Effective Health Care Analytics & Diagnosing using Artificial intelligence Approaches
Recently, effective Medical Image Analysis and diagnosis reported a noticeably improvements due to Deep Learning approach. However, building a deep learning based health care diagnosing system faces many challenges in the Medical Imaging. One of these challenges is the analysis of brain imaging to accurately diagnosis Autism Spectrum Disorder (ASD) which is a brain-based disorder characterized by social deficits and repetitive behaviors. The present paper proposes a new deep learning model to classify individuals with ASD. The proposed model consists of stacked three autoencoders formed to enhance the brain images and obtain important features then classify the individuals as autistic or healthy controlled. A large dataset of ASD patients' known as Autism Brain Imaging Data Exchange dataset (ABIDE) is used as analytic case. The proposed model recorded a promising result of 70% accuracy compared to literature study.
Keywords: Diagnosing system, Autism; Medical Image; Autoencoder; Deep learning.