Now a day’s big data is the fastest and more widely used in every field. With the help of big data medical and health care sectors achieves their growth and with the help of big data benefit of an accurate medical data analysis, early disease prediction, accurate data of a patient can be securely stored and used. Moreover, the accuracy of an analysis can be reduces due to various reason like incomplete medical data. The analysis accuracy of TORCH infection detection is reduced when the quality of medical data is incomplete. To overcome these difficulties of incomplete data with the use of a latent factor model. The progression of deep learning contributes to aid in the decision- making process of experts to diagnose patients with TORCH infection. Propose a new Convolutional Neural Network based Multimodal Disease Risk Prediction (CNN-MDRP) algorithm using structured and unstructured data from hospital for effective prediction of TORCH infection.