Intelligent hospital terminal doctor seeing service method and system

Abstract

The invention discloses an intelligent hospital terminal treatment service method and system, and relates to the technical field of medical informationization, and the method comprises the specific steps: enabling a patient to input personal information through an intelligent hospital terminal system, including the name, age, gender and medical history, and storing the information in a patient database, through the deep learning algorithm and the artificial intelligence technology, a large amount of medical image data and other clinical data can be automatically analyzed, key information is rapidly extracted, accurate auxiliary diagnosis information is provided for doctors, the time for the doctors to process and analyze the data is shortened, the diagnosis process is more rapid and efficient, and meanwhile the diagnosis efficiency is improved. By learning a large amount of case data, more accurate disease prediction and diagnosis suggestions can be provided, the risk of diagnosis errors is effectively reduced, and the accuracy and reliability of diagnosis are improved.

Description

Technical Field

[0001] This invention relates to the field of medical information technology, specifically to a smart hospital terminal medical service method and system. Background Technology

[0002] With the rapid development of medical technology, the medical service field is undergoing unprecedented changes. In modern society, people are paying more and more attention to personal health, and the demand for medical services is also increasing. In order to meet this demand, hospitals need to continuously improve their service quality and efficiency. However, the traditional consultation model often relies on the doctor's personal experience and judgment, which greatly limits the accuracy and efficiency of medical services. Therefore, how to use modern technological means, especially artificial intelligence technology, to assist doctors in disease diagnosis has become an urgent problem to be solved in the current medical service field.

[0003] Although electronic medical record systems and telemedicine platforms have improved the efficiency of medical services to some extent in recent years, they still have many limitations. Not only is the phenomenon of information silos widespread and data exchange between different medical institutions difficult, making it difficult to share patient information across institutions and affecting the provision of continuous medical services, but most clinical decision support systems are based on rule engines, which have limited ability to assist in the diagnosis of complex diseases and are difficult to adapt to the rapid updates of medical knowledge and the needs of personalized diagnosis and treatment. Traditional medical procedures involve many manual operations, which are prone to human error and affect the efficiency and accuracy of diagnosis. The interpretation of medical images is highly dependent on the experience and skills of radiologists, and the diagnostic results are greatly affected by subjective factors. Moreover, in the face of large-scale image data, human processing capacity is stretched thin.

[0004] However, existing AI-assisted diagnostic systems lack comprehensive integration and analysis of patient information, often failing to detect potential disease risks and missing the optimal time for early intervention. Furthermore, the lack of in-depth communication and collaboration with doctors often prevents the full utilization of doctors' clinical experience and expertise, potentially leading to biased diagnostic results. Moreover, the lack of timely response and processing of patient feedback often fails to meet individualized patient needs, reducing the patient's overall experience. Therefore, developing an AI-assisted diagnostic system that comprehensively integrates patient information and fully utilizes AI technology is crucial for improving the accuracy and efficiency of medical services. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a smart hospital terminal medical service method and system. It can improve the efficiency and accuracy of hospital diagnosis through artificial intelligence technology, optimize the patient's medical experience, and provide doctors with real-time auxiliary diagnostic information by using a smart terminal system combined with patient personal information, medical images and other clinical data, so as to achieve rapid and accurate disease diagnosis.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for providing medical services via a smart hospital terminal, the specific steps of which are as follows:

[0007] S100, Patient Information Entry: Patients enter their personal information, including name, age, gender, and medical history, through the smart hospital terminal system, and this information is saved to the patient database;

[0008] S200, Medical Image Upload: Patients upload medical image data to the terminal, which automatically analyzes the medical images and extracts key information;

[0009] S300, Clinical Data Integration: Integrates patient personal information, medical images, clinical laboratory test results, and vital signs data to form complete medical record information;

[0010] S400, Artificial Intelligence-Assisted Diagnosis: Constructs a deep learning-based diagnostic model, trains it on a large amount of labeled medical data, learns the characteristic manifestations of diseases, automatically extracts features, and intelligently assesses the patient's condition. The specific steps of S400 are as follows:

