An optimized deep learning architecture for breast cancer diagnosis based on improved marine predators algorithm

Abstract

Breast cancer is the second leading cause of death in women; therefore, effective early detection of this cancer can reduce its mortality rate. Breast cancer detection and classification in the early phases of development may allow for optimal therapy. Convolutional neural networks (CNNs) have enhanced tumor detection and classification efficiency in medical imaging compared to traditional approaches. This paper proposes a novel classification model for breast cancer diagnosis based on a hybridized CNN and an improved optimization algorithm, along with transfer learning, to help radiologists detect abnormalities efficiently. The marine predators algorithm (MPA) is the optimization algorithm we used, and we improve it using the opposition-based learning strategy to cope with the implied weaknesses of the original MPA. The improved marine predators algorithm (IMPA) is used to find the best values for the hyperparameters of the CNN architecture. The proposed method uses a pretrained CNN model called ResNet50 (residual network). This model is hybridized with the IMPA algorithm, resulting in an architecture called IMPA-ResNet50. Our evaluation is performed on two mammographic datasets, the mammographic image analysis society (MIAS) and curated breast imaging subset of DDSM (CBIS-DDSM) datasets. The proposed model was compared with other state-of-the-art approaches. The obtained results showed that the proposed model outperforms the compared state-of-the-art approaches, which are beneficial to classification performance, achieving 98.32% accuracy, 98.56% sensitivity, and 98.68% specificity on the CBIS-DDSM dataset and 98.88% accuracy, 97.61% sensitivity, and 98.40% specificity on the MIAS dataset. To evaluate the performance of IMPA in finding the optimal values for the hyperparameters of ResNet50 architecture, it compared to four other optimization algorithms including gravitational search algorithm (GSA), Harris hawks optimization (HHO), whale optimization algorithm (WOA), and the original MPA algorithm. The counterparts algorithms are also hybrid with the ResNet50 architecture produce models named GSA-ResNet50, HHO-ResNet50, WOA-ResNet50, and MPA-ResNet50, respectively. The results indicated that the proposed IMPA-ResNet50 is achieved a better performance than other counterparts.

Deep learning and radiomics to predict the mitotic index of gastrointestinal stromal tumors based on multiparametric MRI

“…Compared with previous models, the residual network is easier to optimize and can derive accuracy from a significantly increased depth (43). Many previous studies have used this network to classify tumors and achieved good results (44)(45)(46). The transfer learning method was adopted, and a fully connected layer was added to the hybrid model.…”

Self-adaptive moth flame optimizer combined with crossover operator and Fibonacci search strategy for COVID-19 CT image segmentation

“…In another study, Houssein et al (2022) developed a new improved equilibrium optimizer (I-EO) with the help of dimension learning hunting (DLH) technique. The efficiency of the suggested I-EO was tested by IEEE CEC 2020 test problems and COVID -19 CT image segmentation problem.…”

Breast Cancer Classification from Mammogram Images Using Extreme Learning Machine-Based DenseNet121 Model

“…The output of the fully connected layer is passed through the softmax layer to classifiy the images into cancerous and noncancerous. [15] DDSM IMPA-ResNet50 98.32% Nawaz et al, [16] BreakHis dataset DenseNet 95.4% Khan et al, [17] CBIS-DDSM ResNet50 88% Hameed et al, [18] MIFLUDAN project Xception 97.33% Joseph et al, [19] BreakHis dataset DNN 96.84% Alkassar et al, [20] BreakHis DenseNet and Xception 99% Our proposed method Kaggle dataset DenseNet121+ELM 99.47%…”