracking. At this time, direct imaging is performed and the test strip enters the color - development stage; When 3.0 ≤ cracking response efficiency REI < 5.0, it indicates abnormal cracking. At this time, the thermo - optical coordination - assisted imaging module is triggered; When the cracking response efficiency REI < 3.0, it indicates cracking failure. At this time, the detection is stopped and a prompt for re - sampling is given.

[0028] In this embodiment, the system constructs a non - linear coupling mapping algorithm model through the cracking analysis unit based on the standardized digital set extracted by the central control processor, comprehensively considering the input energy, represented by the SPR response change rate Espr, the sample characteristics, such as the relative viscosity change rate Urel, the cracking release difficulty, such as the transient resistance signal Srupt, and the LED irradiation duration t ledKey parameters such as these. Through this non-linear coupling mapping algorithm model, the system can quantitatively analyze the lysis efficiency at the early stage of chip operation and determine whether the microcapsules have reached the critical condition of sufficient release. The lysis evaluation unit then conducts a preliminary comparative evaluation based on the numerical results of the lysis response efficiency REI and sets a judgment threshold: when the lysis response efficiency REI ≥ 5.0, it indicates sufficient lysis and can directly enter the test strip development imaging process; when the lysis response efficiency REI is between 3.0 - 5.0, it indicates abnormal lysis, and the system automatically triggers the thermoluminescence coordinated auxiliary imaging module to compensate and optimize the subsequent imaging; when the lysis response efficiency REI < 3.0, it is determined that lysis has failed, and the system will abort the detection process and prompt for re-sampling or parameter reset. Through the implementation of this module, the present invention for the first time constructs an adaptive decision-making mechanism with lysis efficiency as the core index, overcoming the problems of fixed optothermal parameters, poor sample adaptability, and inability to warn of lysis failure in traditional microfluidic detection. This mechanism enables the system to achieve intelligent judgment and response control of the lysis state in the early stage, effectively improving the process transparency, judgment accuracy, and abnormal fault tolerance of the detection, significantly enhancing the stability, reliability, and adaptability of the overall system, and is especially suitable for the early cancer screening scenario with a large variety of samples and strong non-invasive detection requirements.

[0029] Example 5: Please refer to Figure 1 , specifically: The thermoluminescence coordinated auxiliary imaging module includes a thermoluminescence offset compensation analysis unit and an execution unit; After abnormal lysis appears in the preliminary comparative evaluation, the thermoluminescence offset compensation analysis unit, based on the currently obtained lysis response efficiency REI at this time, combines the central offset angle Pdef of the imaging area to calculate and output the thermoluminescence offset compensation value Gcoord to adaptively compensate for signal imaging deviation and intensity loss; The thermoluminescence offset compensation value Gcoord is calculated and output through the following algorithm formula; ; In the formula, represents the micro-offset stability constant, with a dimensionless value, calibration value, t corr represents the automatic compensation time, that is, the system execution feedback control time, and T0 represents the standard LED irradiation time.

[0030] The execution unit conducts a secondary comparative evaluation based on the output result of the thermoluminescence offset compensation value Gcoord and executes an adjustment strategy based on the result of the secondary comparative evaluation. The specific evaluation content is as follows; When the thermoluminescence offset compensation value Gcoord ≥ 4.0, the execution adjustment strategy is triggered at this time; When the thermoluminescence offset compensation value Gcoord < 4.0, no adjustment is required at this time, and direct development is performed; The adjustment strategy sends control instructions to the heating element and the LED irradiation window through the central control processor, automatically extending the heating area time of the heating element by +5 s, and simultaneously fine-tuning the eccentric angle of the irradiation window by 5°. After the adjustment, the iterative execution system is carried out until the development is completed iteratively.

[0031] In this embodiment, after the thermal-optical offset compensation analysis unit of the system detects that the cracking response efficiency REI is in the abnormal range in the preliminary evaluation, it calculates and outputs the thermal-optical offset compensation value Gcoord in combination with the central offset angle Pdef of the imaging area. The execution unit conducts a secondary comparative evaluation based on the output result of the thermal-optical offset compensation value Gcoord. When the thermal-optical offset compensation value Gcoord ≥ 4.0, the system determines that the current imaging conditions are not sufficient to directly complete high-quality development. At this time, the central control processor issues instructions to the heating element and the LED irradiation window, automatically extending the heating area time by +5 seconds and fine-tuning the eccentric angle of the irradiation window by 5° to change the local distribution of thermal-optical energy and optimize the coupling relationship between the microcapsule release area and the test strip development band; if the thermal-optical offset compensation value Gcoord < 4.0, it is considered that the offset degree is within the acceptable range, and the system directly enters the imaging stage. All adjustment processes are automatically iteratively executed under the system control until the development is completed and the imaging parameters are stable. Through the implementation of this module, the present invention establishes a thermal-optical compensation algorithm model with REI and the image center offset angle Pdef as the core input variables, which can dynamically adjust the thermal-optical excitation path and heating strategy to achieve an intelligent transition control from insufficient cracking to readable imaging. This mechanism effectively solves the problems of imaging offset, strip skew, or inconsistent signal intensity caused by uneven energy input or sample rheology differences, improves the accuracy, alignment accuracy, and color development integrity of the test strip imaging, enhances the self-repair ability and imaging robustness of the system under abnormal conditions, and ensures that the image quality of the non-invasive cancer detection result is controllable and the data output is reliable.

