ABSTRACT
Cancer is the leading cause of death in Taiwan. According to the Cancer Registration Report of Taiwan's Ministry of Health and Welfare, a total of 13,488 people suffered from lung cancer in 2016, making it the second-most common cancer and the leading cancer in men. Compared with other types of cancer, the incidence of lung cancer is high. In this study, the National Health Insurance Research Database (NHIRDB) was used to determine the diseases and symptoms associated with lung cancer, and a 10-year probability deep neural network prediction model for lung cancer was developed. The proposed model could allow patients with a high risk of lung cancer to receive an earlier diagnosis and support the physicians' clinical decision-making. The study was designed as a cohort study. The subjects were patients who were diagnosed with lung cancer between 2000 and 2009, and the patients' disease histories were back-tracked for a period, extending to ten years before the diagnosis of lung cancer. As a result, a total of 13 diseases were selected as the predicting factors. A nine layers deep neural network model was created to predict the probability of lung cancer, depending on the different pre-diagnosed diseases, and to benefit the earlier detection of lung cancer in potential patients. The model is trained 1000 times, the batch size is set to 100, the SGD (Stochastic gradient descent) optimizer is used, the learning rate is set to 0.1, and the momentum is set to 0.1. The proposed model showed an accuracy of 85.4%, a sensitivity of 72.4% and a specificity of 85%, as well as an 87.4% area under ROC (AUROC) (95%, 0.8604-0.8885) model precision. Based on data analysis and deep learning, our prediction model discovered some features that had not been previously identified by clinical knowledge. This study tracks a decade of clinical diagnostic records to identify possible symptoms and comorbidities of lung cancer, allows early prediction of the disease, and assists more patients with early diagnosis.
ABSTRACT
BACKGROUND: Personal health record (PHR) security, correctness, and protection are essential for health and medical services. Blockchain architecture can provide efficient data retrieval and security requirements. Exchangeable PHRs and the self-management of patient health can offer many benefits to traditional medical services by allowing people to manage their own health records for disease prevention, prediction, and control while reducing resource burdens on the health care infrastructure and improving population health and quality of life. OBJECTIVE: This study aimed to build a blockchain-based architecture for an international health record exchange platform to ensure health record confidentiality, integrity, and availability for health management and used Health Level 7 Fast Healthcare Interoperability Resource international standards as the data format that could allow international, cross-institutional, and patient/doctor exchanges of PHRs. METHODS: The PHR architecture in this study comprised 2 main components. The first component was the PHR management platform, on which users could upload PHRs, view their record content, authorize PHR exchanges with doctors or other medical health care providers, and check their block information. When a PHR was uploaded, the hash value of the PHR would be calculated by the SHA-256 algorithm and the PHR would be encrypted by the Rivest-Shamir-Adleman encryption mechanism before being transferred to a secure database. The second component was the blockchain exchange architecture, which was based on Ethereum to create a private chain. Proof of authority, which delivers transactions through a consensus mechanism based on identity, was used for consensus. The hash value was calculated based on the previous hash value, block content, and timestamp by a hash function. RESULTS: The PHR blockchain architecture constructed in this study is an effective method for the management and utilization of PHRs. The platform has been deployed in Southeast Asian countries via the Asia eHealth Information Network (AeHIN) and has become the first PHR management platform for cross-region medical data exchange. CONCLUSIONS: Some systems have shown that blockchain technology has great potential for electronic health record applications. This study combined different types of data storage modes to effectively solve the problems of PHR data security, storage, and transmission and proposed a hybrid blockchain and data security approach to enable effective international PHR exchange. By partnering with the AeHIN and making use of the network's regional reach and expert pool, the platform could be deployed and promoted successfully. In the future, the PHR platform could be utilized for the purpose of precision and individual medicine in a cross-country manner because of the platform's provision of a secure and efficient PHR sharing and management architecture, making it a reasonable base for future data collection sources and the data analytics needed for precision medicine.
Subject(s)
Blockchain/standards , Health Records, Personal/ethics , Information Storage and Retrieval/methods , Telemedicine/methods , HumansABSTRACT
Taiwan has been confronted with a serious problem of aging in recent years. The prevalence of the chronic diseases caused by aging is increased continuously, which has led to a high percentage of comorbidity and polypharmacy. The proportion of the elderly people with polypharmacy (over three to five kinds of drugs) is 81%. Under the situation of high comorbidity, the potentially inappropriate drug for the elderly have become a series problem. However, in order to promote personal health management, Taiwan's Ministry of Health and Welfare has released a service of "My Health Bank", which contains all the personal medical information issued by the National Health Insurance Department and can be downloaded by any individual person. This study designs a cloud-based personal health management platform to parse and store the information of "My Health Bank", establishes two databases, one for the health insurance drug table and one for the inappropriate medications. A warning of inappropriate personal medication will be generated based on a checking process. We expect that the application will enhance the safety of medication and improve the self-health management of the elderly.