[0011] S401, Data Input: The integrated case information, including basic patient information, medical imaging data, and laboratory test results, is input into the deep learning model;

[0012] S402, Feature Extraction: Use the Convolutional Neural Network (CNN) in the deep learning model to extract features from medical images and capture pathological features in the images;

[0013] S403, Model Training: Select the deep learning framework of convolutional neural network, train the model using a large number of labeled case datasets, learn the correlation between case features and diagnostic results, and optimize the loss function to minimize prediction error;

[0014] S404, Model Validation and Tuning: Validate and evaluate the performance of the model, and adjust the hyperparameters based on the validation results;

[0015] S405, Auxiliary Diagnostic Information Generation: The trained model infers from new case information and outputs auxiliary diagnostic information such as disease characteristics, lesion degree, and treatment plan;

[0016] S500, Doctor's Diagnosis Confirmation: Based on the provided auxiliary diagnostic information and combined with clinical experience, the doctor makes the final diagnosis confirmation. The doctor can enter and modify the diagnosis results.

[0017] S600, Patient Feedback and Optimization: Patients view diagnostic results through the terminal, and doctors adjust and optimize the diagnostic results based on patient feedback.

[0018] Furthermore, the S100 collects basic patient information, symptom descriptions, past medical history, and physiological parameters through the hospital terminal system, including self-service machines, mobile apps, and intelligent triage robots. Patients can also make appointments through the smart hospital terminal, selecting the department, doctor, and appointment time. Based on the number of doctors, the number of patients, and the order of registration, the system automatically determines the appointment time and time period for each patient.

[0019] Furthermore, the specific steps of S200 are as follows:

[0020] S201, Patient Preparation and Upload Selection: Patients select the medical image upload function through the smart hospital terminal system and upload medical image data, including X-rays, MRI, and CT scans;

[0021] S202, File Selection and Upload: Patients and medical staff select medical image files from local storage devices and cloud storage and transfer them to the server;

[0022] S203, Verification and Confirmation: Automatically verify the integrity and correctness of the file format, and display a message to the user indicating successful upload;

[0023] S204, Processing and Storage: On the server side, the uploaded medical image files are further compressed and encrypted before being stored in the database and smart terminal system.

[0024] Furthermore, the specific steps of S300 are as follows:

[0025] S301, Patient Data Reception: Retrieve patient information from the patient database, receive and store medical imaging data uploaded by patients, and obtain other clinical data of patients, including laboratory test results and vital signs data;

[0026] S302, Data Cleaning: Preprocess the received data to remove invalid, incomplete, and erroneous information;

[0027] S303, Data Standardization: Comparing and synthesizing clinical data from different sources and types, and standardizing the data;

[0028] S304, Information Integration: Integrate the cleaned and standardized data together to form a comprehensive patient case file;

[0029] S305, Information Storage: Store the integrated case information in a database.

[0030] Furthermore, during the data cleaning process of S302, judgments are made on eliminating invalid, incomplete, and incorrect information, that is, according to the integrity I c of the information, the validity I e of the information, and the accuracy I a of the information for comprehensive evaluation. The comprehensive evaluation value I v = I c × I e × I a . By analyzing historical case data, the distribution of the comprehensive evaluation values of valid, complete, and accurate information is statistically obtained to determine the evaluation threshold T. For I v < T, it is determined that the information is invalid, incomplete, and incorrect information and is eliminated.

[0031] Furthermore, S303 performs Z-score standardization on the blood pressure and blood glucose level indicators with symmetric distributions. Its algorithm formula is: x i is the original data value, μ is the average value of all sample data, and σ is the standard deviation of the sample data.

