A detection device and method for gastrointestinal tumor screening examination

Abstract

This invention discloses a detection device and method for screening gastrointestinal tumors, comprising: a biological sample collection module, a multidimensional signal sensing module, a feature extraction and processing module, a data transmission and interaction module, a model calculation and analysis module, and a result output and feedback module. Each module establishes a collaborative interaction mechanism with a specific data link through an embedded interface, achieving seamless integration from sample collection to result output. The method relies on the functions of each module of the device to construct a systematic screening system through multidimensional signal capture and transformation, feature factor separation and screening, encrypted data transmission, multi-parameter coupled calculation, and result decoding and presentation. This device and method effectively reduce feature information loss and data transmission risks, ensuring the reliability and consistency of screening results, and providing efficient and accurate technical support for early screening of gastrointestinal tumors.

Description

Technical Field

[0001] This invention relates to the field of gastrointestinal tumor screening technology, and in particular to a detection device and method for gastrointestinal tumor screening. Background Technology

[0002] Gastrointestinal tumors are highly prevalent malignant diseases, and early screening plays a crucial role in improving cure rates and reducing mortality. Current screening methods largely rely on single detection modalities, requiring a collaborative approach involving multiple stages such as biological sample analysis, signal processing, and data computation. Current clinical screening methods involve core processes including sample collection, signal conversion, feature extraction, data transmission, and result analysis; however, the lack of systematic integration in each stage leads to fragmented testing processes, hindering efficient integration from sample collection to result output. As medical testing technologies advance towards precision and integration, there is an urgent need to construct a multi-module collaborative testing system. Through multi-dimensional signal capture and multi-parameter computational analysis, this system can compensate for the shortcomings of traditional screening methods in terms of detection coverage and data processing efficiency, meeting the clinical demands for precise and efficient early screening of gastrointestinal tumors.

[0003] Existing technologies have two significant drawbacks: First, there is a lack of a unified adaptation mechanism for data interaction between various detection modules. The technical connections between sample collection, signal perception, and feature processing are insufficient, leading to loss of feature information during signal conversion and loss of integrity during data transmission, making it impossible to form a coherent and efficient detection chain. Second, the parameter calculation and analysis models lack specificity, do not fully consider the multi-dimensional coupling relationship of gastrointestinal tumor-related biomarkers, and lack the ability to dynamically adjust feature factor screening and weight allocation. This makes it difficult to accurately match the individual differences of different examinees, resulting in limited accuracy of parameter processing during the detection process and affecting the reliability and consistency of screening results. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of the existing technology, the present invention provides a detection device and detection method for screening gastrointestinal tumors.

[0005] The technical solution adopted in this invention is a detection device for screening gastrointestinal tumors, comprising: a biological sample collection module, a multidimensional signal sensing module, a feature extraction and processing module, a data transmission and interaction module, a model calculation and analysis module, and a result output and feedback module; The biological sample acquisition module establishes a bidirectional data link with the multidimensional signal sensing module through a sealed channel. The multidimensional signal sensing module converts the captured multidimensional biological signals into computable electrical signals and transmits them to the feature extraction and processing module. The feature extraction and processing module separates and filters the electrical signals into multidimensional feature factors and then sends them to the data transmission and interaction module. The data transmission and interaction module uses an encrypted transmission protocol to transmit the filtered feature data to the model operation and analysis module. The model operation and analysis module performs multi-parameter coupling operations on the feature data, generates screening analysis results, and sends them to the result output feedback module. The result output feedback module decodes the analysis results, converts the format, and then visualizes them. All modules interact and work together through embedded interfaces to form a complete gastrointestinal tumor screening and detection link.

[0006] Furthermore, the feature extraction and processing module includes: a signal denoising and filtering unit, a feature factor separation unit, a feature priority ranking unit, and an effective feature screening unit. The signal denoising and filtering unit uses multi-band signal separation technology to remove interference signals from the electrical signals transmitted by the multi-dimensional signal sensing module. The feature factor separation unit uses a multi-dimensional feature decomposition algorithm to split the denoised electrical signals into biological feature-related factors. The feature priority ranking unit allocates and ranks feature factor weights based on the correlation between features and gastrointestinal tumors. The effective feature screening unit extracts and retains high-weight feature factors based on the ranking results.

[0007] Furthermore, the data transmission interaction module includes: a transmission protocol adaptation unit, a data encryption processing unit, a transmission status monitoring unit, and a data integrity verification unit. The transmission protocol adaptation unit dynamically matches the transmission protocol according to the interface protocol type of the model operation and analysis module. The data encryption processing unit uses an asymmetric encryption algorithm to encrypt the feature data output by the feature extraction processing module. The transmission status monitoring unit captures the link status parameters during the data transmission process in real time and makes dynamic adjustments. The data integrity verification unit verifies the consistency of the data before and after transmission using checksum comparison technology.

[0008] Furthermore, the model operation and analysis module includes: a parameter initialization and configuration unit, a multi-model collaborative operation unit, an operation result fusion unit, and an analysis result optimization unit. The parameter initialization and configuration unit sets the initial parameters of the operation model according to the range of gastrointestinal tumor screening parameters. The multi-model collaborative operation unit calls a preset operation model to perform parallel operation processing on the feature data. The operation result fusion unit uses weighted fusion technology to integrate the operation results of multiple models. The analysis result optimization unit corrects and adjusts the integrated results according to the fluctuation trend of the feature data.

