A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis
Abstract The application of deep learning algorithms for medical diagnosis in the real world faces challenges with transparency and interpretability. The labeling of large-scale samples leads to costly investment in developing deep learning algorithms. The application of human prior knowledge is an...
Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/03dcc04288584655bc8b9a6fe5671d8e |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:03dcc04288584655bc8b9a6fe5671d8e |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:03dcc04288584655bc8b9a6fe5671d8e2021-12-02T13:34:33ZA hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis10.1038/s41746-021-00417-42398-6352https://doaj.org/article/03dcc04288584655bc8b9a6fe5671d8e2021-03-01T00:00:00Zhttps://doi.org/10.1038/s41746-021-00417-4https://doaj.org/toc/2398-6352Abstract The application of deep learning algorithms for medical diagnosis in the real world faces challenges with transparency and interpretability. The labeling of large-scale samples leads to costly investment in developing deep learning algorithms. The application of human prior knowledge is an effective way to solve these problems. Previously, we developed a deep learning system for glaucoma diagnosis based on a large number of samples that had high sensitivity and specificity. However, it is a black box and the specific analytic methods cannot be elucidated. Here, we establish a hierarchical deep learning system based on a small number of samples that comprehensively simulates the diagnostic thinking of human experts. This system can extract the anatomical characteristics of the fundus images, including the optic disc, optic cup, and appearance of the retinal nerve fiber layer to realize automatic diagnosis of glaucoma. In addition, this system is transparent and interpretable, and the intermediate process of prediction can be visualized. Applying this system to three validation datasets of fundus images, we demonstrate performance comparable to that of human experts in diagnosing glaucoma. Moreover, it markedly improves the diagnostic accuracy of ophthalmologists. This system may expedite the screening and diagnosis of glaucoma, resulting in improved clinical outcomes.Yongli XuMan HuHanruo LiuHao YangHuaizhou WangShuai LuTianwei LiangXiaoxing LiMai XuLiu LiHuiqi LiXin JiZhijun WangLi LiRobert N. WeinrebNingli WangNature PortfolioarticleComputer applications to medicine. Medical informaticsR858-859.7ENnpj Digital Medicine, Vol 4, Iss 1, Pp 1-11 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
Computer applications to medicine. Medical informatics R858-859.7 |
| spellingShingle |
Computer applications to medicine. Medical informatics R858-859.7 Yongli Xu Man Hu Hanruo Liu Hao Yang Huaizhou Wang Shuai Lu Tianwei Liang Xiaoxing Li Mai Xu Liu Li Huiqi Li Xin Ji Zhijun Wang Li Li Robert N. Weinreb Ningli Wang A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis |
| description |
Abstract The application of deep learning algorithms for medical diagnosis in the real world faces challenges with transparency and interpretability. The labeling of large-scale samples leads to costly investment in developing deep learning algorithms. The application of human prior knowledge is an effective way to solve these problems. Previously, we developed a deep learning system for glaucoma diagnosis based on a large number of samples that had high sensitivity and specificity. However, it is a black box and the specific analytic methods cannot be elucidated. Here, we establish a hierarchical deep learning system based on a small number of samples that comprehensively simulates the diagnostic thinking of human experts. This system can extract the anatomical characteristics of the fundus images, including the optic disc, optic cup, and appearance of the retinal nerve fiber layer to realize automatic diagnosis of glaucoma. In addition, this system is transparent and interpretable, and the intermediate process of prediction can be visualized. Applying this system to three validation datasets of fundus images, we demonstrate performance comparable to that of human experts in diagnosing glaucoma. Moreover, it markedly improves the diagnostic accuracy of ophthalmologists. This system may expedite the screening and diagnosis of glaucoma, resulting in improved clinical outcomes. |
| format |
article |
| author |
Yongli Xu Man Hu Hanruo Liu Hao Yang Huaizhou Wang Shuai Lu Tianwei Liang Xiaoxing Li Mai Xu Liu Li Huiqi Li Xin Ji Zhijun Wang Li Li Robert N. Weinreb Ningli Wang |
| author_facet |
Yongli Xu Man Hu Hanruo Liu Hao Yang Huaizhou Wang Shuai Lu Tianwei Liang Xiaoxing Li Mai Xu Liu Li Huiqi Li Xin Ji Zhijun Wang Li Li Robert N. Weinreb Ningli Wang |
| author_sort |
Yongli Xu |
| title |
A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis |
| title_short |
A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis |
| title_full |
A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis |
| title_fullStr |
A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis |
| title_full_unstemmed |
A hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis |
| title_sort |
hierarchical deep learning approach with transparency and interpretability based on small samples for glaucoma diagnosis |
| publisher |
Nature Portfolio |
| publishDate |
2021 |
| url |
https://doaj.org/article/03dcc04288584655bc8b9a6fe5671d8e |
| work_keys_str_mv |
AT yonglixu ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT manhu ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT hanruoliu ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT haoyang ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT huaizhouwang ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT shuailu ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT tianweiliang ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT xiaoxingli ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT maixu ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT liuli ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT huiqili ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT xinji ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT zhijunwang ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT lili ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT robertnweinreb ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT ningliwang ahierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT yonglixu hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT manhu hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT hanruoliu hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT haoyang hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT huaizhouwang hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT shuailu hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT tianweiliang hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT xiaoxingli hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT maixu hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT liuli hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT huiqili hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT xinji hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT zhijunwang hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT lili hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT robertnweinreb hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis AT ningliwang hierarchicaldeeplearningapproachwithtransparencyandinterpretabilitybasedonsmallsamplesforglaucomadiagnosis |
| _version_ |
1718392714732503040 |