Machine learning approaches to identify systemic lupus erythematosus in anti-nuclear antibody-positive patients using genomic data and electronic health records.

BACKGROUND: Although the 2019 EULAR/ACR classification criteria for systemic lupus erythematosus (SLE) has required at least a positive anti-nuclear antibody (ANA) titer (≥ 1:80), it remains challenging for clinicians to identify patients with SLE. This study aimed to develop a machine learning (ML) approach to assist in the detection of SLE patients using genomic data and electronic health records. METHODS: Participants with a positive ANA (≥ 1:80) were enrolled from the Taiwan Precision Medicine Initiative cohort. The Taiwan Biobank version 2 array was used to detect single nucleotide polymorphism (SNP) data. Six ML models, Logistic Regression, Random Forest (RF), Support Vector Machine, Light Gradient Boosting Machine, Gradient Tree Boosting, and Extreme Gradient Boosting (XGB), were used to identify SLE patients. The importance of the clinical and genetic features was determined by Shapley Additive Explanation (SHAP) values. A logistic regression model was applied to identify genetic variations associated with SLE in the subset of patients with an ANA equal to or exceeding 1:640. RESULTS: A total of 946 SLE and 1,892 non-SLE controls were included in this analysis. Among the six ML models, RF and XGB demonstrated superior performance in the differentiation of SLE from non-SLE. The leading features in the SHAP diagram were anti-double strand DNA antibodies, ANA titers, AC4 ANA pattern, polygenic risk scores, complement levels, and SNPs. Additionally, in the subgroup with a high ANA titer (≥ 1:640), six SNPs positively associated with SLE and five SNPs negatively correlated with SLE were discovered. CONCLUSIONS: ML approaches offer the potential to assist in diagnosing SLE and uncovering novel SNPs in a group of patients with autoimmunity.

Machine learning approaches for the genomic prediction of rheumatoid arthritis and systemic lupus erythematosus.

BACKGROUND: Rheumatoid arthritis (RA) and systemic lupus erythematous (SLE) are autoimmune rheumatic diseases that share a complex genetic background and common clinical features. This study's purpose was to construct machine learning (ML) models for the genomic prediction of RA and SLE. METHODS: A total of 2,094 patients with RA and 2,190 patients with SLE were enrolled from the Taichung Veterans General Hospital cohort of the Taiwan Precision Medicine Initiative. Genome-wide single nucleotide polymorphism (SNP) data were obtained using Taiwan Biobank version 2 array. The ML methods used were logistic regression (LR), random forest (RF), support vector machine (SVM), gradient tree boosting (GTB), and extreme gradient boosting (XGB). SHapley Additive exPlanation (SHAP) values were calculated to clarify the contribution of each SNPs. Human leukocyte antigen (HLA) imputation was performed using the HLA Genotype Imputation with Attribute Bagging package. RESULTS: Compared with LR (area under the curve [AUC] = 0.8247), the RF approach (AUC = 0.9844), SVM (AUC = 0.9828), GTB (AUC = 0.9932), and XGB (AUC = 0.9919) exhibited significantly better prediction performance. The top 20 genes by feature importance and SHAP values included HLA class II alleles. We found that imputed HLA-DQA1*05:01, DQB1*0201 and DRB1*0301 were associated with SLE; HLA-DQA1*03:03, DQB1*0401, DRB1*0405 were more frequently observed in patients with RA. CONCLUSIONS: We established ML methods for genomic prediction of RA and SLE. Genetic variations at HLA-DQA1, HLA-DQB1, and HLA-DRB1 were crucial for differentiating RA from SLE. Future studies are required to verify our results and explore their mechanistic explanation.

A Mobile Phone-Based Approach for Hearing Screening of School-Age Children: Cross-Sectional Validation Study.

