Abstract:Introduction: The research of auxiliary diagnosis has always been one of the hotspots in the world. The implementation of auxiliary diagnosis support algorithm for medical text data faces challenges with interpretability and creditability. The improvement of clinical diagnostic techniques means not only the improvement of diagnostic accuracy but also the further study of diagnostic basis. Traditional research methods for diagnostic markers often require a large amount of time and economic costs. Research objects are often dozens of samples, and it is, therefore, difficult to synthesize large amounts of data. Therefore, the comprehensiveness and reliability of traditional methods have yet to be improved. Therefore, the establishment of a model that can automatically diagnose diseases and automatically provide a diagnostic basis at the same time has a positive effect on the improvement of medical diagnostic techniques. Methods: Here, we established an auxiliary diagnostic tool based on attention deep learning algorithm to diagnostic hyperlipemia and automatically predict the corresponding diagnostic markers using hematological parameters. In this paper, we not only demonstrated the ability of the proposed model to automatically diagnose diseases using text-based medical data, such as physiological parameters, but also demonstrated its ability to forecast disease diagnostic markers. Human physiological parameters are used as input to the model, and the doctor's diagnosis results as an output. Through the attention layer, the degree of attention of the model to different physiological parameters can be obtained, that is, the model provides a diagnostic basis. Results: It achieved 94% ACC, 97.48% AUC, 96% sensitivity and 92% specificity with the test dataset. All the above samples are drawn from clinical practice. Moreover, the model predicted the diagnostic markers of hyperlipidemia by the attention mechanism, and the results were fully agreeable to the golden criteria. Discussion: The auxiliary diagnosis system proposed in this paper not only achieves the accurate and robust performance, and can be used for the preliminary diagnosis of patients, but also showing its great potential to discover new diagnostic markers. Therefore, it not only can improve the efficiency of clinical diagnosis but also shorten the research period of researching a diagnosis basis to an extent. It has a positive significance to the development of the medical diagnosis level.

Study On Computer-Aided Diagnosis Of Multi-Parameter Measured By Bio-Chemical Machine For Blood Based On Rough Set

An Interactive Task Analysis Framework and Interactive System Research for Computer Aided Diagnosis

The Preliminary Study of a Pocket Apparatus for Multi-parameter Bio-chemical Measurement of Blood

An Augmented Model with Inferred Blood Features for the Self-diagnosis of Metabolic Syndrome.

Application of a Decision Tree Model for Predicting Diabetic Retinopathy

Machine learning-based prediction of diabetic patients using blood routine data

DMSO_A_198547 637..645

The Application of Rough Set-Neural Network Based on c-Means Clustering in Heart Disease Diagnosation

Research on diabetes prediction method based on electronic medical record data analysis

A Data-experience intelligent model to integrate human judging behavior and statistics for predicting diabetes complications

Forecasting Potential Diabetes Complications.

Predicting coronary heart disease in Chinese diabetics using machine learning

Machine Learning Sensors for Diagnosis of COVID-19 Disease Using Routine Blood Values for Internet of Things Application

An optimal method for diagnosing heart disease using combination of grasshopper evalutionary algorithm and support vector machines

Intelligent Machine Learning Approach for Effective Recognition of Diabetes in E-Healthcare Using Clinical Data

Machine learning-based early detection of diabetes risk factors for improved health management

Rapid diagnosis of diabetes based on ResNet and Raman spectroscopy

Improved logistic regression model for diabetes prediction by integrating PCA and K-means techniques

Deep Learning-Based Method of Diagnosing Hyperlipidemia and Providing Diagnostic Markers Automatically

Rule Mining of Early Diabetes Symptom and Applied Supervised Machine Learning and Cross Validation Approaches based on the Most Important Features to Predict Early-stage Diabetes

Neighborhood rough set-based multi-attribute prediction approach and its application of gout patients