[0009] Furthermore, the feature factor operation expression used by the model operation and analysis module is as follows: in, This is the comprehensive value calculated from the feature factors. , , These are the feature weight coefficients. For adjustment coefficients, These are the protein characteristic values ​​of biological samples. These are nucleic acid sequence characteristic parameters. This is a value associated with metabolite concentration. These are cell morphology characteristic parameters. As a tumor marker-associated factor, The baseline value for normal tissue characteristics, This represents the rate of change of signal strength.

[0010] Furthermore, the signal transformation expression of the multidimensional signal sensing module is as follows: ,in, The converted electrical signal value, The conversion factor is... For the first Bio-like signal intensity, For the first Signal-to-weight conversion, For the first Physiological parameter test values, For the first Class parameter baseline value, The change in signal frequency. For the stable value of the signal period, This represents the number of signal and parameter categories.

[0011] Furthermore, the signal decoding expression of the result output feedback module is: ,in, The decoded and visualized signal values. These are the decoding coefficients. The result is the original signal value. This is the decoding gain coefficient. This is the signal format conversion factor. Signal strength before encoding This is the encoding loss value. Parameters for decoding error correction. For signal synchronization factor, As the decoding base, This is the signal compensation value.

[0012] Furthermore, the sample processing expression of the biological sample acquisition module is as follows: ,in, The effective signal value after sample processing. For processing coefficients, For parameters related to sample collection volume, This is the sample activity retention coefficient. For sample processing time parameters, For sample storage environment parameters, This is a parameter representing the degree of sample homogenization. For the sample temperature change parameter, This is a parameter for adjusting sample concentration. This represents the total number of sample processing parameters. For the k-th sample processing parameter.

[0013] Furthermore, the optimized expression for the transmission efficiency of the data transmission interaction module is as follows: ,in, For optimized transmission efficiency, To optimize the coefficients, For transmission bandwidth parameters, For signal transmission rate, For transmission delay parameters, For data cache capacity, For transmission link stability parameters, For interference signal strength parameters, For anti-interference capability parameters, Optimize the number of parameter categories for transmission.

[0014] A detection method for screening gastrointestinal tumors, applied to a detection device for screening gastrointestinal tumors, includes the following steps: S1, collecting and sealing relevant gastrointestinal biological samples from the subject using a biological sample collection module; S2, capturing and converting multidimensional signals from the collected biological samples using a multidimensional signal sensing module; S3, performing noise reduction and feature factor separation and screening on the converted electrical signals using a feature extraction and processing module; S4, transmitting the screened feature data to a model calculation and analysis module using an encrypted protocol via a data transmission interaction module; S5, processing the feature data using a multi-parameter coupled calculation algorithm in the model calculation and analysis module. The system performs computational processing and generates screening analysis results. In step S6, the results are decoded, format converted, and visualized through the result output feedback module. Each step achieves efficient transmission and accurate computation of feature data through the collaborative operation between modules, forming a complete detection process from sample collection to result output. The computation process employs a dynamic parameter adjustment mechanism to optimize the computational model in real time, ensuring the relevance of feature data processing and the accuracy of analysis results. The transmission process uses link status monitoring technology to dynamically adjust the data transmission link, ensuring the integrity and security of data transmission. Each step is executed sequentially according to the time sequence, and the parameters of each link are adapted and collaboratively optimized through the feedback mechanism.

[0015] Beneficial Effects: This invention proposes a detection device and method for screening gastrointestinal tumors. Six modules, including a biological sample collection module and a multidimensional signal sensing module, establish an efficient data interaction link through a sealed channel and encrypted transmission protocol, forming a unified and compatible technical connection mechanism. This completely solves the problems of fragmentation and lack of data transmission integrity in traditional detection processes, achieving seamless operation from sample collection to result output. It significantly reduces the loss of feature information during signal conversion and data transmission. Utilizing multi-dimensional feature screening in the feature extraction and processing module and multi-parameter coupled computation in the model operation and analysis module, it comprehensively considers the multi-dimensional coupling relationships of gastrointestinal tumor-related biological characteristics. Dynamic weight allocation and multi-model collaborative computation enhance the targeting of feature processing. Simultaneously, combined with the status monitoring of the data transmission interaction module and the visualization of the result output feedback module, it achieves precise matching for individual differences among different examinees, effectively overcoming the bottleneck of limited accuracy in traditional model parameter processing. This significantly improves the reliability and consistency of screening results, ultimately achieving the goal of precise and efficient early screening of gastrointestinal tumors, meeting the core clinical needs for tumor screening. Attached Figure Description

[0016] Figure 1 This is a diagram showing the modular composition of the device of the present invention; Figure 2 This is a flowchart of the method steps of the present invention. Detailed Implementation

[0017] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1 As shown, a detection device for screening gastrointestinal tumors includes: a biological sample collection module, a multidimensional signal sensing module, a feature extraction and processing module, a data transmission and interaction module, a model calculation and analysis module, and a result output and feedback module. The biological sample acquisition module establishes a bidirectional data link with the multidimensional signal sensing module through a sealed channel. The multidimensional signal sensing module converts the captured multidimensional biological signals into computable electrical signals and transmits them to the feature extraction and processing module. The feature extraction and processing module separates and filters the electrical signals into multidimensional feature factors and then sends them to the data transmission and interaction module. The data transmission and interaction module uses an encrypted transmission protocol to transmit the filtered feature data to the model operation and analysis module. The model operation and analysis module performs multi-parameter coupling operations on the feature data, generates screening analysis results, and sends them to the result output feedback module. The result output feedback module decodes the analysis results, converts the format, and then visualizes them. All modules interact and work together through embedded interfaces to form a complete gastrointestinal tumor screening and detection link.

