Thyroid health status detection system
By designing a thyroid health status detection system and utilizing pressure sensor arrays and data processing technology, the problem of inaccurate assessment results in existing technologies has been solved, achieving higher resolution and global thyroid health status assessment, and improving the objectivity and accuracy of the assessment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-20
- Publication Date
- 2026-03-13
AI Technical Summary
Existing thyroid health assessment techniques rely on physician experience, resulting in significant variations in results. Ultrasound elastography and single-point stiffness testing cannot effectively characterize the pressure gradient difference between the nodule edge and normal tissue, leading to insufficient assessment capabilities.
A thyroid health status detection system was designed, including a pressure sensor array, a pressure control end, a data acquisition end, and a data processing end. The pressure sensor array is repeatedly attached to the user's neck skin to collect pressure detection data, and the data processing end is used to extract and analyze features to provide a global thyroid health status assessment.
It improves the resolution and coverage of thyroid detection, provides global information, enhances the objectivity and accuracy of assessment results, and is well tolerated by users.
Smart Images
Figure CN121647601A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, specifically to a thyroid health status detection system. Background Technology
[0002] Accurate assessment of thyroid health is a key clinical focus, involving morphological aspects such as nodule size and location, as well as biomechanical properties such as hardness and elasticity. Current mainstream assessment techniques include palpation, ultrasound elastography, and isostatic hardness testing.
[0003] However, palpation relies on physician experience, and results vary significantly among different operators. Ultrasound elastography and single-point stiffness testing techniques have limitations: they can only acquire local stiffness information and cannot effectively characterize the pressure gradient difference between the nodule edge and normal tissue, resulting in insufficient ability to assess the heterogeneity of thyroid surface stiffness.
[0004] Therefore, designing an objective and accurate thyroid health status detection system has become an urgent problem to be solved in this field. Summary of the Invention
[0005] In view of this, in a first aspect, this application proposes a thyroid health status detection system, the system comprising: The pressure sensor array can be repeatedly applied to the user's neck skin; The pressure control terminal applies pressure to the pressure sensor array so that the pressure sensor array can obtain pressure detection data; The data acquisition terminal is connected to the pressure sensor array to acquire the pressure detection data. The data processing end is connected to the data acquisition end and obtains thyroid health status assessment data based on the pressure detection data.
[0006] Preferably, the pressure sensor array includes: a flexible substrate, an electrode unit, and multiple pressure sensors; The electrode unit is disposed on the surface of the flexible substrate, forming a two-dimensional electrode array with the flexible substrate; Each of the pressure sensors is disposed in the two-dimensional electrode array to form the pressure sensor array.
[0007] Further optimization: The flexible substrate includes: polydimethylsiloxane, or silicone, or hydrogel; Furthermore, the flexible substrate also includes: nano-silica particles.
[0008] Further optimization: The electrode unit is interdigitated or parallel plate-shaped, and its surface is coated with a polypyrrole insulating layer.
[0009] The electrode unit is made of silver nanowire electrode, graphene thin film electrode, or carbon nanotube electrode.
[0010] Further optimization: Each pressure sensor has a self-adhesive layer on its surface and is coated with a coupling agent.
[0011] Preferably, the pressure control terminal includes: Motion end; The control terminal controls the movement of the motion terminal to press the pressure sensor array via a force feedback algorithm.
[0012] More preferably, the moving end is a robotic arm, a pneumatic airbag, or an electric push rod.
[0013] More preferably, the data acquisition terminal is also used to acquire displacement data of the moving end.
[0014] Preferably, the data processing terminal includes: The preprocessing module calibrates and denoises the pressure detection data; The visualization module generates a pressure distribution map based on the calibrated and denoised pressure detection data. The feature extraction and analysis module extracts and analyzes features from the pressure distribution map to obtain the thyroid health status assessment data.
[0015] More preferably, the feature extraction and parsing module uses a convolutional neural network model, a random forest model, or a Transformer model for feature extraction and parsing.
