Novel Disease Prediction System Using Hybrid Deep Learning Techniques.

      Аннотация: This Book has been carried out through three different models with a different combination of feature selection and deep learning techniques. The first model proposed the combination of the new Enhanced Grey-Wolf Optimization-based Feature Selection Algorithm (EGWO-FSA) and Deep Belief Network (DBN) for diagnosing heart, diabetes, and cancer disease. The second model proposed on disease prediction system which is developed by using the novel Genetic Binary Cuckoo Optimization Algorithm (GBCOA) and new Convolutional-Recurrent Neural Network (C-RNN) for identifying the heart, cancer, and diabetic diseases. The third technique implements a novel disease prediction system that has been developed by using the new Incremental Feature Selection Algorithm (IFSA) and novel Convolutional Neural Network with Temporal features (T-CNN) for predicting heart, diabetic, and cancer diseases., The proposed techniques are evaluated by conducting various experiments and achieved better performance in the proposed disease prediction system than the existing systems in terms of prediction accuracy and computation time.

Ключевые слова: Deep Learning, Feature Selection, Artificial Intelligence, Machine Learning, Disease Prediction, Convolutional neural networks, Recurrent Neural Network, Healthcare, IoT, deep belief network