[0019] The biosample collection module features a sterile, sealed design, measuring 18 cm long, 8 cm wide, and 5 cm high, with a weight under 350 grams. It is suitable for collecting core biosamples for gastrointestinal tumor screening, including oral mucosa, feces, and blood. The module incorporates a pressure sensor and a temperature control component. The pressure sensor, with a sensitivity of 0.01 kPa, monitors pressure changes in real time during collection to prevent tissue damage. The temperature control component maintains a stable storage environment of 4 degrees Celsius, ensuring stable sample viability for 24 hours after collection. During implementation, a disposable sterile collection head comes into contact with the target area of ​​the subject. The surface of the collection head is equipped with 12 micron-sized collection holes, each with a diameter of 80 microns, which can accurately capture sample tissue. The collection volume is controlled within the range of 50 to 200 milligrams. After collection, the collection channel is closed by a built-in sealing valve to prevent sample contamination and volatilization. At the same time, the module uploads parameters such as collection time, collection volume, and storage temperature to the system in real time through a data interface, providing basic data support for subsequent testing. Its sealing design and environmental control technology can effectively prevent sample degradation and ensure the accuracy of subsequent test data.

[0020] The multidimensional signal sensing module is equipped with three core sensing components: an optical signal sensor, an electrochemical sensor, and a bioimpedance sensor. The module has a response time of ≤0.3 seconds and a signal detection accuracy of 0.001 units, simultaneously capturing three key signals from biological samples: optical characteristics, electrochemical properties, and impedance changes. The optical signal sensor uses near-infrared spectroscopy with a detection wavelength range of 700 to 1500 nanometers to capture the spectral absorption signals of biomolecules in the sample. The electrochemical sensor employs a three-electrode system with a 2-millimeter spacing between the working electrode, reference electrode, and counter electrode, acquiring electrochemical signals by detecting the redox reaction of the sample. The bioimpedance sensor has a detection frequency range of 100 Hz to 1 MHz, measuring the impedance value and phase angle changes of the sample. During implementation, the module interfaces with the biological sample acquisition module via an adapter. The sample transmitted by the acquisition module undergoes signal contact through the module's built-in sample reaction cell, which has a volume of 50 microliters. The contact time between the sample and the sensing components is controlled within 10 seconds. Three types of sensors simultaneously acquire signals and convert them into computable electrical signals. During the conversion process, the signal amplification factor is set to 1000 times to ensure that weak signals can be effectively identified. The module uses a filtering circuit to initially filter environmental interference signals and transmits the converted electrical signals to the feature extraction and processing module. Its multi-dimensional signal capture capability can comprehensively acquire the biological feature information of the sample, providing a rich data foundation for subsequent feature extraction.

[0021] The feature extraction and processing module incorporates a 32-bit microprocessor with a processing frequency of 1.2 GHz and a 256-megabyte cache, enabling rapid processing of electrical signals transmitted by the multi-dimensional signal sensing module. The module first performs noise reduction on the electrical signals using multi-band signal separation technology, setting eight signal bands, each with a bandwidth of 100 Hz, and removing 50 Hz power frequency interference and high-frequency noise signals. Then, a multi-dimensional feature decomposition algorithm is used to decompose the noise-reduced electrical signals, extracting 12 core feature factors, including spectral feature factors, electrochemical feature factors, and impedance feature factors. Next, weights are assigned based on the correlation between the features and gastrointestinal tumors, with weight values ​​ranging from 0.1 to 0.9. Tumor marker-related feature factors are assigned a high weight range of 0.8 to 0.9. Finally, a threshold screening mechanism is used, setting a feature factor threshold of 0.3, retaining valid feature factors with weights higher than the threshold. The filtered feature data is stored in the cache and awaits transmission. This module uses precise feature extraction and filtering to eliminate invalid interference information, focusing on core biometric features and providing high-quality data input for subsequent model operation and analysis, thereby improving the relevance and reliability of the detection results.

[0022] The data transmission module supports both wired and wireless transmission modes. Wired transmission uses a USB 3.0 interface with a transmission rate of up to 5 gigabits per second. Wireless transmission supports Wi-Fi 6 and Bluetooth 5.2 protocols, with a wireless transmission distance of ≤10 meters and a transmission latency of ≤50 milliseconds. The module incorporates an encryption chip, employing an asymmetric encryption algorithm to encrypt the feature data output by the feature extraction processing module. The key length is 2048 bits, ensuring security during data transmission. It also features a transmission status monitoring unit that monitors parameters such as signal strength and transmission rate in real time. When the signal strength drops below -70 dB / mW, the module automatically switches transmission modes; when the transmission rate drops below 10 megabits per second, a bandwidth optimization mechanism is activated. After data transmission is complete, the module uses a checksum comparison technology to verify the consistency between the received and transmitted data. The checksum length is 128 bits, ensuring data integrity. During implementation, the module first automatically adapts the transmission protocol according to the interface type of the model operation and analysis module. After completing the protocol handshake, it starts the data encryption and transmission process. During the transmission process, the transmission progress and status are fed back in real time. If data loss or transmission interruption occurs, the retransmission mechanism is automatically triggered. The maximum number of retransmissions is 3, ensuring that the feature data is transmitted to the model operation and analysis module in a complete, secure and efficient manner, and ensuring the continuity of the detection link.

