Table 2 This table gives a summary of recent work that has been executed in breast cancer detection using machine learning and deep learning algorithms as discussed in Sect. 2.2
References | Feature extraction | Data | ML/DL | Acc, AUC or ROC (%) |
|---|---|---|---|---|
[49] 2018 | Watershed Segmentation | histopathology images | CNN | 98 |
[49] 2018 | Label encoder, normalization | Wisconsin breast cancer [42] | CNN | 99.6 |
[50] 2018 | Standard scaler | Wisconsin breast cancer [42] | GRU-SVM, Linear Regression, | Â |
| Â | Â | Â | MLP, Nearest Neighbor, | Â |
| Â | Â | Â | Softmax Regression, SVM | 99.0 |
[45] 2018 | Inception V3 | thermogram images [41] | LinearSVC, SVM | 100 |
[51] 2019 | – | ResNet50, VGG16 | 65-97 | |
[48] 2020 | Histogram-sigmoid fuzzy clustering | histopathology images | Deep Neural Network | 97 |
[44] 2019 | filters | whole slide images | CNN | 88 |
[46] 2020 | Hu moment, color histogram, | Â | Â | Â |
| Â | and Haralick textures, ResNet50, | Â | Â | Â |
| Â | VGG16 and VGG19 | BreakHis [54] | RF, SVM, LDA, | Â |
| Â | Â | Â | ResNet50, VGG16, VGG19 | 91.2-93.9 |
[55] 2021 | – | IDC patch images [56] | CNNs,LR,SVM, KNN | 87 |
[47] 2022 | AWS, DenseNet-169 | mammograms [43] | MLP | 93.8 |
[57] 2022 | AlexNet CNN | ultrasound images and histopathological images | Fully connected layer | 96.7-100 |
[58] 2022 | AlexNet CNN | MRI scans [59] | Fully connected layer | 98.1-98.44 |
[60] 2022 | – | Wisconsin Breast Cancer Diagnostic data | deep extreme gradient descent optimization | 98.73 |