Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy
Breast cancer is the leading cause of cancer-related deaths among women worldwide, and early detection and treatment has been shown to significantly reduce fatality rates from severe illness. Moreover, determination of the human epidermal growth factor receptor-2 (HER2) gene amplification by Fluores...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/15/15/3991 |
| _version_ | 1797586927599222784 |
|---|---|
| author | Ching-Wei Wang Kai-Lin Chu Hikam Muzakky Yi-Jia Lin Tai-Kuang Chao |
| author_facet | Ching-Wei Wang Kai-Lin Chu Hikam Muzakky Yi-Jia Lin Tai-Kuang Chao |
| author_sort | Ching-Wei Wang |
| collection | DOAJ |
| description | Breast cancer is the leading cause of cancer-related deaths among women worldwide, and early detection and treatment has been shown to significantly reduce fatality rates from severe illness. Moreover, determination of the human epidermal growth factor receptor-2 (HER2) gene amplification by Fluorescence in situ hybridization (FISH) and Dual in situ hybridization (DISH) is critical for the selection of appropriate breast cancer patients for HER2-targeted therapy. However, visual examination of microscopy is time-consuming, subjective and poorly reproducible due to high inter-observer variability among pathologists and cytopathologists. The lack of consistency in identifying carcinoma-like nuclei has led to divergences in the calculation of sensitivity and specificity. This manuscript introduces a highly efficient deep learning method with low computing cost. The experimental results demonstrate that the proposed framework achieves high precision and recall on three essential clinical applications, including breast cancer diagnosis and human epidermal receptor factor 2 (HER2) amplification detection on FISH and DISH slides for HER2 target therapy. Furthermore, the proposed method outperforms the majority of the benchmark methods in terms of IoU by a significant margin (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>p</mi><mo><</mo><mn>0.001</mn></mrow></semantics></math></inline-formula>) on three essential clinical applications. Importantly, run time analysis shows that the proposed method obtains excellent segmentation results with notably reduced time for Artificial intelligence (AI) training (16.93%), AI inference (17.25%) and memory usage (18.52%), making the proposed framework feasible for practical clinical usage. |
| first_indexed | 2024-03-11T00:30:06Z |
| format | Article |
| id | doaj.art-247aa6953f7f446d926ff11b1b69b94f |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-11T00:30:06Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-247aa6953f7f446d926ff11b1b69b94f2023-11-18T22:44:10ZengMDPI AGCancers2072-66942023-08-011515399110.3390/cancers15153991Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target TherapyChing-Wei Wang0Kai-Lin Chu1Hikam Muzakky2Yi-Jia Lin3Tai-Kuang Chao4Graduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology, Taipei 106335, TaiwanGraduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology, Taipei 106335, TaiwanGraduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology, Taipei 106335, TaiwanDepartment of Pathology, Tri-Service General Hospital, Taipei 11490, TaiwanDepartment of Pathology, Tri-Service General Hospital, Taipei 11490, TaiwanBreast cancer is the leading cause of cancer-related deaths among women worldwide, and early detection and treatment has been shown to significantly reduce fatality rates from severe illness. Moreover, determination of the human epidermal growth factor receptor-2 (HER2) gene amplification by Fluorescence in situ hybridization (FISH) and Dual in situ hybridization (DISH) is critical for the selection of appropriate breast cancer patients for HER2-targeted therapy. However, visual examination of microscopy is time-consuming, subjective and poorly reproducible due to high inter-observer variability among pathologists and cytopathologists. The lack of consistency in identifying carcinoma-like nuclei has led to divergences in the calculation of sensitivity and specificity. This manuscript introduces a highly efficient deep learning method with low computing cost. The experimental results demonstrate that the proposed framework achieves high precision and recall on three essential clinical applications, including breast cancer diagnosis and human epidermal receptor factor 2 (HER2) amplification detection on FISH and DISH slides for HER2 target therapy. Furthermore, the proposed method outperforms the majority of the benchmark methods in terms of IoU by a significant margin (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>p</mi><mo><</mo><mn>0.001</mn></mrow></semantics></math></inline-formula>) on three essential clinical applications. Importantly, run time analysis shows that the proposed method obtains excellent segmentation results with notably reduced time for Artificial intelligence (AI) training (16.93%), AI inference (17.25%) and memory usage (18.52%), making the proposed framework feasible for practical clinical usage.https://www.mdpi.com/2072-6694/15/15/3991breast cancer metastaseshistopathologybreast cancer target therapydilated soft label deep learning |
| spellingShingle | Ching-Wei Wang Kai-Lin Chu Hikam Muzakky Yi-Jia Lin Tai-Kuang Chao Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy Cancers breast cancer metastases histopathology breast cancer target therapy dilated soft label deep learning |
| title | Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy |
| title_full | Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy |
| title_fullStr | Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy |
| title_full_unstemmed | Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy |
| title_short | Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy |
| title_sort | efficient convolution network to assist breast cancer diagnosis and target therapy |
| topic | breast cancer metastases histopathology breast cancer target therapy dilated soft label deep learning |
| url | https://www.mdpi.com/2072-6694/15/15/3991 |
| work_keys_str_mv | AT chingweiwang efficientconvolutionnetworktoassistbreastcancerdiagnosisandtargettherapy AT kailinchu efficientconvolutionnetworktoassistbreastcancerdiagnosisandtargettherapy AT hikammuzakky efficientconvolutionnetworktoassistbreastcancerdiagnosisandtargettherapy AT yijialin efficientconvolutionnetworktoassistbreastcancerdiagnosisandtargettherapy AT taikuangchao efficientconvolutionnetworktoassistbreastcancerdiagnosisandtargettherapy |