[0023] The model operation and analysis module is equipped with an embedded neural network operation unit, achieving an operation accuracy of 6 decimal places and supporting parallel processing, capable of processing 8 channels of feature data simultaneously. The module first sets initial parameters based on the parameter range for gastrointestinal tumor screening, defining 16 operation parameters, including key parameters such as sample feature thresholds, operation step size, and fusion coefficients. These initial parameters are calibrated based on a large amount of clinical sample data. Then, a preset multi-model collaborative operation mechanism is invoked, integrating three types of operation models: feature matching model, parameter coupling model, and result prediction model. The transmitted feature data is processed in parallel, with each model's operation time ≤2 seconds. After the operation is completed, a weighted fusion technique is used to integrate the results of the three models. The fusion weight is dynamically adjusted according to the feature data type: the fusion weight for spectral feature-related results is 0.35, the fusion weight for electrochemical feature-related results is 0.35, and the fusion weight for impedance feature-related results is 0.3. Finally, a result optimization algorithm is used to correct the integrated results, adjusting the optimization coefficient based on the fluctuation range of the feature data. The optimization coefficient is set to 0.15 when the fluctuation range is greater than 0.2, and to 0.05 when the fluctuation range is less than 0.2, ultimately generating the screening analysis results. This module leverages multi-model collaboration and dynamic optimization to fully mine tumor-related information from feature data, thereby improving the accuracy and stability of screening and analysis results.

[0024] The results output feedback module is equipped with a 2.8-inch high-definition display screen with a resolution of 480×640 pixels. It supports touch operation and dual text and graphic modes, and also has a built-in audio output unit and data interface, enabling visualization, voice broadcasting, and data export of results. After receiving the screening analysis results transmitted from the model calculation and analysis module, the module decodes the result data using a signal decoding algorithm with a decoding time of ≤0.5 seconds, converting the encrypted and stored result data into a recognizable signal format. Subsequently, it performs format conversion, transforming the original calculation results into a clinically commonly used screening report format, including core content such as feature parameter summaries, risk level assessment, and key indicator annotations. Finally, it presents the results visually on the display screen, using red, yellow, and green to indicate different risk levels. The audio output unit can loop and broadcast key result information at a speech rate of 180 words per minute, and the data interface supports exporting the complete report to a computer or mobile storage device. During implementation, the module can switch display modes according to user operation commands, supports zooming in to view key parameters, and automatically stores the most recent 100 test results for a period of ≥1 year, facilitating subsequent query and comparison. Its intuitive result presentation lowers the threshold for clinical use and improves the practicality and convenience of testing.

[0025] Preferably, the feature extraction and processing module includes: a signal denoising and filtering unit, a feature factor separation unit, a feature priority ranking unit, and an effective feature screening unit. The signal denoising and filtering unit uses multi-band signal separation technology to remove interference signals from the electrical signals transmitted by the multi-dimensional signal sensing module. The feature factor separation unit uses a multi-dimensional feature decomposition algorithm to split the denoised electrical signals into biological feature-related factors. The feature priority ranking unit allocates and ranks feature factor weights based on the correlation between features and gastrointestinal tumors. The effective feature screening unit extracts and retains high-weight feature factors based on the ranking results.

[0026] Specifically, the signal noise reduction and filtering unit of the feature extraction and processing module adopts an 8-band adaptive filtering technology, with each band having a bandwidth of 100 Hz. This precisely removes 50 Hz power frequency interference and high-frequency noise above 10 kHz from the electrical signal transmitted by the multi-dimensional signal sensing module, achieving a filtering attenuation of ≥40 dB to ensure a signal-to-noise ratio ≥35 dB after noise reduction. The feature factor separation unit, based on a multi-dimensional feature decomposition algorithm, divides the noise-reduced electrical signal into 12 core factors according to spectral features, electrochemical features, impedance features, etc. During the decomposition process, the sampling frequency is maintained at 10 kHz to ensure that no key information is lost from the feature factors. The feature priority ranking unit calculates the priority of each feature factor... The correlation coefficients between the tumor markers and gastrointestinal tumor markers were calculated, with values ​​ranging from 0.1 to 0.9. Weights were assigned based on the coefficient values, with tumor-specific protein-related features having a weight of 0.8 to 0.9 and metabolite-related features having a weight of 0.6 to 0.7. The effective feature screening unit had a weight threshold of 0.3, and features with weights higher than the threshold were selectively extracted. During extraction, the data transmission rate was maintained at 500 kilobits per second. The amount of effective feature data after screening was controlled to be 30% to 40% of the original data, reducing redundant data while retaining core information, providing high-quality data support for subsequent calculations, and significantly improving detection efficiency and accuracy.

[0027] Preferably, the data transmission interaction module includes: a transmission protocol adaptation unit, a data encryption processing unit, a transmission status monitoring unit, and a data integrity verification unit. The transmission protocol adaptation unit dynamically matches the transmission protocol according to the interface protocol type of the model operation and analysis module. The data encryption processing unit uses an asymmetric encryption algorithm to encrypt the feature data output by the feature extraction processing module. The transmission status monitoring unit captures the link status parameters during the data transmission process in real time and makes dynamic adjustments. The data integrity verification unit verifies the consistency of the data before and after transmission using checksum comparison technology.