[0032] Embodiment 6: Please refer to Figure 1 , specifically: The imaging stability analysis module includes an imaging data extraction unit, an imaging analysis unit, and an imaging stability evaluation unit; The imaging data extraction unit analyzes the image intensity and fits the colorimetric rate in the test strip area after development; The image intensity analysis of the test strip area is performed by using an embedded CMOS image sensor to perform gray-scale intensity integration processing on the developed strip image to obtain the average pixel value Iavg of the central area and the background area value Ibg; The colorimetric rate fitting monitors the rising process of the image from no intensity to the maximum intensity, and records the reaction development time tpeak, that is, the time point when the image signal first exceeds 95% of the maximum value; Data normalization is performed on the average pixel value Iavg, the background region value Ibg, and the reaction development time tpeak to eliminate the parameter dimension.

[0033] Based on the photothermal offset compensation value number Gcoord output after development, the imaging analysis unit combines the obtained average pixel value Iavg, background region value Ibg, and reaction development time tpeak to calculate and output the imaging stability evaluation index FSI, and analyzes the readability and model uniformity of the image strip; The imaging stability evaluation index FSI is calculated and output through the following algorithm formula; ; In the formula, ln represents the natural logarithm function, k0 represents the colorimetric response constant, which is determined by the reagent concentration and the imaging amplification system, a represents the colorimetric time index, represents the fluid correction coefficient, and the above values are all dimensionless parameters; represents the streptavidin binding strength, which is used to quantitatively reflect the reaction activity strength of the biotin released from the microcapsule binding to the streptavidin immobilized on the test strip, and is one of the core reference quantities for imaging intensity and diagnostic reliability.

[0034] Based on the output result of the imaging stability evaluation index FSI, the imaging stability evaluation unit conducts a comprehensive evaluation to judge the imaging stability. The specific evaluation content is as follows; When the imaging stability evaluation index FSI ≥ 2.5, it indicates that the imaging is stable, the image is clear, the alignment is accurate, the strip has high contrast, and the result is automatically output, and the current imaging content is remotely transmitted to the diagnostic user terminal; When the imaging stability evaluation index FSI < 2.5, it indicates that the imaging is abnormal, there are image blurring, distortion, unreadable, and mild deviation from the steady state. At this time, the transmission is blocked, and the microfluidic chip is restarted for re-imaging.

[0035] In this embodiment, after the imaging data extraction unit of the system completes the development of the test strip, it uses an embedded CMOS image sensor to collect images of the developed area, and adopts the gray intensity integration method to extract the average pixel value Iavg of the strip area and the gray value Ibg of the background area, and further calculates the signal intensity contrast. At the same time, through the colorimetric rate fitting method, the development process of the image intensity from the initial state to reaching 95% of the maximum value is monitored in real time, and the reaction development time tpeak is extracted. These three constitute the core characteristic parameters of the image response. Then, the imaging analysis unit based on the above extracted parameters, combined with the photothermal offset compensation value Gcoord output by the photothermal coordination assisted imaging module in the previous stage, comprehensively calculates the imaging stability evaluation index FSI. Finally, the imaging stability evaluation unit sets a grading determination mechanism according to the output result of FSI: when the imaging stability evaluation index FSI ≥ 2.5, it is judged that the image quality is stable, clear, and the alignment is accurate. The system automatically outputs the detection result and remotely transmits the test strip image to the diagnostic user terminal through the wireless module; when the imaging stability evaluation index FSI < 2.5, it is determined that the imaging is abnormal, and there may be phenomena such as image blurring, distortion, or mild offset. The system aborts the output and automatically restarts the microfluidic chip for secondary imaging. Through the implementation of this module, the present invention realizes an intelligent judgment mechanism for image quality based on development data, and establishes a complete set of stability guarantee links from the physical imaging signal to the final diagnostic output. Compared with the method of manual judgment or simple threshold judgment of image results in traditional detection methods, this module realizes self-evaluation, self-screening, and self-correction of imaging data through multi-dimensional parameter fusion and non-linear calculation models, significantly improving the system's control ability for outputting interpretable images, the ability to identify imaging abnormalities, and the ability to guarantee the credibility of terminal results, thereby improving the overall automation, reliability, and remote adaptability of the system in the early cancer screening scenario.