[0016] More preferably, the system further includes: a supplementary detection terminal; The data acquisition terminal is also used to acquire supplementary thyroid test data; The visualization module is also used to perform multimodal fusion of the pressure distribution map and the thyroid supplementary test data to obtain comprehensive test data; The feature extraction and parsing module is also used to extract and parse features from the comprehensive detection data to obtain the thyroid health status assessment data.
[0017] More preferably, the supplementary detection end includes one or more of the following: an ultrasonic detection end, a nuclear magnetic resonance detection end, and an optical detection end.
[0018] The thyroid health status detection system provided in this application includes: a pressure sensor array that can be repeatedly attached to the user's neck skin; a pressure control end for applying pressure to the pressure sensor array; a data acquisition end for acquiring pressure detection data; and a data processing end for obtaining thyroid health status assessment data based on the pressure detection data. This application is non-invasive, and users tolerated it well in clinical trials. At the same time, the pressure sensor array can provide global information to reflect the heterogeneity of the thyroid surface hardness, thereby improving the objectivity and accuracy of the assessment results.
[0019] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description
[0020] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application, and the illustrative embodiments and descriptions thereof are used to explain this application. In the drawings: Figure 1 A schematic diagram of a thyroid health status detection system according to a preferred embodiment of this application; Figure 2 This is a schematic diagram of a pressure sensor array according to a preferred embodiment of this application.
[0021] Reference numerals: 1-Pressure sensor array; 11-Flexible substrate; 12-Electrode; 2-Pressure sensor; 3-Data acquisition end; 4-Data processing end. Detailed Implementation
[0022] The technical solution of this application will now be described in detail with reference to the accompanying drawings and embodiments.
[0023] This application provides a thyroid health status detection system, such as Figure 1 As shown, the system includes: a pressure sensor array 1, a pressure control terminal 2, a data acquisition terminal 3, and a data processing terminal 4. The pressure sensor array 1 comprises multiple pressure sensors arranged in an array to cover the area of the thyroid gland to be tested. The entire sensor array can be repeatedly applied to the user's neck skin, enabling multi-point, repeated measurements. The pressure control terminal 2 applies pressure to the pressure sensor array 1, allowing it to obtain pressure detection data. The data acquisition terminal 3 is connected to the pressure sensor array 1 and is used to acquire the pressure detection data. The data processing terminal 4 is connected to the data acquisition terminal 3 and obtains thyroid health status assessment data based on the pressure detection data.
[0024] The pressure detection data mentioned above is the data detected by the sensor array and the data of a non-single-point sensor. The data processing terminal 4 also obtains thyroid health status assessment data based on the data detected by the sensor array and the data of a non-single-point sensor. This thyroid health status assessment data can more comprehensively reflect the thyroid health status in the covered detection area, rather than the thyroid health status at a single measurement point.
[0025] In other words, the thyroid health status detection system provided in this application can apply pressure to the pressure sensor array 1 attached to the user's neck via the pressure control end 2, causing a change in the capacitance value of the pressure sensor. The data processing end 4 derives the hardness from the pressure value and obtains the thyroid health status assessment result based on the hardness value. This solution can improve the resolution and coverage of thyroid detection, and is also reusable.
[0026] For pressure sensor array 1, in some preferred embodiments, such as Figure 2 As shown, the pressure sensor array 1 includes a flexible substrate 11, electrode units 12, and multiple pressure sensors 13. The electrode units 12 are disposed on the surface of the flexible substrate 11, forming a two-dimensional electrode array together with the flexible substrate 11. Each pressure sensor 13 is disposed within the two-dimensional electrode array, forming the pressure sensor array 1. The two-dimensional electrode array can be understood as multiple electrode nodes arranged on a plane, forming a grid structure with "rows" and "columns." Each pressure sensor is arranged on the electrode nodes of the aforementioned grid structure, forming the pressure sensor array 1. For example, the electrode nodes can have a one-to-one mapping relationship with the sensors, or the M×N electrode array can be divided into several blocks, with each block mapping to one sensor.
