Evolving convolutional neural network parameters through the genetic algorithm for the breast cancer classification problem

Abstract

Breast cancer is the most frequently diagnosed cancer and the leading cause of cancer mortality in women around the world. However, it can be controlled effectively by early diagnosis, followed by effective treatment. Clinical specialists take the advantages of computer-aided diagnosis (CAD) systems to make their diagnosis as accurate as possible. Deep learning techniques, such as the convolutional neural network (CNN), due to their classification capabilities on learned feature methods and ability of working with complex images, have been widely adopted in CAD systems. The parameters of the network, including the weights of the convolution filters and the weights of the fully connected layers, play a crucial role in the classification accuracy of any CNN model. The back-propagation technique is the most frequently used approach for training the CNN. However, this technique has some disadvantages, such as getting stuck in local minima. In this study, we propose to optimize the weights of the CNN using the genetic algorithm (GA). The work consists of designing a CNN model to facilitate the classification process, training the model using three different optimizers (mini-batch gradient descent, Adam, and GA), and evaluating the model through various experiments on the BreakHis dataset. We show that the CNN model trained through the GA performs as well as the Adam optimizer with a classification accuracy of 85%.

A Deep Learning Computer-Aided Diagnosis Approach for Breast Cancer

“…They conducted the various experiments in a similar manner to our study; however, they did not target the possibility of developing more than one model in the experiments. Our top performing model has achieved higher accuracy values than that achieved by Davoudi et al [ 10 ] in their conducted experiments. The efficacy of their model was tested only in magnification-independent binary classification.…”

“…Davoudi et al [ 10 ] design and implement a CNN for the detection of binary classes of BreakHis dataset independent of the magnification factors. The main contribution of their study is to try to optimize the weights of the CNN using genetic algorithms (GAs) instead of the normal optimizers.…”

Nonlinear modeling and prediction of forklift acoustic annoyance based on the improved neural networks

“…The weights and thresholds of the neural network structure can be optimized using the global search ability of GA. 12,16,23 The detailed implementation steps of GANN in Figure 5 are as follows: (1) determine the initial structure of neural network; (2) set relevant parameters and chromosome code; (3) based on fitness values, perform genetic operations including selection, crossover, and mutation; (4) obtain the optimal weights and thresholds, and then predict annoyance.…”

“…To solve nonlinear modeling problems, artificial neural network (ANN) has satisfied self-adaptation and fault tolerance with widespread applications in prediction, such as temperature estimation, 15 feature classification, 16 material mechanics, 17 and especially vehicle noise evaluation. 8,12,18 For example, Huang et al 19 built a novel intelligent acoustic quality evaluation model for pure electric vehicles using a deep ANN algorithm; Wang et al 20 used an ANN model based on wavelet theory to predict subjective vehicle noise quality and proved its effectiveness in nonlinear mapping.…”

A Two-Stage Model for Breast Cancer Detection and Classification from Mammograms

“…It is made up of the input, hidden and output layers. The input layer is the image; hidden layer consists of the convolution, pooling, flattening and fully connected layers (Davoudi & Thulasiraman, 2021). The convolution layer converts the input image to a feature map by performing a linear operation known as convolution.…”