[0032] Furthermore, S402 uses a convolutional neural network CNN to perform a layer-by-layer feature learning process on the input medical image data during feature extraction. Through the structure of the convolutional layer, pooling layer, and fully connected layer, it automatically extracts features related to clinical data from the original data and constructs a diagnostic model. The convolutional layer is responsible for extracting features in the input medical image data, and multiple convolutional kernels are set to detect different features, that is: C(i,j) is the result of the convolution operation at position (i,j), I is the input medical image, K represents the convolutional kernel, m and n are index parameters used to determine the position where the convolutional kernel slides on the input medical image, and M and N are the size parameters of the convolutional kernel in two dimensions. The pooling layer downsamples the output of the convolutional layer, that is: P(i,j) is the result after pooling. The fully connected layer matches and classifies the obtained feature map with the label, that is: y = f(Wx + b), where y is the output vector, W is the weight matrix, x is the input vector, and b is the bias vector. By traversing the entire input data, a complete feature map is obtained.

[0033] Furthermore, S400 compares the predicted output of the diagnostic model with the actual diagnostic results, and calculates the loss value using the mean squared error loss function, i.e.: Where L represents the loss value, n represents the number of case data points used to calculate the loss value, i is the index parameter used to iterate through the samples in the case dataset, and y i p represents the model's predicted value for the disease characteristics of the i-th sample. i Let represent the actual probability of disease occurrence in the i-th sample. The calculated loss value is propagated to each layer of the model through the backpropagation algorithm to calculate the gradient of each parameter, i.e.: Where w is the weight, used to determine the connection strength between the gradients of each parameter in the deep learning model. It is the learning rate. Based on the calculated gradient, the optimization algorithm is used to update the model's parameters, that is: The model parameters are continuously iterated and optimized until the loss value converges. The trained model is then used to predict new medical images, generate diagnostic results, and evaluate the model's performance.

[0034] A smart hospital terminal medical service system, the system comprising:

[0035] User Information Management Module: Responsible for managing patients' basic information. Patients can enter and update their personal information through the smart hospital terminal system, and the patient database is updated in real time.

[0036] Image data analysis module: Receives and processes medical image data uploaded by patients, including X-rays, MRI, and CT, and automatically analyzes the image data to extract diagnostic information;

[0037] Comprehensive data analysis module: Integrates patient personal information, medical imaging data, and other clinical data to form a comprehensive and accurate case information database;

[0038] Intelligent Diagnostic Assistance Module: Based on deep learning algorithms, it automatically analyzes the case information output by the comprehensive data analysis module to assist doctors in making diagnostic decisions;

[0039] Expert Diagnosis Confirmation Module: Doctors can view the auxiliary diagnostic information provided by the intelligent diagnostic assistance module and make a final diagnosis confirmation based on their own clinical experience and professional knowledge. The system supports doctors in entering, modifying and confirming diagnostic results.

[0040] The doctor-patient interaction optimization module provides patients with real-time diagnostic progress and result query functions. Patients can ask questions and provide feedback to doctors through the terminal system, and doctors can adjust and optimize the diagnostic results based on the patient's feedback.

[0041] Furthermore, the medical service system is equipped with hardware facilities such as a smart hospital terminal system, medical image processing equipment, servers, and network equipment. In terms of software design, it develops software systems with modules for user information management, image data analysis, comprehensive data analysis, intelligent diagnostic assistance, expert diagnosis confirmation, and doctor-patient interaction optimization.

[0042] Compared with existing technologies, this smart hospital terminal medical service method and system has the following beneficial effects:

[0043] I. This invention utilizes deep learning algorithms and artificial intelligence technology to automatically analyze large amounts of medical imaging data and other clinical data, quickly extract key information, and provide doctors with accurate auxiliary diagnostic information. This reduces the time doctors spend processing and analyzing data, making the diagnostic process faster and more efficient. At the same time, through learning from a large amount of case data, it can provide more accurate disease predictions and diagnostic suggestions, effectively reducing the risk of diagnostic errors and improving the accuracy and reliability of diagnoses.

[0044] Second, this invention allows patients to conveniently input personal information and upload medical images through a smart hospital terminal system, and view diagnostic progress and results in real time, reducing waiting time and cumbersome procedures, making the medical process more convenient and efficient. In addition, patients can also communicate with doctors in a timely manner through the system, provide feedback on their condition and questions, and obtain more personalized and precise medical services, thereby improving patient satisfaction and medical experience.

