A Medical-History-Based Potential Disease Prediction Algorithm

Wenxing Hong,Nannan Zheng,Yang Weng,Ziang Xiong

doi:10.1109/access.2019.2940644

Abstract

As an important application of medical informatization, healthcare big data analysis has been extensively researched in the fields of intelligent consultation, disease diagnosis, intelligent question-answering doctors, and medical assistant decision support, and have made many achievements. In order to improve the comprehensiveness and pertinence of the medical examination, this paper intends to use healthcare big data analysis combined with deep learning technology to provide patients with potential diseases which is usually neglected for lacking of professional knowledge, so that patients can do targeted medical examinations to prevent health condition from getting worse. Inspired by the existing recommendation methods, this paper proposes a novel deep-learning-based hybrid recommendation algorithm, which is called medical-history-based potential disease prediction algorithm. The algorithm predicts the patient's possible disease based on the patient's medical history, providing a reference to patients and doctors to reduce the problem of delaying treatment due to unclear description of the symptom or limited professional knowledge. The experimental results show that our approach improves the accuracy of the potential diseases prediction.

Highlights

With the rapid development of the internet, electronic medical information has become popular with many cities around the world, such as electronic medical records(EMR) to replace traditional paper medical records, online appointments, and online reports, accumulated large-scale of healthcare data
Evaluation Metrics: We adopt two metrics to evaluate the performance of models in our experiment: Hit Ratio at rank k (HR@k) [26] and Normalized Discounted Cumulative Gain at rank k (NDCG@k) [19], [28], set k = 10
In this paper, we proposed a medical-history-based potential disease prediction algorithm, and overcame the shortcomings of the state-of-the-art models listed in section IV.B with combining the advantages of FM and deep neural networks (DNN)

Summary

Introduction

With the rapid development of the internet, electronic medical information has become popular with many cities around the world, such as electronic medical records(EMR) to replace traditional paper medical records, online appointments, and online reports, accumulated large-scale of healthcare data. The most common categories include: healthcare services data [1], [2], biomedical data [3]–[6], health insurance data, medical research and management data [7], public health data [8], [9], behavioral and emotional data, health statistics, population management data, and environmental data. Many approaches of these fields are developed upon machine learning [10], it is usually used to deal with classification, regression and feature extraction [11] problems. Potential disease prediction can help patient do targeted medical examinations by pointing the possible diseases, otherwise patients may miss significant medical examinations for lacking of professional knowledge, which will lead to severe health problems

Methods

Results

Conclusion