NoSQL for Health Informatics Data Analytics

Health care information system usually consists of patient’s data related to clinical and medical activities (Yu, Kollipara, Penmetsa & Elliadka, 2013). Usually the data is consisted of several components like radiology information system, picture archiving and communication system, laboratory information system, and policy and procedure management system. Among these, radiology information system consisted of large medical imaging and video data from ultrasound scans, x-rays, CT-scan, MRIs and so on. Some of the activities in handling the medical imaging require processing, formatting, storing, tracking and retrieval. Usually radiology information data is in structured as well as unstructured data format with requirement of large storage space (2013). Electronic health records (EHS) consisted of patients data such as medical issue, prescriptions, demographics, lab data, reports, clinical notes, and payment information (Vanathi & Khadir, 2017)). Biomedical signal data is collected from human body like electroneurogram (ENG), electromyogram (EMG), electrocardiogram (ECG), electroencephalogram (EEG), electrogastrogram (EGG) phonocardiogram (PCG), etc. Genomic data analysis provides deeper understanding of patient’s medical issue at gene level, which can help in offering tailored medications per individual’s gene pattern (2017).

NoSQL databases are schema-less, mostly open-source, low latency and horizontally scalable technology suited for health care information data (Yu et al., 2013). It can handle data in either structured or unstructured format. Some of the popular NoSQL database management systems are MongoDB, HBase and Cassandra. AUTHORS developed a n architecture for radiology information system consisted of No SQL data storage layer hosted on a private cloud, application layer catering radiology information system on an Amazon EC2 instance connected to end-user’s web browser and mobile application using internet. Patients and doctors can log into the radiology information system residing in Amazon’s EC2 instance and can fetch patient specific information like medical history, lab results, medications, appointments, test results etc. The application will run the query on the back-end NoSQL MongoDB database hosted on a private cloud to fetch patient’s health data (2013). Vanathi and Khadir (2017) outlined a big data framework using Cassandra for healthcare information data due to scalability and high availability features without compromising performance. Overall architecture consisted of Apache Kafka with Apache Storm for stream computing and integrated with NoSQL Cassandra database. Other part for batch processing is based on Hadoop with HBase database. In this way, the architecture able to support both real time streaming and batch processing of health care data (2017).

References:

Vanathi, R. & Khadir, A. (2017). A robust architectural framework for big data stream computing in personal healthcare real time analytics. 2017 World Congress on Computing and Communication Technologies.

Yu, W., Kollipara, M., Penmetsa, R., & Elliadka, S. (2013). A distributed storage solution for cloud-based e-healthcare information system. 2013 IEEE International Conference on e-Health Networking, Applications and Services.