BACKGROUND: Pure-tone screening (PTS) is considered as the gold standard for hearing screening programs in school-age children. Mobile devices, such as mobile phones, have the potential for audiometric testing. OBJECTIVE: This study aimed to demonstrate a new approach to rapidly screen hearing status and provide stratified test values, using a smartphone-based hearing screening app, for each screened ear of school-age children. METHOD: This was a prospective cohort study design. The proposed smartphone-based screening method and a standard sound-treated booth with PTS were used to assess 85 school-age children (170 ears). Sound-treated PTS involved applying 4 test tones to each tested ear: 500 Hz at 25 dB and 1000 Hz, 2000 Hz, and 4000 Hz at 20 dB. The results were classified as pass (normal hearing in the ear) or fail (possible hearing impairment). The proposed smartphone-based screening employs 20 stratified hearing scales. Thresholds were compared with those of pure-tone average (PTA). RESULTS: A total of 85 subjects (170 ears), including 38 males and 47 females, aged between 11 and 12 years with a mean (SD) of 11 (0.5) years, participated in the trial. Both screening methods produced comparable pass and fail results (pass in 168 ears and fail in 2 ears). The smartphone-based screening detected moderate or worse hearing loss (average PTA>25 dB) accurately. Both the sensitivity and specificity of the smartphone-based screening method were calculated at 100%. CONCLUSIONS: The results of the proposed smartphone-based self-hearing test demonstrated high concordance with conventional PTS in a sound-treated booth. Our results suggested the potential use of the proposed smartphone-based hearing screening in a school-age population.

Implementation of Online Veterinary Hospital on Cloud Platform.

Pet markets involve in great commercial possibilities, which boost thriving development of veterinary hospital businesses. The service tends to intensive competition and diversified channel environment. Information technology is integrated for developing the veterinary hospital cloud service platform. The platform contains not only pet medical services but veterinary hospital management and services. In the study, QR Code andcloud technology are applied to establish the veterinary hospital cloud service platform for pet search by labeling a pet's identification with QR Code. This technology can break the restriction on veterinary hospital inspection in different areas and allows veterinary hospitals receiving the medical records and information through the exclusive QR Code for more effective inspection. As an interactive platform, the veterinary hospital cloud service platform allows pet owners gaining the knowledge of pet diseases and healthcare. Moreover, pet owners can enquire and communicate with veterinarians through the platform. Also, veterinary hospitals can periodically send reminders of relevant points and introduce exclusive marketing information with the platform for promoting the service items and establishing individualized marketing. Consequently, veterinary hospitals can increase the profits by information share and create the best solution in such a competitive veterinary market with industry alliance.

Access Scheme for Controlling Mobile Agents and its Application to Share Medical Information.

This study is showing the advantage of mobile agents to conquer heterogeneous system environments and contribute to a virtual integrated sharing system. Mobile agents will collect medical information from each medical institution as a method to achieve the medical purpose of data sharing. Besides, this research also provides an access control and key management mechanism by adopting Public key cryptography and Lagrange interpolation. The safety analysis of the system is based on a network attacker's perspective. The achievement of this study tries to improve the medical quality, prevent wasting medical resources and make medical resources access to appropriate configuration.

Design of a wireless sensor network platform for tele-homecare.

The problem of an ageing population has become serious in the past few years as the degeneration of various physiological functions has resulted in distinct chronic diseases in the elderly. Most elderly are not willing to leave home for healthcare centers, but caring for patients at home eats up caregiver resources, and can overwhelm patients' families. Besides, a lot of chronic disease symptoms cause the elderly to visit hospitals frequently. Repeated examinations not only exhaust medical resources, but also waste patients' time and effort. To make matters worse, this healthcare system does not actually appear to be effective as expected. In response to these problems, a wireless remote home care system is designed in this study, where ZigBee is used to set up a wireless network for the users to take measurements anytime and anywhere. Using suitable measuring devices, users' physiological signals are measured, and their daily conditions are monitored by various sensors. Being transferred through ZigBee network, vital signs are analyzed in computers which deliver distinct alerts to remind the users and the family of possible emergencies. The system could be further combined with electric appliances to remotely control the users' environmental conditions. The environmental monitoring function can be activated to transmit in real time dynamic images of the cared to medical personnel through the video function when emergencies occur. Meanwhile, in consideration of privacy, the video camera would be turned on only when it is necessary. The caregiver could adjust the angle of camera to a proper position and observe the current situation of the cared when a sensor on the cared or the environmental monitoring system detects exceptions. All physiological data are stored in the database for family enquiries or accurate diagnoses by medical personnel.