[0028] Specifically, the data transmission interaction module's transmission protocol adaptation unit has a built-in database of 16 mainstream communication protocols. It can dynamically match protocols within 0.1 seconds based on the interface type of the model calculation and analysis module, supporting conversion between multiple protocol formats such as USB 3.0, Wi-Fi 6, and Bluetooth 5.2. The data encryption processing unit uses an asymmetric encryption algorithm with a key length of 2048 bits. The encryption process takes ≤0.3 seconds, providing end-to-end encryption protection for characteristic data to prevent data leakage during transmission. The transmission status monitoring unit collects parameters such as signal strength, transmission rate, and bit error rate of the transmission link in real time. The signal strength monitoring accuracy reaches 0.1 dB / mW, the transmission rate monitoring range is 1 Mbps to 5 Gbps, and the bit error rate threshold is set at [value missing]. When the parameter exceeds the threshold, an early warning is triggered immediately. The data integrity verification unit adopts 128-bit cyclic redundancy check code technology to compare the data before and after transmission byte by byte. The verification time is ≤0.2 seconds. If data is missing or incorrect, the retransmission mechanism is automatically started. The maximum number of retransmissions is 3, which ensures that the feature data is transmitted to the model operation and analysis module completely and securely, and ensures the continuity and reliability of the detection link.

[0029] Preferably, the model operation and analysis module includes: a parameter initialization and configuration unit, a multi-model collaborative operation unit, an operation result fusion unit, and an analysis result optimization unit. The parameter initialization and configuration unit sets the initial parameters of the operation model according to the range of gastrointestinal tumor screening parameters. The multi-model collaborative operation unit calls a preset operation model to perform parallel operation processing on the feature data. The operation result fusion unit uses weighted fusion technology to integrate the operation results of the multiple models. The analysis result optimization unit corrects and adjusts the integrated results according to the fluctuation trend of the feature data.

[0030] Specifically, the parameter initialization configuration unit of the model operation and analysis module calibrates the initial values ​​of 16 core operation parameters based on 100,000 clinical screening data, including sample feature thresholds, operation step size, fusion coefficients, etc., with parameter adjustment accuracy to 6 decimal places, and can automatically adapt the initial parameter range according to different sample types; the multi-model collaborative operation unit integrates three types of operation models: feature matching model, parameter coupling model, and result prediction model, adopting a parallel operation architecture, with each model operation core frequency of 1.2 GHz, capable of processing 8 channels of feature data simultaneously, with a single model operation time ≤ 2 seconds, and the total time for simultaneous operation of the three types of models ≤ 3 seconds; the operation result fusion unit adopts dynamic weighted fusion technology, based on feature data... The fusion weights are adjusted in real time based on the data type. The fusion weights for spectral feature-related results range from 0.3 to 0.4, for electrochemical feature-related results from 0.3 to 0.4, and for impedance feature-related results from 0.2 to 0.3. During the fusion process, the data processing precision is maintained to 6 decimal places. The analysis result optimization unit monitors the fluctuation range of feature data, with a fluctuation range calculation precision of 0.01. When the fluctuation range is greater than 0.2, the optimization coefficient is set to 0.15; when the fluctuation range is less than 0.2, the optimization coefficient is set to 0.05. This targeted correction of the integrated calculation results effectively reduces the impact of data fluctuations on the screening results and improves the accuracy and stability of the analysis results.

[0031] Preferably, the feature factor operation expression used by the model operation and analysis module is: in, This is the comprehensive value calculated from the feature factors. , , These are the feature weight coefficients. For adjustment coefficients, These are the protein characteristic values ​​of biological samples. These are nucleic acid sequence characteristic parameters. This is a value associated with metabolite concentration. These are cell morphology characteristic parameters. As a tumor marker-associated factor, The baseline value for normal tissue characteristics, This represents the rate of change of signal strength.

[0032] Specifically, feature factor computation is based on the synergistic mechanism of multiple biomarkers in the development and progression of gastrointestinal tumors. Through systematic analysis of core tumor-related biomarkers in biological samples, it is clarified that the influence of various indicators on tumor screening differs. Therefore, a multi-parameter coupling algorithm is used to construct the model, introducing weight coefficients and adjustment coefficients to balance the influence of different biomarkers and the overall computational scale. The weight coefficients are set according to the degree of correlation between various biomarkers and tumors; the higher the correlation, the larger the coefficient value. The adjustment coefficients are calibrated based on a large amount of clinical screening data to calibrate computational biases caused by different sample types. In implementation, the feature extraction processing module first separates and extracts various biomarker parameters from the converted electrical signals. These parameters have been validated with clinical samples and can truly reflect the biological characteristics of the samples. After substituting the parameters into the model, the comprehensive computational value of the feature factors is obtained through step-by-step coupling computation. The model is logically designed to align with the intrinsic link between biological characteristics and tumor development. By comprehensively integrating multi-dimensional biological information, it avoids the limitations of single-feature analysis, enhances the sensitivity of tumor-related feature identification, and provides accurate and comprehensive data support for subsequent screening analysis. It ensures that the calculation results can objectively reflect the tumor risk correlation of the samples. Its construction approach considers both the independence of biological characteristics and the interaction between various features, making the calculation results more in line with actual clinical screening needs.