[0027] The flexible substrate 11 comprises polydimethylsiloxane and nano-silica particles. The nano-silica particles enhance mechanical strength, while the polydimethylsiloxane provides flexibility. Alternatively, the polydimethylsiloxane can be replaced with silicone with a Shore A hardness of 30, or a hydrogel containing hyaluronic acid and exhibiting higher biocompatibility, thereby improving skin affinity while maintaining flexibility.
[0028] The electrode unit 12 has an interdigitated or parallel plate shape, and its surface is coated with a polypyrrole insulating layer to reduce contact resistance. Of course, to form a two-dimensional electrode array, the plate-shaped electrode unit 12 needs to have a grid composed of multiple independently addressable microelectrode points inside. Alternatively, the electrode unit 12 can use silver nanowire electrodes, or graphene film electrodes with better conductivity and transmittance greater than 90%, or carbon nanotube electrodes with higher mechanical strength and fracture strain greater than 15%.
[0029] The pressure sensor 13 has a self-adhesive layer on its surface and is coated with a coupling agent to enhance its fit to the curved surface of the neck.
[0030] In one specific embodiment, the flexible substrate 11 uses polydimethylsiloxane with added nano-silica particles as the substrate, wherein the mass fraction of the nano-silica particles is 5%, and the overall elongation at break of the flexible substrate 11 is greater than 200%. Two layers of interdigitated silver nanowire electrode units 12 are sputtered onto the surface of the flexible substrate 11. Each electrode unit 12 has a linewidth of 50 μm and a spacing of 50 μm, forming a two-dimensional electrode array with a unit spacing of 500 μm and a total of 10 × 10 electrode nodes (one electrode node corresponds to a pair of interdigitated electrodes). Miniature pressure sensors 13 with an accuracy between ±0.01 N are integrated on the electrode nodes of the two-dimensional electrode array. Simultaneously, a 0.1 mm thick medical-grade silicone self-adhesive layer is attached to the back of each pressure sensor 13, and the surface is coated with sodium carboxymethyl cellulose medical coupling agent, resulting in a curvature radius of 3 to 5 cm for the flexible substrate 11. In another specific embodiment, the flexible substrate 11 and the electrode unit 12 can be made of shape memory alloy springs. When combined with the flat electrode unit 12, they can also form an adaptive neck surface effect. The structure is simpler and the manufacturing cost is lower.
[0031] In some preferred embodiments, the pressure control end 2 includes a motion end and a control end. The control end controls the motion end to press the pressure sensor array 1 using a force feedback algorithm, so that the pressure sensor array 1 obtains pressure detection data. The motion end can be in the form of a robotic arm, a pneumatic airbag, or an electric push rod, etc.
[0032] In one specific implementation, the moving end is a robotic arm capable of performing actions such as gripping and pressing, with a repeatability accuracy within ±0.1mm. The control end uses a proportional-integral-derivative (PID) algorithm to convert the voltage control signal into a displacement correction amount for the robotic arm, controlling the robotic arm to apply a standardized pressure of 0.5 to 2N in steps of 0.1N to the pressure sensor array 1, and the pressure value is fed back in real time by the pressure sensor 13.
[0033] In some preferred embodiments, the data acquisition terminal 3 simultaneously acquires the pressure detection data output by the pressure control terminal 2 and the displacement data of the moving end, so as to distinguish different detection positions and corresponding pressure values.
[0034] In one specific implementation, the data acquisition terminal 3 includes an analog-to-digital converter of model LTC2498, which has a resolution of 24 bits and a sampling frequency of 50Hz, used to convert analog signals into digital signals.
