article · PLoS ONE
An optimised computer-aided diagnosis system has been developed to classify medical images for breast cancer detection. The approach incorporates pre-trained convolutional neural networks, specifically DenseNet-121 and VGG-16, to extract image features. Bidirectional long short-term memory layers are employed to capture temporal features, whilst support vector machines and random forest algorithms perform the final classification. To enhance diagnostic precision, hyperparameters within the pre-trained networks are tuned using a modified grey wolf optimisation technique. Testing on benchmark mammographic collections demonstrated high performance. When evaluated on the Mammographic Image Analysis Society dataset, the VGG-16 configuration achieved 99.86 per cent accuracy and an area under the curve of 1.0. Performance remained robust on the INbreast dataset, reaching 99.4 per cent accuracy along with high levels of sensitivity and specificity.
Early detection of breast tumours is essential for effective treatment and improved patient outcomes. Computer-aided diagnosis systems assist radiologists in reading complex mammograms quickly and accurately. Demonstrating high classification accuracy and sensitivity across benchmark datasets highlights the potential for combining deep learning feature extractors with metaheuristic optimisation to strengthen automated medical imaging tools.
The model could be integrated into computer-aided diagnosis software to assist radiologists during breast cancer screenings. The primary end users are hospital radiology units, diagnostic imaging centres, and medical software vendors. Based on testing on two standard retrospective datasets, the technology is at an applied and tested research stage, requiring clinical trials and integration into medical imaging pipelines before reaching market.
AI-generated from the published abstract. Always read the original work before citing.
Medical image classification (IC) is a method for categorizing images according to the appropriate pathological stage. It is a crucial stage in computer-aided diagnosis (CAD) systems, which were created to help radiologists with reading and analyzing medical images as well as with the early detection of tumors and other disorders. The use of convolutional neural network (CNN) models in the medical industry has recently increased, and they achieve great results at IC, particularly in terms of high performance and robustness. The proposed method uses pre-trained models such as Dense Convolutional Network (DenseNet)-121 and Visual Geometry Group (VGG)-16 as feature extractor networks, bidirectional long short-term memory (BiLSTM) layers for temporal feature extraction, and the Support Vector Machine (SVM) and Random Forest (RF) algorithms to perform classification. For improved performance, the selected pre-trained CNN hyperparameters have been optimized using a modified grey wolf optimization method. The experimental analysis for the presented model on the Mammographic Image Analysis Society (MIAS) dataset shows that the VGG16 model is powerful for BC classification with overall accuracy, sensitivity, specificity, precision, and area under the ROC curve (AUC) of 99.86%, 99.9%, 99.7%, 97.1%, and 1.0, respectively, on the MIAS dataset and 99.4%, 99.03%, 99.2%, 97.4%, and 1.0, respectively, on the INbreast dataset.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1371/journal.pone.0304868
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
