Abstract Wearable sensor (WS) technology and social media platforms (SMPs) are increasingly being used to monitor healthcare, and this has opened up new possibilities as chronic illnesses become more common. The development of a novel technique for gathering patient data for effective healthcare monitoring relies heavily on WS and SMP. But WS-based continuous patient monitoring produces a great deal of healthcare data. Furthermore, the user-generated health information on social networking sites is unstructured and produced in massive amounts. The current healthcare monitoring systems struggle to adequately analyze the useful information that may be gleaned from social networking and sensor data. Additionally, processing healthcare big data for anomaly prediction does not work well with conventional machine learning methods. Therefore, the purpose of this study is to propose an automated healthcare monitoring system for better categorization accuracy via accurate data storage and analysis. A novel fuzzy binary temporal long short-term memory (FBTLSTM) approach is used to appropriately categorize healthcare data and forecast drug side effects and abnormal conditions in patients. Patients' healthcare data, including diabetes, blood pressure, mental health, and drug evaluations, is used to categorize their health status in the proposed system. The findings demonstrate that the suggested model accurately manages heterogeneous data, enhances the categorization of medical conditions, and improves the predictability of drug side effects.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|