Breast cancer diagnosis based on a new improved Elman neural network optimized by meta‐heuristics

Abstract

In this article, a new optimized method for diagnosing and analyzing breast cancer from the mammography images is presented. In this regard, preprocessing is used to remove the Gaussian noises that are used to happen in the mammography images and also to remove the additional areas. Then, image segmentation is performed on the images to determine the areas where the contrast material is perceptible. Afterward, combined feature extraction based on a discrete wavelet transform and gray‐level co‐occurrence matrix is proposed for extracting the important information fromthe images. Finally, a new classification model based on an improved Elman neural network (ENN) has been proposed. The ENN is optimized by an improvedversion of the collective animal behavior algorithm. Simulation results areimplemented to the Mammographic Image Analysis Society database and the resultsare compared with three different methods.

Survival Risk Prediction of Esophageal Squamous Cell Carcinoma Based on BES-LSSVM

“…The results show that the proposed method performs better than existing methods. In addition, random forest [ 26 ], extreme learning machines [ 27 ], BP neural networks [ 28 , 29 ], and Elman neural networks [ 30 ] have achieved satisfactory results in the prognosis and diagnosis of certain cancers.…”

Automatic COVID‐19 CT segmentation using U‐Net integrated spatial and channel attention mechanism

“…A number of artificial intelligence (AI) systems based on deep learning have been proposed and results have been shown to be quite promising in medical image analysis 13‐16 . Compared to the traditional imaging workflow heavily relies on the human labors, AI enables more safe, accurate, and efficient imaging solutions.…”