A New Scalable Deep Learning Model of Pattern Recognition for Medical Diagnosis Using Model Aggregation and Model Selection

Abstract

In recent years, pattern recognition has become a research area with increasing importance using several techniques. One of the most common techniques used is deep learning. This paper presents a new deep learning model to pattern recognition for medical diagnosis. The uncovering of hidden structures is performed by feature selection, model aggregation, and model selection. The deep learning model has the ability to reach the optimal solution and create complex decision boundaries when used to look for and diagnose breast cancer. The evaluation, based on 10-fold cross-validation, showed that the proposed model, which is named BaggingSMF, yielded good results and performed better than radial basis function, bidirectional associative memory, and ELMAN neural networks. Experimental studies demonstrate the multidisciplinary applications of the model.