Learning deep neural networks' architectures using differential evolution. Case study: Medical imaging processing

• Published in: Computers in biology and medicine Vol. 146; p. 105623
• Main Author: Belciug, Smaranda
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.07.2022; Elsevier Limited; The Author. Published by Elsevier Ltd
• Subjects: Accuracy; Algorithms; Artificial Intelligence; Artificial neural networks; Automation; Brain cancer; Brain research; Cancer; Cancer histopathological image; Cancer MRI scan; Classification; Colorectal cancer; Computer architecture; Coronaviruses; COVID-19; Datasets; Deep Learning; Design of experiments; Diagnosis; Diagnostic Imaging; Differential evolution; Evolutionary algorithms; Evolutionary computation; Fetal monitoring; Fetuses; Health care; Humans; Image classification; Internal Medicine; Lung cancer; Machine learning; Magnetic resonance imaging; Maternal-fetal ultrasound; Medical imaging; Medical personnel; Medical screening; Model accuracy; Mutation; Neural networks; Neural Networks, Computer; Other; Pandemics; Patients; Pregnancy complications; Recombination; Regulatory approval; Statistical assessment; Tumors; Ultrasonic imaging; Deep learning; Differential evolution; Cancer histopathological image; Maternal-fetal ultrasound; Statistical assessment; Cancer MRI scan
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2022.105623

The COVID-19 pandemic has changed the way we practice medicine. Cancer patient and obstetric care landscapes have been distorted. Delaying cancer diagnosis or maternal-fetal monitoring increased the number of preventable deaths or pregnancy complications. One solution is using Artificial Intelligence to help the medical personnel establish the diagnosis in a faster and more accurate manner. Deep learning is the state-of-the-art solution for image classification. Researchers manually design the structure of fix deep learning neural networks structures and afterwards verify their performance. The goal of this paper is to propose a potential method for learning deep network architectures automatically. As the number of networks architectures increases exponentially with the number of convolutional layers in the network, we propose a differential evolution algorithm to traverse the search space. At first, we propose a way to encode the network structure as a candidate solution of fixed-length integer array, followed by the initialization of differential evolution method. A set of random individuals is generated, followed by mutation, recombination, and selection. At each generation the individuals with the poorest loss values are eliminated and replaced with more competitive individuals. The model has been tested on three cancer datasets containing MRI scans and histopathological images and two maternal-fetal screening ultrasound images. The novel proposed method has been compared and statistically benchmarked to four state-of-the-art deep learning networks: VGG16, ResNet50, Inception V3, and DenseNet169. The experimental results showed that the model is competitive to other state-of-the-art models, obtaining accuracies between 78.73% and 99.50% depending on the dataset it had been applied on. •The architecture of a DL is learned through neuroevolution.•Differential evolution optimizes the CNN's architecture.•Experimental results on 5 datasets are presented.•A statistical analysis of the proposed model versus state-of-the-art ones is performed.