[0045] Third, this invention, through the networking of telemedicine technology and equipment, enables the smart hospital terminal medical service system to achieve optimized allocation and sharing of medical resources. Doctors can remotely view and analyze patients' medical records, providing timely and professional medical services to patients, effectively alleviating the problem of strained medical resources. At the same time, the system can also share patients' medical records with other medical institutions and experts, promoting the exchange and dissemination of medical knowledge and experience, improving the overall level of medical services, and helping to achieve a balanced distribution and efficient utilization of medical resources, providing high-quality medical services to more patients.

[0046] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0048] Figure 1 This is an operation flowchart for a smart hospital terminal medical service method.

[0049] Figure 2 This is a flowchart of a smart hospital terminal medical service system. Detailed Implementation

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0051] Example 1

[0052] A method for providing smart hospital terminal medical services includes the following steps: patient information entry, medical image uploading, clinical data integration, artificial intelligence-assisted diagnosis, doctor diagnosis confirmation, and patient feedback and optimization.

[0053] First, the patient information entry stage (S100) begins. Patients use the smart terminal system provided by the hospital, including self-service machines, mobile applications, and intelligent triage robots, to complete the entry of their personal information. The entered information includes basic information such as name, age, gender, and past medical history. This information is securely stored in the patient database to provide basic data for subsequent diagnosis. At the same time, patients can make appointments through the terminal system, independently selecting the department, doctor, and time of visit. The system automatically arranges the appointment time according to hospital resources.

[0054] Then, the medical image uploading stage (S200) begins. Patients prepare X-ray, MRI, and CT medical images at the hospital and at home, and select the upload function through the smart hospital terminal. They choose the corresponding image files from local storage and cloud storage and upload them to the system. The system automatically verifies the integrity and format of the files. After ensuring that there are no errors, the system notifies the patient that the upload was successful and compresses and encrypts the files on the server side before storing them in the database.

[0055] Next, the system enters the clinical data integration phase (S300). It extracts patient information and medical imaging data from the database, combines this with laboratory test results and vital signs, performs data cleaning to remove invalid information, and then uses Z-score standardization to eliminate the influence of dimensions for symmetrically distributed indicators such as blood pressure and blood glucose levels. This makes data comparisons more direct and facilitates subsequent statistical analysis and machine learning model training. The algorithm formula is as follows: x i These are the raw data values, μ is the average of all sample data, and σ is the standard deviation of the sample data. These data are integrated into a comprehensive case information archive and stored in the database for subsequent analysis.

[0056] Subsequently, the AI-assisted diagnosis stage (S400) begins. The integrated case information is input into a deep learning model, which uses a convolutional neural network (CNN) to extract features from medical images and automatically identify pathological features. The model is trained using a large number of labeled case datasets to learn the relationship between case features and diagnostic results, optimize the loss function, and reduce prediction errors. After validation and performance evaluation, the model parameters are tuned to improve diagnostic accuracy. The trained model then infers on new cases and outputs the probability of disease, the degree of lesion, and treatment suggestions.

[0057] Finally, the process moves to the doctor's diagnosis confirmation stage (S500). The doctor reviews the information provided by the AI-assisted diagnosis, combines it with their personal clinical experience, and makes the final diagnosis confirmation. The doctor can enter and modify the diagnosis results to ensure the accuracy and authority of the diagnosis.

[0058] In addition, during the patient feedback and optimization adjustment phase (S600), patients can view the diagnostic results through their smart terminals. If they have any questions or need adjustments, they can immediately communicate with their doctors. Based on the patient's feedback, the doctors will make timely adjustments and optimizations to the diagnostic results to ensure patient satisfaction.

[0059] Through the smart hospital terminal medical service method described in this embodiment, patients can conveniently enter information, upload images, and make appointments. Doctors can obtain comprehensive patient information and intelligent auxiliary diagnostic information, thereby improving the efficiency and accuracy of medical services. At the same time, through timely communication between patients and doctors, diagnostic results can be further optimized, and patient satisfaction can be improved.

