A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping
Background: Breast cancer in different molecular subtypes, which is determined by the overexpression rates of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), progesterone receptor (PR), and Ki67, exhibit distinct symptom characteristics and sensitivity to different treatment...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-04-01
|
| Series: | Neoplasia: An International Journal for Oncology Research |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1476558624000137 |
| _version_ | 1797253965442711552 |
|---|---|
| author | Lin Fan Jiahe Liu Baoyang Ju Doudou Lou Yushen Tian |
| author_facet | Lin Fan Jiahe Liu Baoyang Ju Doudou Lou Yushen Tian |
| author_sort | Lin Fan |
| collection | DOAJ |
| description | Background: Breast cancer in different molecular subtypes, which is determined by the overexpression rates of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), progesterone receptor (PR), and Ki67, exhibit distinct symptom characteristics and sensitivity to different treatment. The immunohistochemical method, one of the most common detecting tools for tumour markers, is heavily relied on artificial judgment and in clinical practice, with an inherent limitation in interpreting stability and operating efficiency. Here, a holistic intelligent breast tumour diagnosis system has been developed for tumour-markeromic analysis, combining the automatic interpretation and clinical suggestion. Methods: The holistic intelligent breast tumour diagnosis system included two main modules. The interpreting modules were constructed based on convolutional neural network, for comprehensively extracting and analyzing the multi-features of immunostaining. Referring to the clinical classification criteria, the interpreting results were encoded in a low-dimensional feature representation in the subtyping module, to efficiently output a holistic detecting result of the critical tumour-markeromic with diagnosis suggestions on molecular subtypes. Results: The overexpression rates of HER2, ER, PR, and Ki67, as well as an effective determination of molecular subtypes were successfully obtained by this diagnosis system, with an average sensitivity of 97.6 % and an average specificity of 96.1 %, among those, the sensitivity and specificity for interpreting HER2 were up to 99.8 % and 96.9 %. Conclusion: The holistic intelligent breast tumour diagnosis system shows improved performance in the interpretation of immunohistochemical images over pathologist-level, which can be expected to overcome the limitations of conventional manual interpretation in efficiency, precision, and repeatability. |
| first_indexed | 2024-03-07T20:04:54Z |
| format | Article |
| id | doaj.art-429482d69a6a422f852156b7d4bfe31e |
| institution | Directory Open Access Journal |
| issn | 1476-5586 |
| language | English |
| last_indexed | 2024-04-24T21:42:26Z |
| publishDate | 2024-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Neoplasia: An International Journal for Oncology Research |
| spelling | doaj.art-429482d69a6a422f852156b7d4bfe31e2024-03-21T05:35:56ZengElsevierNeoplasia: An International Journal for Oncology Research1476-55862024-04-0150100976A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtypingLin Fan0Jiahe Liu1Baoyang Ju2Doudou Lou3Yushen Tian4School of Integrated Circuit Science and Engineering (Industry-Education Integration School), Nanjing University of Posts and Telecommunications, Nanjing 210023, PR China; State Key Laboratory of Bioelectronics, Jiangsu Key Laboratory for Biomaterials and Devices, School of Biological Science and Medical Engineering & Collaborative Innovation Center of Suzhou Nano Science and Technology, Southeast University, Nanjing 210096, PR China; Medical School of Nanjing University, Nanjing 210093, PR China; Corresponding authors.School of Integrated Circuit Science and Engineering (Industry-Education Integration School), Nanjing University of Posts and Telecommunications, Nanjing 210023, PR ChinaSchool of Integrated Circuit Science and Engineering (Industry-Education Integration School), Nanjing University of Posts and Telecommunications, Nanjing 210023, PR ChinaNanjing Institute for Food and Drug Control, Nanjing, Jiangsu 211198, PR ChinaSchool of Electrical Engineering, Shenyang University of Technology, Shenyang, Liaoning 110870, PR China; Corresponding authors.Background: Breast cancer in different molecular subtypes, which is determined by the overexpression rates of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), progesterone receptor (PR), and Ki67, exhibit distinct symptom characteristics and sensitivity to different treatment. The immunohistochemical method, one of the most common detecting tools for tumour markers, is heavily relied on artificial judgment and in clinical practice, with an inherent limitation in interpreting stability and operating efficiency. Here, a holistic intelligent breast tumour diagnosis system has been developed for tumour-markeromic analysis, combining the automatic interpretation and clinical suggestion. Methods: The holistic intelligent breast tumour diagnosis system included two main modules. The interpreting modules were constructed based on convolutional neural network, for comprehensively extracting and analyzing the multi-features of immunostaining. Referring to the clinical classification criteria, the interpreting results were encoded in a low-dimensional feature representation in the subtyping module, to efficiently output a holistic detecting result of the critical tumour-markeromic with diagnosis suggestions on molecular subtypes. Results: The overexpression rates of HER2, ER, PR, and Ki67, as well as an effective determination of molecular subtypes were successfully obtained by this diagnosis system, with an average sensitivity of 97.6 % and an average specificity of 96.1 %, among those, the sensitivity and specificity for interpreting HER2 were up to 99.8 % and 96.9 %. Conclusion: The holistic intelligent breast tumour diagnosis system shows improved performance in the interpretation of immunohistochemical images over pathologist-level, which can be expected to overcome the limitations of conventional manual interpretation in efficiency, precision, and repeatability.http://www.sciencedirect.com/science/article/pii/S1476558624000137Breast cancerMolecular subtypeImmunohistochemical microimagingConvolutional neural networkIntelligent diagnosis |
| spellingShingle | Lin Fan Jiahe Liu Baoyang Ju Doudou Lou Yushen Tian A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping Neoplasia: An International Journal for Oncology Research Breast cancer Molecular subtype Immunohistochemical microimaging Convolutional neural network Intelligent diagnosis |
| title | A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping |
| title_full | A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping |
| title_fullStr | A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping |
| title_full_unstemmed | A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping |
| title_short | A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping |
| title_sort | deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping |
| topic | Breast cancer Molecular subtype Immunohistochemical microimaging Convolutional neural network Intelligent diagnosis |
| url | http://www.sciencedirect.com/science/article/pii/S1476558624000137 |
| work_keys_str_mv | AT linfan adeeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT jiaheliu adeeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT baoyangju adeeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT doudoulou adeeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT yushentian adeeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT linfan deeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT jiaheliu deeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT baoyangju deeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT doudoulou deeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping AT yushentian deeplearningbasedholisticdiagnosissystemforimmunohistochemistryinterpretationandmolecularsubtyping |