Abstract: Big data is growing exponentially, and along with that is the technology and new algorithms being developed. With the accumulation of big data, Machine learning and Artificial Intelligence are getting implemented into newer spheres. One of the fields in the sphere is Healthcare and Biomedical. Early disease prediction, patient care, and community services can be made possible using this accumulation of big data, with the help of Machine Learning. Though predicting diseases using Machine Learning can be implemented, but the accuracy of prediction is reduced due to incomplete medical data. Moreover different regions have different chronic diseases depending on the geographical conditions and community, an outbreak of disease. To overcome the problem of incomplete data, an approach of the latent factor model is used to reconstruct the missing data. In this paper, we propose a new Convolutional Neural Network which is based on Multimodal Disease Risk Prediction algorithm (CNN-MDRP) with the help of structured and unstructured data directly from hospital and research institutes. None of the existing work is focused on both data types in the field of medical big data analytics. Compared to several previous prediction algorithms, our prediction algorithm has the prediction accuracy of approximately 95% and speed faster than that of CNN-UDRP, a Uni modal Disease Risk Prediction based prediction algorithm.

Keywords: CNN-MDRP, Big Data, healthcare, Naive Bayes Algorithm, Decision Tree algorithm, K-nearest neighbor Algorithm

PDF | DOI: 10.17148/IJARCCE.2019.8328

Open chat
Chat with IJARCCE