Experiences of building a medical data acquisition system based on two-level modeling

Electronic health records in Brazil: Prospects and technological challenges

Electronic Health Records (EHR) are critical tools for advancing digital health worldwide. In Brazil, EHR development must follow specific standards, laws, and guidelines that contribute to implementing beneficial resources for population health monitoring. This paper presents an audit of the main approaches used for EHR development in Brazil, thus highlighting prospects, challenges, and existing gaps in the field. We applied a systematic review protocol to search for articles published from 2011 to 2021 in seven databases (Science Direct, Web of Science, PubMed, Springer, IEEE Xplore, ACM Digital Library, and SciELO). Subsequently, we analyzed 14 articles that met the inclusion and quality criteria and answered our research questions. According to this analysis, 78.58% (11) of the articles state that interoperability between systems is essential for improving patient care. Moreover, many resources are being designed and deployed to achieve this communication between EHRs and other healthcare systems in the Brazilian landscape. Besides interoperability, the articles report other considerable elements: (i) the need for increased security with the deployment of permission resources for viewing patient data, (ii) the absence of accurate data for testing EHRs, and (iii) the relevance of defining a methodology for EHR development. Our review provides an overview of EHR development in Brazil and discusses current gaps, innovative approaches, and technological solutions that could potentially address the related challenges. Lastly, our study also addresses primary elements that could contribute to relevant components of EHR development in the context of Brazil's public health system. Systematic review registration: PROSPERO, identifier CRD42021233219, https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021233219.

High-Risk Prediction of Cardiovascular Diseases via Attention-Based Deep Neural Networks

High-risk prediction of cardiovascular disease is of great significance and impendency in medical fields with the increasing phenomenon of sub-health these years. Most existing pathological methods for the prognosis prediction are either costly or prone to misjudgement. Therefore, plenty of automated models based on machine learning have been proposed to predict the onset of cardiovascular disease with the premorbid information of patients extracted from their historical Electronic Health Records (EHRs). However, it is a tough job to select proper features from longitudinal and heterogeneous EHRs, and also a great challenge to obtain accurate and robust representations for patients. In this paper, we propose an entirely end-to-end model called DeepRisk based on attention mechanism and deep neural networks, which can not only learn high-quality features automatically from EHRs, but also efficiently integrate heterogeneous and time-ordered medical data, and finally predict patients' risk of cardiovascular diseases. Experiments are carried out on a real medical dataset and results show that DeepRisk can significantly improve the high-risk prediction accuracy for cardiovascular disease compared with state-of-the-art approaches.

Experience and reflection from China's Xiangya medical big data project

The construction of medical big data includes several problems that need to be solved, such as integration and data sharing of many heterogeneous information systems, efficient processing and analysis of large-scale medical data with complex structure or low degree of structure, and narrow application range of medical data. Therefore, medical big data construction is not only a simple collection and application of medical data but also a complex systematic project. This paper introduces China's experience in the construction of a regional medical big data ecosystem, including the overall goal of the project; establishment of policies to encourage data sharing; handling the relationship between personal privacy, information security, and information availability; establishing a cooperation mechanism between agencies; designing a polycentric medical data acquisition system; and establishing a large data centre. From the experience gained from one of China's earliest established medical big data projects, we outline the challenges encountered during its development and recommend approaches to overcome these challenges to design medical big data projects in China more rationally. Clear and complete top-level design of a project requires to be planned in advance and considered carefully. It is essential to provide a culture of information sharing and to facilitate the opening of data, and changes in ideas and policies need the guidance of the government. The contradiction between data sharing and data security must be handled carefully, that is not to say data openness could be abandoned. The construction of medical big data involves many institutions, and high-level management and cooperation can significantly improve efficiency and promote innovation. Compared with infrastructure construction, it is more challenging and time-consuming to develop appropriate data standards, data integration tools and data mining tools.

DAQBroker - A general purpose instrument monitoring framework

The current scientific environment has experimentalists and system administrators allocating large amounts of time for data access, parsing and gathering as well as instrument management. This is a growing challenge since there is an increasing number of large collaborations with significant amount of instrument resources, remote instrumentation sites and continuously improved and upgraded scientific instruments. DAQBroker is a new software framework adopted by the CLOUD experiment at CERN. This framework was designed to monitor CLOUD’s network of various architectures and operating systems and collect data from any instrument while also providing simple data access to any user. Data can be stored in one or several local or remote databases running on any of the most popular relational databases (MySQL, PostgreSQL, Oracle). It also provides the necessary tools for creating and editing the meta data associated with different instruments, perform data manipulation and generate events based on instrument measurements, regardless of the user’s know-how of individual instruments. This submission will present an overview of each of DAQBroker’s components as well as provide preliminary performance results of the application running on high and low performance machines.

Prediction of Hemodialysis Timing Based on LVW Feature Selection and Ensemble Learning

We propose an improved model based on LVW embedded model feature extractor and ensemble learning for improving prediction accuracy of hemodialysis timing in this paper. Due to this drawback caused by feature extraction models, we adopt an enhanced LVW embedded model to search the feature subset by stochastic strategy, which can find the best feature combination that are most beneficial to learner performance. In the model application, we present an improved integrated learners for model fusion to reduce errors caused by overfitting problem of the single classifier. We run several state-of-the-art Q&A methods as contrastive experiments. The experimental results show that the ensemble learning model based on LVW has better generalization ability (97.04%) and lower standard error (± 0.04). We adopt the model to make high-precision predictions of hemodialysis timing, and the experimental results have shown that our framework significantly outperforms several strong baselines. Our model provides strong clinical decision support for physician diagnosis and has important clinical implications.