[0060] Example 2

[0061] This embodiment aims to describe in detail a smart hospital terminal medical service system. The system includes modules for user information management, image data analysis, comprehensive data analysis, intelligent diagnostic assistance, expert diagnosis confirmation, and doctor-patient interaction optimization. Through hardware facilities, the smart terminal system, medical image processing equipment, server and network equipment work together with a specially designed software system to achieve intelligent management of the entire process from information entry to assisted diagnosis and doctor-patient interaction.

[0062] In practice, patients first enter and update their basic personal information, such as name, age, gender, and medical history, through self-service machines, mobile apps, and intelligent triage robots in the smart hospital terminal system. This information is synchronized to the patient database in real time, ensuring that doctors can access the latest and most complete patient information to provide basic information for diagnosis. At the same time, patients can choose the department, doctor, and time of their visit, and the system will automatically arrange their visit based on resource allocation.

[0063] During the image data upload process, the system provides clear operation instructions to ensure that patients or medical staff can successfully select and upload medical image files from their local devices or the cloud. After upload, the system's built-in image processing algorithm automatically performs quality control and format verification to ensure the usability of the image data. Then, it identifies and marks possible abnormal areas and uses intelligent algorithms to preliminarily interpret image features, such as masses, inflammation, and other pathological changes, providing doctors with intuitive auxiliary information and speeding up the diagnosis.

[0064] The comprehensive data analysis module integrates a series of data processing algorithms. First, it deeply integrates various types of clinical data received, and removes irrelevant and erroneous information through data cleaning. Then, it applies Z-score standardization technology to enable data from different sources to be compared with each other, providing a unified benchmark for subsequent machine learning models. After data standardization, the system integrates all relevant information into a structured case file, making it easier for doctors to quickly and comprehensively understand the patient's health status.

[0065] The deep learning-based intelligent diagnostic assistance module uses a convolutional neural network (CNN) to perform in-depth analysis of integrated case information. The CNN model, with its carefully designed convolutional, pooling, and fully connected layers, extracts complex pathological features from medical images and combines them with clinical data. Through extensive training, the model learns to identify key characteristics of various diseases. During the training phase, the model optimizes the loss function to reduce prediction errors and undergoes cross-validation tuning to ensure high diagnostic accuracy. For new cases, the model can quickly output disease probability, lesion severity assessment, and treatment suggestions, providing doctors with scientific decision-making support. After reviewing the information provided by the intelligent diagnostic assistance module, doctors use their professional judgment to make a final diagnosis. The system allows doctors to directly enter and modify diagnostic results on the platform, ensuring the authority of the diagnosis. The doctor-patient interaction module, through instant messaging, allows patients to check the diagnostic progress at any time and ask questions about anything unclear. Doctors can respond quickly based on patient feedback and adjust diagnostic strategies when necessary. This instant feedback mechanism greatly enhances doctor-patient communication, improves the targeted nature of treatment, and increases patient satisfaction.

[0066] Finally, the system relies on advanced hardware facilities, including high-speed server clusters to handle massive amounts of data and model calculations, medical image processing equipment to ensure high-quality reading of image data, and stable and reliable network equipment to ensure secure and efficient data transmission. On the software side, the various functional modules work closely together, and the cloud service architecture enables real-time data synchronization and processing, ensuring the stable operation and rapid response of the system.

[0067] Based on the above implementation steps, a smart hospital terminal medical service system is introduced. Through technological innovation, this system effectively shortens the diagnosis cycle, improves the accuracy of diagnosis, optimizes doctor-patient interaction, and provides patients with more convenient, efficient, and personalized medical services. At the same time, it also provides medical institutions with more efficient resource management and medical service capabilities.

[0068] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that various other embodiments can be derived from it without departing from the spirit or essential characteristics of the invention.

[0069] His specific form implements the invention. Therefore, the embodiments should be considered from any point of view.

[0070] This invention is to be regarded as exemplary and not restrictive; the scope of the invention is defined by the appended claims.

[0071] Rather than being limited by the foregoing description, it is therefore intended to interpret the meaning of the equivalent elements falling under the claims and

[0072] All variations within the scope are included within this invention. No reference numerals in the claims should be included.