[0035] In some preferred embodiments, the data processing terminal 4 includes a preprocessing module, a visualization module, and a feature extraction and analysis module. The preprocessing module calibrates and denoises the stress detection data. The visualization module obtains a stress distribution map based on the calibrated and denoised stress detection data. The feature extraction and analysis module uses a convolutional neural network model, a random forest model, or a Transformer model to extract and analyze features from the stress distribution map, obtaining thyroid health status assessment data. Specifically, the convolutional neural network model can analyze the data through training and learning from samples; the random forest model can analyze the data by processing nonlinear stress distribution features; and the Transformer model can analyze the data by capturing long-range dependent stress gradients.
[0036] In one specific implementation, the preprocessing module first pre-calibrates the "pressure-capacity" curve based on conditions of using a silicone standard block with a hardness of 50 kPa and a deformation of 5 mm, in order to quantify the sensor's "scale". At the same time, it combines a Kalman filter algorithm to remove high-frequency noise such as breathing interference, and uses a PT100 thermal resistor with an accuracy within ±0.1℃ to perform real-time data correction based on a preset temperature compensation coefficient.
[0037] The visualization module converts the capacitance changes in the pressure detection data into a pressure heatmap and generates a two-dimensional pressure distribution map using the Python matplotlib library. The coordinate accuracy of the distribution map is within ±0.5 mm.
[0038] The feature extraction and parsing module utilizes a convolutional neural network (CNN) model to extract parameters such as the maximum pressure value, hardness heterogeneity index, and nodule boundary pressure gradient from the pressure distribution map. The hardness heterogeneity index is obtained by dividing the pressure standard deviation by the pressure mean, and the nodule boundary pressure gradient is obtained by dividing the pressure difference by the nodule width. The CNN model then outputs thyroid health status assessment data based on these features. The input layer of this CNN model is a 10×10 pressure matrix, the hidden layers consist of three stacked convolutional layers with a kernel size of 3×3, and the output layer is the assessment result for benign / malignant classification.
[0039] To further improve the accuracy of the assessment, in some preferred embodiments, the thyroid health status detection system provided in this application also includes a supplementary detection end. This supplementary detection end serves as a complement to the stress detection method, and by combining the stress detection data and the supplementary detection data, more accurate detection data can be obtained. This supplementary detection end includes one or more of the following: an ultrasound detection end, an MRI detection end, and an optical detection end.
[0040] Based on this, data acquisition terminal 3 is also used to acquire supplementary thyroid test data. The visualization module is also used to perform multimodal fusion of the pressure distribution map and the supplementary thyroid test data to obtain comprehensive test data. The feature extraction and analysis module is also used to extract and analyze features from the comprehensive test data to obtain thyroid health status assessment data and ultrasound grayscale matrix.
[0041] In one specific implementation, the supplementary detection end is an ultrasonic detection end, and the data acquisition end 3 synchronously records ultrasonic deformation data at a sampling frequency of 40Hz. After obtaining the two-dimensional pressure distribution map, the visualization module also overlays the ultrasonic image and marks the nodule location. The input layer of the convolutional neural network model in the feature extraction and analysis module also includes an ultrasonic grayscale matrix, so that the convolutional neural network model can simultaneously analyze pressure features and ultrasonic features.
[0042] The thyroid health status detection system provided in this application includes: a pressure sensor array that can be repeatedly attached to the user's neck skin; a pressure control end for applying pressure to the pressure sensor array; a data acquisition end for acquiring pressure detection data; and a data processing end for obtaining thyroid health status assessment data based on the pressure detection data. This application is non-invasive, and users tolerated it well in clinical trials. At the same time, the pressure sensor array can provide global information to reflect the heterogeneity of the thyroid surface hardness, thereby improving the objectivity and accuracy of the assessment results.