[0036] Although embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it is understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A non-invasive fully automatic cancer diagnosis system based on microfluidic technology, characterized in that: It includes a sample loading module, a sample data processing module, a dynamic photothermal response adaptive lysis module, a thermo-optical coordinated auxiliary imaging module, and an imaging stability analysis module; The sample loading module sets a microfluidic chip and sets data acquisition nodes in the microfluidic chip to collect sample data in real time. A central control processor is set in the microfluidic chip to wirelessly transmit the sample data to the central control processor; The sample data processing module preprocesses the sample data in the central control processor to obtain a standardized digital set, constructs a sample database, and stores the standardized digital set; The dynamic photothermal response adaptive lysis module extracts the standardized digital set, calculates and outputs the lysis response efficiency REI, and based on the output result of the lysis response efficiency REI, conducts a preliminary comparison and evaluation to judge the lysis situation; When the thermo-optical coordinated auxiliary imaging module preliminarily compares and evaluates that the lysis is abnormal, it calculates and outputs the photothermal offset compensation value number Gcoord based on the lysis response efficiency REI, and conducts a secondary comparison and evaluation based on the output result of the photothermal offset compensation value number Gcoord to judge the imaging conditions; The imaging stability analysis module calculates and outputs the imaging stability evaluation index FSI based on the photothermal offset compensation value number Gcoord, and conducts a comprehensive evaluation based on the output result of the imaging stability evaluation index FSI to judge the imaging stability.

2. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 1, wherein: The sample loading module includes a sample loading unit and a sample transmission unit; The sample loading unit loads a saliva sample into a disposable microfluidic chip, sets data acquisition nodes in the microfluidic chip, and collects sample data in real time; An electrothermal element and an LED irradiation window are embedded in the microfluidic chip, and microcapsules are set in the detection channel of the microfluidic chip; The data acquisition nodes include point A, point B, point C, and point D; Among them, point A represents the sample pretreatment area, point B represents the SPR photothermal cavity, point C represents the release channel, and point D represents the front end of the test strip reaction; The sample data includes flow resistance, pre-irradiation reflectivity, post-irradiation reflectivity, micro-resistance fluctuation, channel resistance, and the imaging area center offset angle Pdef; At point A, a high-frequency oscillation viscosity sensing array is integrated at the entrance of the microfluidic chip. When the saliva sample flows through, the saliva sample is slightly heated by the electrothermal element to a temperature of 2 degrees Celsius. At the same time, a frequency oscillation is applied through the high-frequency oscillation viscosity sensing array to detect the damping change of the sample to the oscillation signal and obtain the flow resistance; At point B, a micro-optical detection probe group is set in the gold nanoparticle distribution area in the microfluidic chip. After the saliva sample is loaded, the electrothermal element and the LED irradiation window are started to irradiate the gold nanoparticles, convert the light energy into local heat energy, heat the microcapsules, release biotin by lysis, and stimulate the surface plasmon resonance (SPR) effect. The position and intensity of the PR reflection peak are detected in real time with the change of the interaction between the gold nanoparticles and the light by the micro-optical detection probe group, and the pre-irradiation reflectivity and the post-irradiation reflectivity are recorded; At point C, by embedding an embedded microelectrode pair inside the release channel in the microfluidic chip, after the microcapsules are lysed, the released contents are measured for micro-resistance fluctuations by the embedded microelectrode pair at a sampling frequency of 10 kHz; At point D, by setting an embedded CMOS image sensor upstream of the test strip in the microfluidic chip, the center position of the stripe generated by the imaging reaction of the released contents of the microcapsules is tracked in real time, and the deviation angle between the ideal central axis and the actual imaging position is compared to obtain the central offset angle Pdef of the imaging area; The sample transmission unit sets a central control processor in the microfluidic chip and transmits the real-time acquired sample data to the central control processor through Bluetooth communication in the microfluidic chip.

3. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 2, characterized in that: The sample data processing module includes a preprocessing unit and a data storage unit; The preprocessing unit receives the sample data in real time in the central control processor and preprocesses the sample data to obtain a standardized digital set; The preprocessing includes feature extraction and standardization processing; The standardized digital set includes the relative viscosity change rate Urel, the local SPR response change rate Espr, the transient lysis resistance signal Srupt, and the central offset angle Pdef of the imaging area; The feature extraction obtains the relative viscosity change rate Urel, the local SPR response change rate Espr, and the transient lysis resistance signal Srupt by performing combined calculations on the sample data set in the central control processor; The standardization processing uses the Z-Score standardization method to perform standardization processing on the relative viscosity change rate Urel, the local SPR response change rate Espr, and the transient lysis resistance signal Srupt obtained by feature extraction, combined with the central offset angle Pdef of the imaging area, to remove the dimensional influence between the parameters, convert them into computer numbers, and then summarize the standardized parameters to obtain a standardized digital set; The data storage unit constructs a sample database in the microfluidic chip, connects the sample database to the central control processor, and sets an automatic write port and an automatic read port for the sample database. After preprocessing, the real-time acquired standardized digital set is automatically stored in the sample database.

4. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 3, wherein: The dynamic photothermal response adaptive lysis module includes a lysis analysis unit and a lysis evaluation unit; The lysis analysis unit constructs a non-linear coupling mapping algorithm model. The non-linear coupling mapping algorithm model is comprehensively modeled by inputting energy, sample characteristics, release difficulty, and heating time. Then, through the automatic read port, the real-time acquired standardized digital set is extracted and input into the non-linear coupling mapping algorithm model for calculation to output the lysis response efficiency REI and analyze the effectiveness of microcapsule lysis.

5. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 4, wherein: The lysis evaluation unit performs a preliminary comparative evaluation based on the output result of the lysis response efficiency REI to judge the lysis situation of the microcapsules in the microfluidic chip, and triggers the thermo-optical coordinated auxiliary imaging module based on the evaluation result. The specific evaluation content is as follows; When the lysis response efficiency REI≥5.0, it indicates normal lysis. At this time, direct imaging is performed and the test strip enters the color development stage; When 3.0 ≤ cracking response efficiency REI < 5.0, it indicates abnormal cracking. At this time, the thermoluminescence coordinated auxiliary imaging module is triggered; When the cracking response efficiency REI < 3.0, it indicates cracking failure. At this time, the detection is stopped and a prompt to resample is given.

6. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 4, characterized in that: The thermoluminescence coordinated auxiliary imaging module includes a thermoluminescence offset compensation analysis unit and an execution unit; After the thermoluminescence offset compensation analysis unit detects abnormal cracking during the preliminary comparison and evaluation, based on the currently obtained cracking response efficiency REI and combined with the imaging area center offset angle Pdef, it calculates and outputs the optothermal offset compensation value Gcoord to adaptively compensate for signal imaging deviation and intensity loss.

7. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 6, characterized in that: The execution unit conducts a secondary comparison and evaluation based on the output result of the optothermal offset compensation value Gcoord, and executes an adjustment strategy based on the result of the secondary comparison and evaluation. The specific evaluation content is as follows; When the optothermal offset compensation value Gcoord ≥ 4.0, the execution of the adjustment strategy is triggered at this time; When the optothermal offset compensation value Gcoord < 4.0, no adjustment is required at this time, and direct development is performed; The adjustment strategy sends a control command to the electrothermal element and the LED irradiation window through the central control processor, automatically extends the heating area time of the electrothermal element by +5s, and simultaneously finely adjusts the eccentric angle of the irradiation window by 5°. After the adjustment, the iterative execution system is carried out until the development ends the iteration.

8. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 7, characterized in that: The imaging stability analysis module includes an imaging data extraction unit, an imaging analysis unit, and an imaging stability evaluation unit; The imaging data extraction unit analyzes the image intensity and colorimetric rate fitting of the test strip area after development; The image intensity analysis of the test strip area processes the gray intensity integration of the developed strip image using an embedded CMOS image sensor to obtain the average pixel value Iavg of the central area and the background area value Ibg; The colorimetric rate fitting records the reaction development time tpeak; And the average pixel value Iavg, the background area value Ibg, and the reaction development time tpeak are subjected to data normalization processing to eliminate the parameter dimension.

9. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 8, characterized in that: The imaging analysis unit calculates and outputs the imaging stability evaluation index FSI based on the optothermal offset compensation value Gcoord output after development, combined with the obtained average pixel value Iavg, background area value Ibg, and reaction development time tpeak, and analyzes the readability and model uniformity of the image strip.

10. The non-invasive fully automatic cancer diagnosis system based on microfluidic technology according to claim 9, characterized in that: The imaging stability evaluation unit conducts a comprehensive evaluation based on the output result of the imaging stability evaluation index FSI to judge the stability of the imaging. The specific evaluation content is as follows; When the imaging stability evaluation index FSI ≥ 2.5, it indicates stable imaging, and the automatic result is output, and the current imaging content is remotely transmitted to the diagnostic user terminal; When the imaging stability evaluation index FSI < 2.5, it indicates abnormal imaging. At this time, the transmission is blocked and the microfluidic chip is restarted for re-imaging.

Images