[0073] The term is considered to limit the scope of the claims involved.

Claims

1. A method for providing medical services via a smart hospital terminal, characterized in that, The specific steps of this medical service method are as follows: S100, Patient Information Entry: Patients enter their personal information, including name, age, gender, and medical history, through the smart hospital terminal system, and this information is saved to the patient database; S200, Medical Image Upload: Patients upload medical image data to the terminal, which automatically analyzes the medical images and extracts key information; S300, Clinical Data Integration: Integrates patient personal information, medical images, clinical laboratory test results, and vital signs data to form complete medical record information; S400, Artificial Intelligence-Assisted Diagnosis: Constructs a deep learning-based diagnostic model, trains it on a large amount of labeled medical data, learns the characteristic manifestations of diseases, automatically extracts features, and intelligently assesses the patient's condition. The specific steps of S400 are as follows: S401, Data Input: The integrated case information, including basic patient information, medical imaging data, and laboratory test results, is input into the deep learning model; S402, Feature Extraction: Use the Convolutional Neural Network (CNN) in the deep learning model to extract features from medical images and capture pathological features in the images; S403, Model Training: Select the deep learning framework of convolutional neural network, train the model using a large number of labeled case datasets, learn the correlation between case features and diagnostic results, and optimize the loss function to minimize prediction error; S404, Model Validation and Tuning: Validate and evaluate the performance of the model, and adjust the hyperparameters based on the validation results; S405, Auxiliary Diagnostic Information Generation: The trained model infers from new case information and outputs auxiliary diagnostic information such as disease characteristics, lesion degree, and treatment plan; S500, Doctor's Diagnosis Confirmation: Based on the provided auxiliary diagnostic information and combined with clinical experience, the doctor makes the final diagnosis confirmation. The doctor can enter and modify the diagnosis results. S600, Patient Feedback and Optimization: Patients view diagnostic results through the terminal, and doctors adjust and optimize the diagnostic results based on patient feedback.

2. The method for providing medical services via a smart hospital terminal according to claim 1, characterized in that, The S100 collects basic patient information, symptom descriptions, past medical history, and physiological parameters through the hospital terminal system, including self-service machines, mobile apps, and intelligent triage robots. Patients can also make appointments through the smart hospital terminal, selecting the department, doctor, and appointment time. Based on the number of doctors, patients, and the order of registration, the system automatically determines the appointment time and time period for each patient.

3. The method for providing smart hospital terminal medical services according to claim 1, characterized in that, The specific steps of S200 are as follows: S201, Patient Preparation and Upload Selection: Patients select the medical image upload function through the smart hospital terminal system and upload medical image data, including X-rays, MRI, and CT scans; S202, File Selection and Upload: Patients and medical staff select medical image files from local storage devices and cloud storage and transfer them to the server; S203, Verification and Confirmation: Automatically verify the integrity and correctness of the file format, and display a message to the user indicating successful upload; S204, Processing and Storage: On the server side, the uploaded medical image files are further compressed and encrypted before being stored in the database and smart terminal system.

4. The method for providing medical services via a smart hospital terminal according to claim 1, characterized in that, The specific steps of S300 are as follows: S301, Patient Data Reception: Retrieve patient information from the patient database, receive and store medical imaging data uploaded by patients, and obtain other clinical data of patients, including laboratory test results and vital signs data; S302, Data Cleaning: Preprocess the received data to remove invalid, incomplete, and erroneous information; S303, Data Standardization: Comparing and synthesizing clinical data from different sources and types, and standardizing the data; S304, Information Integration: Integrating cleaned and standardized data to form a comprehensive patient case file; S305, Information Storage: Store the integrated case information in the database.

5. A method for providing medical services via a smart hospital terminal according to claim 4, characterized in that, During the data cleaning process in S302, it is determined whether to eliminate invalid, incomplete, and incorrect information, that is, according to the integrity I of the information c , the validity I of the information e and the accuracy I of the information a are comprehensively evaluated, and its comprehensive evaluation value I v = I c × I e × I a . By analyzing the historical case data, the distribution of the comprehensive evaluation values of valid, complete, and accurate information is statistically obtained to determine the evaluation threshold T. For I v < T, it is determined that the information is invalid, incomplete, and incorrect information and is eliminated.