[0043] Verification has shown that the pressure sensor array unit spacing of this application, with a spacing of 500 μm, achieves a spatial resolution of 0.5 mm, enabling it to capture hardness abrupt changes in the nodule edge ranging from 0.1 to 0.3 mm. The system's pressure detection error is less than 5%, and the coefficient of variation for multiple measurements by the same user is less than 3%. For micronodules smaller than 1 cm, a pressure heterogeneity index significantly higher than that of normal tissue can be detected, which is beneficial for early screening of abnormal conditions. Furthermore, when combined with pressure distribution and ultrasound deformation data, this application achieves an accuracy rate of over 92% in assessing a thyroid condition in a sample of 50 cases.
[0044] The preferred embodiments of this application have been described in detail above. However, this application is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this application, various simple modifications can be made to the technical solution of this application, and these simple modifications all fall within the protection scope of this application.
[0045] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this application will not describe the various possible combinations separately.
[0046] Furthermore, various different embodiments of this application can be combined in any way, as long as they do not violate the spirit of this application, they should also be regarded as the content disclosed by this invention.
Claims
1. A thyroid health status detection system, characterized in that, The system includes: A pressure sensor array (1) can be repeatedly applied to the user's neck skin; The pressure control terminal (2) applies pressure to the pressure sensor array (1) so that the pressure sensor array (1) can obtain pressure detection data; The data acquisition terminal (3) is connected to the pressure sensor array (1) to acquire the pressure detection data; The data processing terminal (4) is connected to the data acquisition terminal (3) and obtains thyroid health status assessment data based on the pressure detection data.
2. The system according to claim 1, characterized in that, The pressure sensor array (1) includes: a flexible substrate (11), an electrode unit (12), and multiple pressure sensors (13). The electrode unit (12) is disposed on the surface of the flexible substrate (11) and forms a two-dimensional electrode array with the flexible substrate (11); Each of the pressure sensors (13) is disposed in the two-dimensional electrode array to form the pressure sensor array (1).
3. The system according to claim 2, characterized in that: The flexible substrate (11) comprises: polydimethylsiloxane, or silicone, or hydrogel; Furthermore, the flexible substrate (11) also includes: nano-silica particles.
4. The system according to claim 2, characterized in that: The electrode unit (12) is interdigitated or parallel plate-shaped, and its surface is coated with a polypyrrole insulating layer. The electrode unit (12) is a silver nanowire electrode, a graphene thin film electrode, or a carbon nanotube electrode.
5. The system according to claim 2, characterized in that: Each pressure sensor (13) has a self-adhesive layer on its surface and is coated with a coupling agent.
6. The system according to claim 1, characterized in that, The pressure control terminal (2) includes: Motion end; The control end controls the movement of the motion end through a force feedback algorithm to press the pressure sensor array (1).
7. The system according to claim 6, characterized in that, The moving end is a robotic arm, a pneumatic airbag, or an electric push rod.
8. The system according to claim 6, characterized in that, The data acquisition terminal (3) is also used to acquire displacement data of the moving end.
9. The system according to claim 1, characterized in that, The data processing terminal (4) includes: The preprocessing module calibrates and denoises the pressure detection data; The visualization module generates a pressure distribution map based on the calibrated and denoised pressure detection data. The feature extraction and analysis module extracts and analyzes features from the pressure distribution map to obtain the thyroid health status assessment data.
10. The system according to claim 9, characterized in that, The feature extraction and parsing module uses a convolutional neural network model, a random forest model, or a Transformer model for feature extraction and parsing.
11. The system according to claim 9, characterized in that, The system also includes: a supplementary detection terminal; The data acquisition terminal (3) is also used to acquire supplementary thyroid test data; The visualization module is also used to perform multimodal fusion of the pressure distribution map and the thyroid supplementary test data to obtain comprehensive test data; The feature extraction and parsing module is also used to extract and parse features from the comprehensive detection data to obtain the thyroid health status assessment data.
12. The system according to claim 11, characterized in that, The supplementary detection end includes one or more of the following: an ultrasonic detection end, a nuclear magnetic resonance detection end, and an optical detection end.