Diagnostics, Vol. 13, Pages 103: Efficient Breast Cancer Classification Network with Dual Squeeze and Excitation in Histopathological Images

Conceptualisation, M.M.K.S. and F.A.; Data curation, M.M.K.S. and F.A.; Formal analysis, M.M.K.S. and F.A.; Investigation, M.M.K.S., F.A. and V.K.S.; Methodology, M.M.K.S., F.A. and V.K.S.; Literature review, M.A., V.K.S. and M.M.K.S.; Project administration, M.M.K.S., F.A. and V.K.S.; Resources, M.M.K.S. and E.E.; Software, M.M.K.S., F.A. and V.K.S.; Supervision, M.M.K.S., F.A., V.K.S. and E.E.; Validation, M.M.K.S. and F.A.; Visualisation, M.M.K.S., F.A. and V.K.S.; Writing—original draft, M.M.K.S., F.A., M.A., V.K.S. and R.Y.; Writing—review and editing, M.M.K.S., F.A., M.A., V.K.S. and R.Y. All authors have read and agreed to the published version of the manuscript.

Figure 1. Illustration of H&E-stained breast tumour patch examples extracted from the different magnifications of whole slide image (WSI). Note that (a) benign and (b) malignant correspond to the BN and ML, respectively. It contains four histological subtypes of benign breast tumours: adenosis (A), fibroadenoma (F), phyllodes tumour (PT), and tubular adenona (TA); and four malignant tumours (breast cancer): carcinoma (DC), lobular carcinoma (LC), mucinous carcinoma (MC), and papillary carcinoma (PC). These sub-categorized breast tumours are distinct in their visual, textural patterns and color features.

Figure 1. Illustration of H&E-stained breast tumour patch examples extracted from the different magnifications of whole slide image (WSI). Note that (a) benign and (b) malignant correspond to the BN and ML, respectively. It contains four histological subtypes of benign breast tumours: adenosis (A), fibroadenoma (F), phyllodes tumour (PT), and tubular adenona (TA); and four malignant tumours (breast cancer): carcinoma (DC), lobular carcinoma (LC), mucinous carcinoma (MC), and papillary carcinoma (PC). These sub-categorized breast tumours are distinct in their visual, textural patterns and color features.

Diagnostics 13 00103 g001

Figure 2. Overview of the proposed classification model.

Figure 2. Overview of the proposed classification model.

Diagnostics 13 00103 g002

Figure 3. Illustration of the binary class classification performance achieved by the proposed model compared to the EfficientNetV2 network under multiple magnification settings.

Figure 3. Illustration of the binary class classification performance achieved by the proposed model compared to the EfficientNetV2 network under multiple magnification settings.

Diagnostics 13 00103 g003

Figure 4. Illustration of the binary class confusion matrix obtained through the proposed model on the test set. Note that (ad) refer to the 40×, 100×, 200×, and 400× magnifications, respectively.

Figure 4. Illustration of the binary class confusion matrix obtained through the proposed model on the test set. Note that (ad) refer to the 40×, 100×, 200×, and 400× magnifications, respectively.

Diagnostics 13 00103 g004

Figure 5. Illustration of the ROC curve obtained through the proposed model on the test set. Note that (ad) refer to the 40×, 100×, 200×, and 400× magnifications, respectively.

Figure 5. Illustration of the ROC curve obtained through the proposed model on the test set. Note that (ad) refer to the 40×, 100×, 200×, and 400× magnifications, respectively.

Diagnostics 13 00103 g005aDiagnostics 13 00103 g005b

Figure 6. Illustration of the proposed model multi-class classification performance improvement respective to EfficientNetV2.

Figure 6. Illustration of the proposed model multi-class classification performance improvement respective to EfficientNetV2.

Diagnostics 13 00103 g006

Figure 7. Illustration of multi-class confusion matrix obtained through the proposed model on the test set. Note that (ad) refer to the 40×, 100×, 200×, and 400× magnifications, respectively.

Figure 7. Illustration of multi-class confusion matrix obtained through the proposed model on the test set. Note that (ad) refer to the 40×, 100×, 200×, and 400× magnifications, respectively.

Diagnostics 13 00103 g007

Figure 8. Illustration of the ROC curve of eight classes obtained through the proposed model on the test set. Note that (ad) refer to the 40×, 100×, 200×, and 400× magnifications, respectively.

Figure 8. Illustration of the ROC curve of eight classes obtained through the proposed model on the test set. Note that (ad) refer to the 40×, 100×, 200×, and 400× magnifications, respectively.

Diagnostics 13 00103 g008

Figure 9. GradCam visualisation of the proposed model for H&E-stained breast tumour patch examples of benign (a,b) and malignant (c,d) classes.

Figure 9. GradCam visualisation of the proposed model for H&E-stained breast tumour patch examples of benign (a,b) and malignant (c,d) classes.

Diagnostics 13 00103 g009

Table 1. A breakdown of the publicly available breast cancer classification BreakHis dataset. This dataset contains multiple image magnifications defined for benign and malignant categories.

Table 1. A breakdown of the publicly available breast cancer classification BreakHis dataset. This dataset contains multiple image magnifications defined for benign and malignant categories.

DatasetMagnificationBenignMalignantTotalBreakHis [12]40×65213701995100×64414372081200×62313902031400×58812321820Total of Images248054297909

Table 2. Classification performance of the proposed model compared with three state-of-the-art diagnoses of breast cancer benign and malignant types in histopathology images. Note that ‘PR’ stands for ‘Precision’, ‘RE’ stands for ‘Recall’, and ‘F1’ represents the ‘F1-score’. The best significant result is highlighted in bold.

Table 2. Classification performance of the proposed model compared with three state-of-the-art diagnoses of breast cancer benign and malignant types in histopathology images. Note that ‘PR’ stands for ‘Precision’, ‘RE’ stands for ‘Recall’, and ‘F1’ represents the ‘F1-score’. The best significant result is highlighted in bold.

ModelsPerformance MetricsMagnifications40×100×200×400×PRREF1PRREF1PRREF1PRREF1ResNet1010.87610.83950.85430.83890.8360.83740.90520.91210.90850.84670.8370.8415InceptionResnetV20.97320.98110.9770.95610.96730.96140.98080.98490.98280.93290.9290.9309EfficientNetV20.98470.98050.98250.94020.93650.93830.99640.99210.99420.95140.96290.9568Proposed (EfficientNetV2 + DSE)0.98580.96830.97640.97400.97000.97200.99650.99210.99430.9960.99150.9937

Table 3. Classification performance of the proposed model compared with three state-of-the-art models to diagnose eight types of benign and malignant breast cancer in histopathology images. The most significant results are highlighted in bold.

Table 3. Classification performance of the proposed model compared with three state-of-the-art models to diagnose eight types of benign and malignant breast cancer in histopathology images. The most significant results are highlighted in bold.

ModelsPerformance MetricsMagnifications40×100×200×400×PRREF1PRREF1PRREF1PRREF1ResNet1010.77430.78080.76980.74860.74630.73880.72550.68950.70520.79370.76650.7712InceptionResnetV20.86840.83360.84850.83850.83240.83190.77670.75820.76090.78350.69560.7229EfficientNetV20.89180.87270.8810.81760.84370.82910.79920.76910.78090.79370.76650.7712Proposed (EfficientNetV2 + DSE)0.96160.95820.95990.9440.94870.94590.89650.81330.84750.92750.92020.919

留言 (0)

沒有登入
gif