[0033] Preferably, the signal transformation expression of the multidimensional signal sensing module is: ,in, The converted electrical signal value, The conversion factor is... For the first Bio-like signal intensity, For the first Signal-to-weight conversion, For the first Physiological parameter test values, For the first Class parameter baseline value, The change in signal frequency. For the stable value of the signal period, This represents the number of signal and parameter categories.

[0034] Specifically, the signal conversion model is based on the physical laws governing the conversion of multidimensional biological signals into electrical signals. It incorporates the physical characteristics of optical, electrochemical, and impedance signals, and considers that factors such as signal intensity, physiological parameters, and frequency variations all affect the conversion results. Furthermore, it recognizes the inherent differences in conversion efficiency among different types of biological signals. Therefore, a combination of linear superposition and nonlinear computation is used to construct the model, and conversion coefficients are set to calibrate the conversion process for different signal types. These conversion coefficients are calibrated based on the conversion efficiency and stability of various signals, ensuring that the converted electrical signals have a uniform and comparable scale. During implementation, the multidimensional signal sensing module captures multiple biological signals from biological samples, inputs various signal parameters into the model, and completes the signal conversion through the model's pre-set computational logic. During the conversion process, the inherent differences between different signal types are eliminated through conversion coefficients, strictly controlling conversion errors. The model is established in strict accordance with the physical principles of signal conversion. Through scientific operation and coefficient calibration, it achieves standardized and unified conversion of different types of biological signals, avoiding deviations in subsequent feature extraction due to differences in signal types. This lays the foundation for the integrated analysis of multi-dimensional features. Its construction logic takes into account both the accuracy and efficiency of signal conversion, and can quickly provide high-quality electrical signal data for the feature extraction and processing module.

[0035] Preferably, the signal decoding expression of the result output feedback module is: in, The decoded and visualized signal values. These are the decoding coefficients. The result is the original signal value. This is the decoding gain coefficient. This is the signal format conversion factor. Signal strength before encoding This is the encoding loss value. Parameters for decoding error correction. For signal synchronization factor, As the decoding base, This is the signal compensation value.

[0036] Specifically, signal decoding is based on the inverse process of signal encoding and decoding. Considering the format requirements for visualizing screening results, it fully takes into account the impact of factors such as the original signal, encoding loss, and decoding error on the decoding effect. Therefore, a multi-step operational model is constructed, and the decoding weights of different parameters are adjusted through decoding coefficients to ensure that the decoded signal meets the visualization requirements. The decoding coefficients are set according to the difficulty of signal format conversion, error correction requirements, and signal synchronization requirements. Different coefficient values ​​correspond to different parameters to balance the impact of each step on the decoding result. During implementation, the result output feedback module receives the encoded screening results transmitted by the model operation and analysis module, extracts key information such as the original signal, encoding loss, and error parameters, inputs these parameters into the model, and completes signal decoding, error correction, and format conversion according to preset steps, ultimately obtaining a signal that can be directly visualized. The model's construction closely matches the actual needs of decoding and visualization. Through multi-step operations and coefficient adjustment, it effectively compensates for signal loss during the encoding process, reduces decoding errors, and ensures the accuracy and clarity of the visualization results. Its construction approach not only follows the basic principles of signal processing but also fully adapts to clinical application scenarios, allowing screening results to be intuitively understood and applied, thus improving the practicality of the detection device.

[0037] Preferably, the sample processing expression of the biological sample acquisition module is: ,in, The effective signal value after sample processing. For processing coefficients, For parameters related to sample collection volume, This is the sample activity retention coefficient. For sample processing time parameters, For sample storage environment parameters, This is a parameter representing the degree of sample homogenization. For the sample temperature change parameter, This is a parameter for adjusting sample concentration. This represents the total number of sample processing parameters. For the k-th sample processing parameter.

[0038] Specifically, the sample processing system revolves around the core requirement of preserving effective signals after biological sample collection. It systematically analyzes the impact of factors such as sample volume, activity retention, processing time, storage environment, and homogenization on the extraction of effective signals. Considering that some factors have a linear relationship with the results while others have a non-linear relationship, a model combining linear and non-linear operations is constructed, introducing processing coefficients to optimize the weights of different factors. These processing coefficients are calibrated based on the degree of influence of each factor on the effective signal, combined with clinical sample processing data, ensuring that the model accurately reflects the role of each parameter during sample processing. During implementation, after the biological sample collection module completes sample collection, it records key information such as sample volume correlation parameters, activity retention coefficients, and processing time parameters in real time. This information is input into the model, which calculates the effective signal value after sample processing. Based on this value, the parameter settings in the sample processing workflow are adjusted. This model focuses on the core objective of sample processing, comprehensively integrating the influencing factors of each stage of sample processing to avoid the loss of effective signals due to considering only one factor, thus improving the targeting and effectiveness of sample processing. Its construction logic not only conforms to the biological laws of sample processing but also provides high-quality samples for subsequent signal sensing, ensuring the data reliability of the entire detection chain.

[0039] Preferably, the optimized expression for the transmission efficiency of the data transmission interaction module is: ,in, For optimized transmission efficiency, To optimize the coefficients, For transmission bandwidth parameters, For signal transmission rate, For transmission delay parameters, For data cache capacity, For transmission link stability parameters, For interference signal strength parameters, For anti-interference capability parameters, Optimize the number of parameter categories for transmission.

