Chasing a Better Decision Margin for Discriminative Histopathological Breast Cancer Image Classification
When considering a large dataset of histopathologic breast images captured at various magnification levels, the process of distinguishing between benign and malignant cancer from these images can be time-intensive. The automation of histopathological breast cancer image classification holds signific...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-10-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/12/20/4356 |
| _version_ | 1797573984710033408 |
|---|---|
| author | Pendar Alirezazadeh Fadi Dornaika Abdelmalik Moujahid |
| author_facet | Pendar Alirezazadeh Fadi Dornaika Abdelmalik Moujahid |
| author_sort | Pendar Alirezazadeh |
| collection | DOAJ |
| description | When considering a large dataset of histopathologic breast images captured at various magnification levels, the process of distinguishing between benign and malignant cancer from these images can be time-intensive. The automation of histopathological breast cancer image classification holds significant promise for expediting pathology diagnoses and reducing the analysis time. Convolutional neural networks (CNNs) have recently gained traction for their ability to more accurately classify histopathological breast cancer images. CNNs excel at extracting distinctive features that emphasize semantic information. However, traditional CNNs employing the softmax loss function often struggle to achieve the necessary discriminatory power for this task. To address this challenge, a set of angular margin-based softmax loss functions have emerged, including angular softmax (A-Softmax), large margin cosine loss (CosFace), and additive angular margin (ArcFace), each sharing a common objective: maximizing inter-class variation while minimizing intra-class variation. This study delves into these three loss functions and their potential to extract distinguishing features while expanding the decision boundary between classes. Rigorous experimentation on a well-established histopathological breast cancer image dataset, BreakHis, has been conducted. As per the results, it is evident that CosFace focuses on augmenting the differences between classes, while A-Softmax and ArcFace tend to emphasize augmenting within-class variations. These observations underscore the efficacy of margin penalties on angular softmax losses in enhancing feature discrimination within the embedding space. These loss functions consistently outperform softmax-based techniques, either by widening the gaps among classes or enhancing the compactness of individual classes. |
| first_indexed | 2024-03-10T21:17:51Z |
| format | Article |
| id | doaj.art-b2b4fda4cef84109a6cdd4cc6c0083a4 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T21:17:51Z |
| publishDate | 2023-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-b2b4fda4cef84109a6cdd4cc6c0083a42023-11-19T16:20:33ZengMDPI AGElectronics2079-92922023-10-011220435610.3390/electronics12204356Chasing a Better Decision Margin for Discriminative Histopathological Breast Cancer Image ClassificationPendar Alirezazadeh0Fadi Dornaika1Abdelmalik Moujahid2Department of Informatics, University of the Basque Country, 20008 Donostia-San Sebastian, SpainDepartment of Informatics, University of the Basque Country, 20008 Donostia-San Sebastian, SpainHigh School of Engineering and Technology, Universidad Internacional de la Rioja, Avenida de la Paz 137, 26006 Logroño, SpainWhen considering a large dataset of histopathologic breast images captured at various magnification levels, the process of distinguishing between benign and malignant cancer from these images can be time-intensive. The automation of histopathological breast cancer image classification holds significant promise for expediting pathology diagnoses and reducing the analysis time. Convolutional neural networks (CNNs) have recently gained traction for their ability to more accurately classify histopathological breast cancer images. CNNs excel at extracting distinctive features that emphasize semantic information. However, traditional CNNs employing the softmax loss function often struggle to achieve the necessary discriminatory power for this task. To address this challenge, a set of angular margin-based softmax loss functions have emerged, including angular softmax (A-Softmax), large margin cosine loss (CosFace), and additive angular margin (ArcFace), each sharing a common objective: maximizing inter-class variation while minimizing intra-class variation. This study delves into these three loss functions and their potential to extract distinguishing features while expanding the decision boundary between classes. Rigorous experimentation on a well-established histopathological breast cancer image dataset, BreakHis, has been conducted. As per the results, it is evident that CosFace focuses on augmenting the differences between classes, while A-Softmax and ArcFace tend to emphasize augmenting within-class variations. These observations underscore the efficacy of margin penalties on angular softmax losses in enhancing feature discrimination within the embedding space. These loss functions consistently outperform softmax-based techniques, either by widening the gaps among classes or enhancing the compactness of individual classes.https://www.mdpi.com/2079-9292/12/20/4356BreakHisbreast cancer image classificationdiscriminative deep embeddingmargin penalties on angular softmax lossescompactness and separabilitydeep learning |
| spellingShingle | Pendar Alirezazadeh Fadi Dornaika Abdelmalik Moujahid Chasing a Better Decision Margin for Discriminative Histopathological Breast Cancer Image Classification Electronics BreakHis breast cancer image classification discriminative deep embedding margin penalties on angular softmax losses compactness and separability deep learning |
| title | Chasing a Better Decision Margin for Discriminative Histopathological Breast Cancer Image Classification |
| title_full | Chasing a Better Decision Margin for Discriminative Histopathological Breast Cancer Image Classification |
| title_fullStr | Chasing a Better Decision Margin for Discriminative Histopathological Breast Cancer Image Classification |
| title_full_unstemmed | Chasing a Better Decision Margin for Discriminative Histopathological Breast Cancer Image Classification |
| title_short | Chasing a Better Decision Margin for Discriminative Histopathological Breast Cancer Image Classification |
| title_sort | chasing a better decision margin for discriminative histopathological breast cancer image classification |
| topic | BreakHis breast cancer image classification discriminative deep embedding margin penalties on angular softmax losses compactness and separability deep learning |
| url | https://www.mdpi.com/2079-9292/12/20/4356 |
| work_keys_str_mv | AT pendaralirezazadeh chasingabetterdecisionmarginfordiscriminativehistopathologicalbreastcancerimageclassification AT fadidornaika chasingabetterdecisionmarginfordiscriminativehistopathologicalbreastcancerimageclassification AT abdelmalikmoujahid chasingabetterdecisionmarginfordiscriminativehistopathologicalbreastcancerimageclassification |