A novel assessment of flexibility by microcirculatory signals.

Flexibility testing is one of the most important fitness assessments. It is generally evaluated by measuring the range of motion (RoM) of body segments around a joint center. This study presents a novel assessment of flexibility in the microcirculatory aspect. Eighteen college students were recruited for the flexibility assessment. The flexibility of the leg was defined according to the angle of active ankle dorsiflexion measured by goniometry. Six legs were excluded, and the remaining thirty legs were categorized into two groups, group H (n = 15 with higher flexibility) and group L (n = 15 with lower flexibility), according to their RoM. The microcirculatory signals of the gastrocnemius muscle on the belly were monitored by using Laser-Doppler Flowmetry (LDF) with a noninvasive skin probe. Three indices of nonpulsatile component (DC), pulsatile component (AC) and perfusion pulsatility (PP) were defined from the LDF signals after signal processing. The results revealed that both the DC and AC values of the group H that demonstrated higher stability underwent muscle stretching. In contrast, these indices of group L had interferences and became unstable during muscle stretching. The PP value of group H was a little higher than that of group L. These primary findings help us to understand the microcirculatory physiology of flexibility, and warrant further investigations for use of non-invasive LDF techniques in the assessment of flexibility.

A mobile agent approach for secure integrated medical information systems.

Different patient-related information in medical organizations is the primary reference for medical personnel diagnosing, treating, and caring patients. With the rapid development of information technology, paper-based medical records have gradually been changed to electronic forms. However, different medical organizations present individual system specifications and data-saving formats so that the medical information of the same patient cannot be exchanged, shared, and securely accessed. In order not to largely change the present medical information systems as well as not to increase abundant costs, Virtual Integrated Medical-information Systems (VIMS) is proposed to assist various hospitals in information exchange. Furthermore, with Mobile Agent, the dispersed medical information can be securely integrated. It presents confidentiality, non-repudiation, source authentication, and integrity in network transmission. Virtual Integrated Medical-information Systems (VIMS) is a virtual electronic integration system combined with Mobile Agent technology. With the features of independence, adaptability, mobility, objectives, and autonomy, Mobile Agent is applied to overcome the problems from heterogeneous systems. With the features, the over-dispersed medical records can be integrated. Moreover, Mobile Agent can ensure the instantaneity and usability of medical records from which doctors can make the most appropriate evaluation and diagnoses. It will avoid the waste of medical resources, such as repetition medication, as well as become the reference of further consultation or health check. Not only can it improve the medical care quality, but it can be provided for medical research.

Deployment of secure mobile agents for medical information systems.

Changes in global population and demography, and advances in medicine have led to elderly population growth, creating aging societies from which elderly medical care has evolved. In addition, with the elderly susceptible to chronic diseases, this together with the changing lifestyles of young adults have not only pushed up patient numbers of chronic diseases, but also effected into younger patients. These problems have become the major focus for the health care industry. In response to patient demand and the huge shortage of medical resources, we propose remote healthcare medical information systems that combine patient physiological data acquisition equipment with real-time health care analyses. Since remote health care systems are structured around the Internet, in addition to considering the numerous public systems spread across insecure heterogeneous networks, compatibility among heterogeneous networks will also be another concern. To address the aforementioned issues, mobile agents are adopted. With a mobile agent's characteristics of easy adaptability to heterogeneity and autonomy, the problem of heterogeneous network environments can be tackled. To construct a hierarchical safe access control mechanism for monitoring and control of patient data in order to provide the most appropriate medical treatment, we also propose to use the Chinese Remainder Theorem and discrete logarithm to classify different levels of monitoring staff and hence, to grant permission and access according to their authorized levels. We expect the methods proposed can improve medical care quality and reduce medical resource wastage, while ensuring patient privacy. Finally, security analysis of the system is conducted by simulating a variety of typical attacks, from which it can be concluded that the constructed remote healthcare information system be secure.