[0040] Specifically, the transmission efficiency optimization model, based on core data transmission performance requirements, deeply analyzes the impact of factors such as transmission bandwidth, transmission rate, latency, buffer capacity, link stability, and interference signal strength on transmission efficiency. It clarifies that the mechanisms of action of different factors vary, with some having a positive impact and others a negative one. Therefore, a multi-parameter weighted calculation is used to construct the model, balancing the effects of each factor through optimization coefficients to maximize transmission efficiency. The optimization coefficients are calibrated based on the weight of each factor's impact on transmission efficiency, combined with actual data from different transmission scenarios, ensuring the model is adaptable to both wired and wireless transmission modes. During implementation, the data transmission interaction module monitors key parameters such as bandwidth, rate, latency, and interference strength in real time during transmission. These parameters are input into the model, and the model calculates the optimized transmission efficiency value. Based on this value, the transmission protocol, bandwidth allocation, and anti-interference strategies are dynamically adjusted. The model is designed to meet the actual needs of data transmission. Through multi-factor integration and coefficient optimization, it effectively improves transmission rate, reduces latency, and minimizes interference, ensuring the integrity and efficiency of feature data transmission. Its construction approach not only follows the technical principles of data transmission but also adapts to the multi-scenario usage needs of detection devices, ensuring the continuity and stability of the detection link and providing timely and complete data support for subsequent calculation and analysis.

[0041] like Figure 2 As shown, a detection method for gastrointestinal tumor screening is applied to a detection device for gastrointestinal tumor screening, comprising the following steps: S1, collecting and sealing relevant gastrointestinal biological samples from the subject using a biological sample collection module; S2, capturing and converting multidimensional signals from the collected biological samples using a multidimensional signal sensing module; S3, performing noise reduction and feature factor separation and screening on the converted electrical signals using a feature extraction and processing module; S4, transmitting the screened feature data to a model calculation and analysis module using an encryption protocol via a data transmission interaction module; S5, processing the feature data using a multi-parameter coupled calculation algorithm in the model calculation and analysis module. The data is processed and screening analysis results are generated. In step S6, the screening analysis results are decoded, converted, and visualized through the result output feedback module. Each step achieves efficient transmission and accurate calculation of feature data through the collaborative operation between modules, forming a complete detection process from sample collection to result output. The calculation process adopts a dynamic parameter adjustment mechanism to optimize the calculation model in real time, ensuring the targeting of feature data processing and the accuracy of analysis results. The transmission process adopts link status monitoring technology to dynamically adjust the data transmission link, ensuring the integrity and security of data transmission. Each step is executed sequentially according to the time sequence, and the parameters of each link are adapted and optimized collaboratively through the feedback mechanism.

[0042] A detection device and method for screening gastrointestinal tumors are disclosed. The device comprises six modules, including a biological sample acquisition module and a multidimensional signal sensing module, which form a unified and compatible interactive system through embedded interfaces and a sealed data channel. Coupled with an encrypted transmission protocol and a data integrity verification mechanism, it achieves seamless integration of each step from sample acquisition to result output, completely solving the problems of feature information loss during signal conversion and data transmission loss in traditional detection methods. This modular and collaborative architecture not only makes the detection process more coherent and efficient but also enhances the stability of the overall detection system through the complementary functions of each module, ensuring the integrity and security of data during transmission and processing, and breaking the limitations of isolated operation of each technical step in traditional screening methods.

[0043] This invention utilizes a feature extraction and processing module for multi-dimensional feature separation, filtering, and prioritization, combined with a model computation and analysis module for multi-parameter coupling operations and dynamic weight allocation. This comprehensively considers the complex correlations among biomarkers related to gastrointestinal tumors, accurately matching individual differences among examinees. Simultaneously, through multi-model collaborative computation and result optimization, it effectively compensates for the shortcomings of traditional models in comprehensively considering feature coupling relationships, significantly improving the targeting and accuracy of feature processing and computational analysis. Combined with signal decoding and visualization in the result output feedback module, the screening results are more intuitive and reliable, completely changing the problem of poor result consistency caused by the limited accuracy of parameter processing in traditional technologies. This meets the core clinical needs for precise and efficient early screening of gastrointestinal tumors.

[0044] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0045] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A detection device for screening gastrointestinal tumors, comprising: Biological sample acquisition module, multidimensional signal sensing module, feature extraction and processing module, data transmission and interaction module, model calculation and analysis module, and result output and feedback module; The biological sample acquisition module establishes a bidirectional data link with the multidimensional signal sensing module through a sealed channel. The multidimensional signal sensing module converts the captured multidimensional biological signals into computable electrical signals and transmits them to the feature extraction and processing module. The feature extraction and processing module separates and filters the electrical signals into multidimensional feature factors and then sends them to the data transmission and interaction module. The data transmission and interaction module uses an encrypted transmission protocol to transmit the filtered feature data to the model operation and analysis module. The model operation and analysis module performs multi-parameter coupling operations on the feature data, generates screening analysis results, and sends them to the result output feedback module. The result output feedback module decodes the analysis results, converts the format, and then visualizes them. All modules interact and work together through embedded interfaces to form a complete gastrointestinal tumor screening and detection link.