6. The method for providing a smart hospital terminal medical service according to claim 4, characterized in that, S303 performs Z-score standardization on symmetrically distributed blood pressure and blood glucose levels, and the algorithm formula is as follows: x i σ is the original data value, μ is the mean of all sample data, and σ is the standard deviation of the sample data.

7. A method for providing medical services via a smart hospital terminal according to claim 1, characterized in that, The S402 utilizes a Convolutional Neural Network (CNN) to perform a layer-by-layer feature learning process on the input medical image data during feature extraction. Through a structure of convolutional layers, pooling layers, and fully connected layers, it automatically extracts clinically relevant features from the original data and constructs a diagnostic model. The convolutional layers are responsible for extracting features from the input medical image data, and multiple convolutional kernels are set to detect different features. C(i,j) is the result of the convolution operation at position (i,j), I is the input medical image, K represents the convolution kernel, m and n are index parameters used to determine the position of the convolution kernel sliding on the input medical image, M and N are the size parameters of the convolution kernel in two dimensions, and the pooling layer downsamples the output of the convolution layer, i.e.: P(i,j) is the result after pooling. The fully connected layer matches and classifies the obtained feature maps with the labels, i.e.: y = f(Wx + b), where y is the output vector, W is the weight matrix, x is the input vector, and b is the bias vector. By traversing the entire input data, the complete feature map is obtained.

8. A method for providing medical services via a smart hospital terminal according to claim 1, characterized in that, The S400 compares the predicted output of the diagnostic model with the actual diagnostic results, and calculates the loss value using the mean squared error loss function, i.e.: Where L represents the loss value, n represents the number of case data points used to calculate the loss value, i is the index parameter used to iterate through the samples in the case dataset, and y i p represents the model's predicted value for the disease characteristics of the i-th sample. i Let represent the actual probability of disease occurrence in the i-th sample. The calculated loss value is propagated to each layer of the model through the backpropagation algorithm to calculate the gradient of each parameter, i.e.: Where w is the weight, used to determine the connection strength between the gradients of each parameter in the deep learning model. It is the learning rate. Based on the calculated gradient, the optimization algorithm is used to update the model's parameters, that is: The model parameters are continuously iterated and optimized until the loss value converges. The trained model is then used to predict new medical images, generate diagnostic results, and evaluate the model's performance.

9. A smart hospital terminal medical service system according to any one of claims 1-8, characterized in that, The system components include: User Information Management Module: Responsible for managing patients' basic information. Patients can enter and update their personal information through the smart hospital terminal system, and the patient database is updated in real time. Image data analysis module: Receives and processes medical image data uploaded by patients, including X-rays, MRI, and CT, and automatically analyzes the image data to extract diagnostic information; Comprehensive data analysis module: Integrates patient personal information, medical imaging data, and other clinical data to form a comprehensive and accurate case information database; Intelligent Diagnostic Assistance Module: Based on deep learning algorithms, it automatically analyzes the case information output by the comprehensive data analysis module to assist doctors in making diagnostic decisions; Expert Diagnosis Confirmation Module: Doctors can view the auxiliary diagnostic information provided by the intelligent diagnostic assistance module and make a final diagnosis confirmation based on their own clinical experience and professional knowledge. The system supports doctors in entering, modifying and confirming diagnostic results. The doctor-patient interaction optimization module provides patients with real-time diagnostic progress and result query functions. Patients can ask questions and provide feedback to doctors through the terminal system, and doctors can adjust and optimize the diagnostic results based on the patient's feedback.

10. A smart hospital terminal medical service system according to claim 9, characterized in that, The medical service system is equipped with hardware facilities such as a smart hospital terminal system, medical image processing equipment, servers and network equipment. In terms of software design, it develops software systems with modules for user information management, image data analysis, comprehensive data analysis, intelligent diagnostic assistance, expert diagnosis confirmation, and doctor-patient interaction optimization.

Images