The enhancement of security in healthcare information systems.

With the progress and the development of information technology, the internal data in medical organizations have become computerized and are further established the medical information system. Moreover, the use of the Internet enhances the information communication as well as affects the development of the medical information system that a lot of medical information is transmitted with the Internet. Since there is a network within another network, when all networks are connected together, they will form the "Internet". For this reason, the Internet is considered as a high-risk and public environment which is easily destroyed and invaded so that a relevant protection is acquired. Besides, the data in the medical network system are confidential that it is necessary to protect the personal privacy, such as electronic patient records, medical confidential information, and authorization-controlled data in the hospital. As a consequence, a medical network system is considered as a network requiring high security that excellent protections and managerial strategies are inevitable to prevent illegal events and external attacks from happening. This study proposes secure medical managerial strategies being applied to the network environment of the medical organization information system so as to avoid the external or internal information security events, allow the medical system to work smoothly and safely that not only benefits the patients, but also allows the doctors to use it more conveniently, and further promote the overall medical quality. The objectives could be achieved by preventing from illegal invasion or medical information being stolen, protecting the completeness and security of medical information, avoiding the managerial mistakes of the internal information system in medical organizations, and providing the highly-reliable medical information system.

A study on agent-based secure scheme for electronic medical record system.

Patient records, including doctors' diagnoses of diseases, trace of treatments and patients' conditions, nursing actions, and examination results from allied health profession departments, are the most important medical records of patients in medical systems. With patient records, medical staff can instantly understand the entire medical information of a patient so that, according to the patient's conditions, more accurate diagnoses and more appropriate in-depth treatments can be provided. Nevertheless, in such a modern society with booming information technologies, traditional paper-based patient records have faced a lot of problems, such as lack of uniform formats, low data mobility, slow data transfer, illegible handwritings, enormous and insufficient storage space, difficulty of conservation, being easily damaged, and low transferability. To improve such drawbacks, reduce medical costs, and advance medical quality, paper-based patient records are modified into electronic medical records and reformed into electronic patient records. However, since electronic patient records used in various hospitals are diverse and different, in consideration of cost, it is rather difficult to establish a compatible and complete integrated electronic patient records system to unify patient records from heterogeneous systems in hospitals. Moreover, as the booming of the Internet, it is no longer necessary to build an integrated system. Instead, doctors can instantly look up patients' complete information through the Internet access to electronic patient records as well as avoid the above difficulties. Nonetheless, the major problem of accessing to electronic patient records cross-hospital systems exists in the security of transmitting and accessing to the records in case of unauthorized medical personnels intercepting or stealing the information. This study applies the Mobile Agent scheme to cope with the problem. Since a Mobile Agent is a program, which can move among hosts and automatically disperse arithmetic processes, and moves from one host to another in heterogeneous network systems with the characteristics of autonomy and mobility, decreasing network traffic, reducing transfer lag, encapsulating protocol, availability on heterogeneous platforms, fault-tolerance, high flexibility, and personalization. However, since a Mobile Agent contacts and exchanges information with other hosts or agents on the Internet for rapid exchange and access to medical information, the security is threatened. In order to solve the problem, this study proposes a key management scheme based on Lagrange interpolation formulas and hierarchical management structure to make Mobile Agents a more secure and efficient access control scheme for electronic patient record systems when applied to the access of patients' personal electronic patient records cross hospitals. Meanwhile, with the comparison of security and efficacy analyses being the feasibility of validation scheme and the basis of better efficiency, the security of Mobile Agents in the process of operation can be guaranteed, key management efficacy can be advanced, and the security of the Mobile Agent system can be protected.

