Downloads: 0 | Views: 2
Research Paper | Information Technology | India | Volume 9 Issue 12, December 2020 | Popularity: 2.6 / 10
Machine Learning in Modern Healthcare: Leveraging Big Data for Early Disease Detection and Patient Monitoring
Karthik Chava
Abstract: Healthcare has witnessed significant transformations in recent years, fueled by developments in smart technologies and the Internet of Things. Modern healthcare is distinguished by remote monitoring, wearable biosensors, and timely personalized healthcare services in a rapid feedback loop. Recently, smart healthcare has opened applications in disease diagnosis and disease treatment, such as health basic assessment and detection of daily human activities. This has motivated healthcare organizations and different organizations to invest in the new healthcare sector of big data analytics. Big data in smart healthcare has unique and influential characteristics, which give rise to specific and challenging issues, and also requires new techniques, technologies, and platforms for outbreak area monitoring and emerging disease detection. In addition to the above challenges, a newly raised challenge for sensing-based healthcare applications is disease diagnosis. Rapid detection of human activities and its associated illness is helpful for taking timely actions for patients? safety assurance. In certain cases, the human activity is too complicated to be solely defined by experts, or operate according to experts? rules. New techniques relying on the concept of computational epidemiology are broadly applied, such as correlation measure, spread people simulation process involving parameter estimation, and state-space model-based and distance model-based spread event forecasting. All such techniques rely on event categorization on the very first step. However, outbreak detection in a new geographical area hardly needs human-doing-based categorization processes because of matching the geography and available data set?s inference before consulting experts. These challenges make the design of a big data analytical platform for fault data storage management, rapid emergency detection, and disease diagnosis in exactly the new area difficult and also necessary. This paper proposes an efficient and robust big data analytical platform for processing real-time and sensing-based healthcare applications. The platform deploys big data and related technologies to addressing the challenges raised by irregular and large quantity data streams from biosensors. The platform consists of many layers, which include storage management, real-time control, decision-making, patient classification, disease diagnosis, and data retrieval. Each layer is analyzed in detail, and case scenarios are taken to demonstrate the platform?s applicability and plausibility on real health data.
Keywords: Machine learning, modern healthcare, big data analytics, early disease detection, patient monitoring, predictive modeling, healthcare informatics, real-time health data, clinical decision support, medical diagnostics, AI in healthcare, health data mining, personalized healthcare, electronic health records (EHR), anomaly detection, health prediction models, remote patient monitoring, chronic disease management, supervised learning, healthcare data integration
Edition: Volume 9 Issue 12, December 2020
Pages: 1899 - 1910
DOI: https://www.doi.org/10.21275/SR201212164722
Please Disable the Pop-Up Blocker of Web Browser
Verification Code will appear in 2 Seconds ... Wait