Classification of breast cancer using a manta-ray foraging optimized transfer learning framework

Abstract

Due to its high prevalence and wide dissemination, breast cancer is a particularly dangerous disease. Breast cancer survival chances can be improved by early detection and diagnosis. For medical image analyzers, diagnosing is tough, time-consuming, routine, and repetitive. Medical image analysis could be a useful method for detecting such a disease. Recently, artificial intelligence technology has been utilized to help radiologists identify breast cancer more rapidly and reliably. Convolutional neural networks, among other technologies, are promising medical image recognition and classification tools. This study proposes a framework for automatic and reliable breast cancer classification based on histological and ultrasound data. The system is built on CNN and employs transfer learning technology and metaheuristic optimization. The Manta Ray Foraging Optimization (MRFO) approach is deployed to improve the framework’s adaptability. Using the Breast Cancer Dataset (two classes) and the Breast Ultrasound Dataset (three-classes), eight modern pre-trained CNN architectures are examined to apply the transfer learning technique. The framework uses MRFO to improve the performance of CNN architectures by optimizing their hyperparameters. Extensive experiments have recorded performance parameters, including accuracy, AUC, precision, F1-score, sensitivity, dice, recall, IoU, and cosine similarity. The proposed framework scored 97.73% on histopathological data and 99.01% on ultrasound data in terms of accuracy. The experimental results show that the proposed framework is superior to other state-of-the-art approaches in the literature review.

A concentrated machine learning-based classification system for age-related macular degeneration (AMD) diagnosis using fundus images

“…In the field of image processing and texture analysis, feature extraction plays a crucial role in quantifying the characteristics of an image. The study worked on extracting both first-order and second-order features using GLCM (Gray-Level Co-occurrence Matrix) and GLRLM (Gray-Level Run-Length Matrix) methods for each extracted contour resulted from the pre-processing step 41 .…”

SNSVM: SqueezeNet-Guided SVM for Breast Cancer Diagnosis

“…SNSVM is compared with state-of-the-art models, which include WEE + LRC [27], SVM [28], GoogleNet [29], CNNDL [30], MRFOTL [31], and SWAE [32], as shown in Table 4. A comparison of the results reveals that in each of the indicators, the proposed SNSVM gets the best results, but there is some fluctuation in the multiple validations.…”

“…The proposed customdesigned CNN network [30] achieved impressive accuracy in distinguishing between metastatic and non-metastatic cells. The proposed framework by Baghdadi et al [31] for breast cancer classification uses CNN and transfer learning. Eight classical pre-trained models were compared to arrive at the optimal choice.…”

Integrating Artificial Intelligence and Big Data into Smart Healthcare Systems: A Comprehensive Review of Current Practices and Future Directions

“…Authors in [38] proposed an innovative automated, and dependable breast cancer classification framework. The system utilized CNN and incorporated TL technology along with Metaheuristic optimization techniques.…”

“…The framework under consideration achieved an accuracy of 97.73% on histopathological data and 99.01% on ultrasound data, as reported. However, the proposed systems in [37][38] are limited to a single modality and can be extended to other modalities and cough sounds.…”