Mobile agent application and integration in electronic anamnesis system.

Electronic anamnesis is to transform ordinary paper trails to digitally formatted health records, which include the patient's general information, health status, and follow-ups on chronic diseases. Its main purpose is to let the records could be stored for a longer period of time and could be shared easily across departments and hospitals. Which means hospital management could use less resource on maintaining ever-growing database and reduce redundancy, so less money would be spent for managing the health records. In the foreseeable future, building up a comprehensive and integrated medical information system is a must, because it is critical to hospital resource integration and quality improvement. If mobile agent technology is adopted in the electronic anamnesis system, it would help the hospitals to make the medical practices more efficiently and conveniently. Nonetheless, most of the hospitals today are still using paper-based health records to manage the medical information. The reason why the institutions continue using traditional practices to manage the records is because there is no well-trusted and reliable electronic anamnesis system existing and accepted by both institutions and patients. The threat of privacy invasion is one of the biggest concerns when the topic of electronic anamnesis is brought up, because the security threats drag us back from using such a system. So, the medical service quality is difficult to be improved substantially. In this case, we have come up a theory to remove such security threats and make electronic anamnesis more appealing for use. Our theory is to integrate the mobile agent technology with the backbone of electronic anamnesis to construct a hierarchical access control system to retrieve the corresponding information based upon the permission classes. The system would create a classification for permission among the users inside the medical institution. Under this framework, permission control center would distribute an access key to each user, so they would only allow using the key to access information correspondingly. In order to verify the reliability of the proposed system framework, we have also conducted a security analysis to list all the possible security threats that may harm the system and to prove the system is reliable and safe. If the system is adopted, the doctors would be able to quickly access the information while performing medical examinations. Hence, the efficiency and quality of healthcare service would be greatly improved.

A secure integrated medical information system.

The rapid rise and development of the internet has made digitization of our everyday life common. E-medicine, including electronic prescription records, electronic prescriptions, diagnosis information systems, and others are now being regarded as future trends. As development on the structure and format of electronic patient records and prescriptions matures, the implementation of a comprehensive medical information system is imperative, one which is constructed from integrating the various electronic information systems that is being developed. It is important to allow the implementation of such a system applicable to the present medical environment, which facilitates the integration of electronic patient record from all levels of medical centers and clinics, secures the transmission of these integrated patient records between them, enables the combined use of electronic prescriptions with patients' medications, and permits anonymous or confidential transmission of patients' private data. To put the ideas into practice, in this study, we would like to propose an Integrated Medical Information System.

Identification of a novel cell cycle regulated gene, HURP, overexpressed in human hepatocellular carcinoma.

An analytic strategy was followed to identify putative regulatory genes during the development of human hepatocellular carcinoma (HCC). This strategy employed a bioinformatics analysis that used a database search to identify genes, which are differentially expressed in human HCC and are also under cell cycle regulation. A novel cell cycle regulated gene (HURP) that is overexpressed in HCC was identified. Full-length cDNAs encoding the human and mouse HURP genes were isolated. They share 72 and 61% identity at the nucleotide level and amino-acid level, respectively. Endogenous levels of HURP mRNA were found to be tightly regulated during cell cycle progression as illustrated by its elevated expression in the G(2)/M phase of synchronized HeLa cells and in regenerating mouse liver after partial hepatectomy. Immunofluorescence studies revealed that hepatoma up-regulated protein (HURP) localizes to the spindle poles during mitosis. Overexpression of HURP in 293T cells resulted in an enhanced cell growth at low serum levels and at polyhema-based, anchorage-independent growth assay. Taken together, these results strongly suggest that HURP is a potential novel cell cycle regulator that may play a role in the carcinogenesis of human cancer cells.