2. The detection device for screening gastrointestinal tumors according to claim 1, characterized in that, The feature extraction and processing module includes: a signal denoising and filtering unit, a feature factor separation unit, a feature priority ranking unit, and an effective feature screening unit. The signal denoising and filtering unit uses multi-band signal separation technology to remove interference signals from the electrical signals transmitted by the multi-dimensional signal sensing module. The feature factor separation unit uses a multi-dimensional feature decomposition algorithm to split the denoised electrical signals into biological feature-related factors. The feature priority ranking unit assigns and ranks feature factors according to their correlation with gastrointestinal tumors. The effective feature screening unit extracts and retains high-weight feature factors based on the ranking results.

3. The detection device for screening gastrointestinal tumors according to claim 1, characterized in that, The data transmission interaction module includes: a transmission protocol adaptation unit, a data encryption processing unit, a transmission status monitoring unit, and a data integrity verification unit. The transmission protocol adaptation unit dynamically matches the transmission protocol according to the interface protocol type of the model operation and analysis module. The data encryption processing unit uses an asymmetric encryption algorithm to encrypt the feature data output by the feature extraction processing module. The transmission status monitoring unit captures the link status parameters in real time during the data transmission process and makes dynamic adjustments. The data integrity verification unit verifies the consistency of the data before and after transmission using checksum comparison technology.

4. The detection device for screening gastrointestinal tumors according to claim 1, characterized in that, The model operation and analysis module includes: a parameter initialization and configuration unit, a multi-model collaborative operation unit, an operation result fusion unit, and an analysis result optimization unit. The parameter initialization and configuration unit sets the initial parameters of the operation model according to the range of gastrointestinal tumor screening parameters. The multi-model collaborative operation unit calls the preset operation model to perform parallel operation processing on the feature data. The operation result fusion unit uses weighted fusion technology to integrate the operation results of multiple models. The analysis result optimization unit corrects and adjusts the integrated results according to the fluctuation trend of the feature data.

5. The detection device for screening gastrointestinal tumors according to claim 1, characterized in that, The feature factor operation expression used by the model operation and analysis module is: in, This is the comprehensive calculation value of the feature factors. , , These are the feature weight coefficients. For adjustment coefficients, These are the protein characteristic values ​​of biological samples. These are nucleic acid sequence characteristic parameters. This is a value associated with metabolite concentration. These are cell morphology characteristic parameters. As a tumor marker-associated factor, The baseline value for normal tissue characteristics, This represents the rate of change of signal strength.

6. The detection device for screening gastrointestinal tumors according to claim 1, characterized in that, The signal transformation expression of the multidimensional signal sensing module is: ,in, The converted electrical signal value, The conversion factor is... For the first Bio-like signal intensity, For the first Signal-to-weight conversion, For the first Physiological parameter test values, For the first Class parameter baseline value, The change in signal frequency. For the stable value of the signal period, This represents the number of signal and parameter categories.

7. The detection device for screening gastrointestinal tumors according to claim 1, characterized in that, The signal decoding expression of the result output feedback module is: ,in, The decoded and visualized signal values. These are the decoding coefficients. The result is the original signal value. This is the decoding gain coefficient. This is the signal format conversion factor. Signal strength before encoding This is the encoding loss value. Parameters for decoding error correction. For signal synchronization factor, As the decoding base, This is the signal compensation value.

8. The detection device for screening gastrointestinal tumors according to claim 1, characterized in that, The sample processing expression of the biological sample acquisition module is: ,in, The effective signal value after sample processing. For processing coefficients, For parameters related to sample collection volume, This is the sample activity retention coefficient. For sample processing time parameters, For sample storage environment parameters, This is a parameter representing the degree of sample homogenization. For the sample temperature change parameter, This is a parameter for adjusting sample concentration. This represents the total number of sample processing parameters. For the k-th sample processing parameter.

9. A detection device for screening gastrointestinal tumors according to claim 1, characterized in that, The efficiency optimization expression for the data transmission interaction module is as follows: ,in, For optimized transmission efficiency, To optimize the coefficients, For transmission bandwidth parameters, For signal transmission rate, For transmission delay parameters, For data cache capacity, For transmission link stability parameters, For interference signal strength parameters, For anti-interference capability parameters, Optimize the number of parameter categories for transmission.

10. A detection method for screening gastrointestinal tumors, characterized in that, This method is applied to the detection device for gastrointestinal tumor screening as described in claim 1, comprising the following steps: S1, collecting and sealing gastrointestinal biological samples from the subject at designated locations using a biological sample collection module; S2, capturing and converting multidimensional signals into electrical signals from the collected biological samples using a multidimensional signal sensing module; S3, performing noise reduction and feature factor separation and screening on the converted electrical signals using a feature extraction and processing module; S4, transmitting the screened feature data to a model operation and analysis module using an encryption protocol via a data transmission interaction module; and S5, performing operation and processing on the feature data using a multi-parameter coupled operation algorithm in the model operation and analysis module. The process generates screening analysis results. In step S6, the results are decoded, format converted, and visualized through the result output feedback module. Each step achieves efficient transmission and accurate calculation of feature data through the collaborative operation between modules, forming a complete detection process from sample collection to result output. The calculation process uses a dynamic parameter adjustment mechanism to optimize the calculation model in real time, ensuring the relevance of feature data processing and the accuracy of analysis results. The transmission process uses link status monitoring technology to dynamically adjust the data transmission link, ensuring the integrity and security of data transmission. Each step is executed sequentially according to the time sequence, and the parameters of each link are adapted and optimized collaboratively through the feedback mechanism.

Images