Noninvasive identification of Benign and malignant eyelid tumors using clinical images via deep learning system
Abstract Eyelid tumors accounts for 5–10% of skin tumors. It is important but difficult to identify malignant eyelid tumors from benign lesions in a cost-effective way. Traditional screening methods for malignancy in eyelid tumors require laborious and time-consuming histopathological process. There...
Main Authors: | , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2022-06-01
|
Series: | Journal of Big Data |
Subjects: | |
Online Access: | https://doi.org/10.1186/s40537-022-00634-y |
_version_ | 1811334670355267584 |
---|---|
author | Shiqi Hui Li Dong Kai Zhang Zihan Nie Xue Jiang Heyan Li Zhijia Hou Jingwen Ding Yue Wang Dongmei Li |
author_facet | Shiqi Hui Li Dong Kai Zhang Zihan Nie Xue Jiang Heyan Li Zhijia Hou Jingwen Ding Yue Wang Dongmei Li |
author_sort | Shiqi Hui |
collection | DOAJ |
description | Abstract Eyelid tumors accounts for 5–10% of skin tumors. It is important but difficult to identify malignant eyelid tumors from benign lesions in a cost-effective way. Traditional screening methods for malignancy in eyelid tumors require laborious and time-consuming histopathological process. Therefore, we aimed to develop a deep learning (DL)-based image analysis system for automatic identification of benign and malignant eyelid tumors. Using a common digital camera, we collected clinical images from patients who were histopathologically diagnosed with eyelid tumors. We trained 8 convolutional neural network (CNN) models to identify benign and malignant eyelid tumors, including ResNet-50, ResNet-101, InceptionV3, and InceptionResNetV2. Another group of patients with eyelid tumors were also collected as the prospective validation dataset. Performance of DL models and human clinicians in prospective validation dataset were evaluated and compared. A total of 309 images from 209 patients were used for training DL system, all eight models reached an average accuracy greater than 0.958 in the internal cross-validation. 36 images from 36 patients were included for the prospective validation, the models reached the best performance in accuracy, sensitivity, specificity, and area under curve (AUC) of 0.889 (95% CI 0.747–0.956), 0.933 (95% CI 0.702–0.988), 0.857 (95% CI 0.654–0.950), and 0.966 (95% CI 0.850–0.993), respectively. DL system had a similar performance as the senior ophthalmologists, and outreached the performance of junior ophthalmologists and medical students. DL system can identify benign and malignant tumors through common clinical images, with a better performance than most ophthalmologists. Combining DL system with smartphone may enable patients’ self-monitoring for eyelid tumors and assist in doctors’ clinical decision making. |
first_indexed | 2024-04-13T17:11:32Z |
format | Article |
id | doaj.art-148803da5f3f44b08e473892c57556d2 |
institution | Directory Open Access Journal |
issn | 2196-1115 |
language | English |
last_indexed | 2024-04-13T17:11:32Z |
publishDate | 2022-06-01 |
publisher | SpringerOpen |
record_format | Article |
series | Journal of Big Data |
spelling | doaj.art-148803da5f3f44b08e473892c57556d22022-12-22T02:38:15ZengSpringerOpenJournal of Big Data2196-11152022-06-019111410.1186/s40537-022-00634-yNoninvasive identification of Benign and malignant eyelid tumors using clinical images via deep learning systemShiqi Hui0Li Dong1Kai Zhang2Zihan Nie3Xue Jiang4Heyan Li5Zhijia Hou6Jingwen Ding7Yue Wang8Dongmei Li9Beijing Ophthalmology & Visual Science Key Laboratory, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityBeijing Key Laboratory of Intraocular Tumor Diagnosis and Treatment, Beijing Ophthalmology & Visual Sciences Key Lab, Medical Artificial Intelligence Research and Verification Key Laboratory of the Ministry of Industry and Information Technology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityInferVision Healthcare Science and Technology Limited CompanyBeijing Ophthalmology & Visual Science Key Laboratory, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityBeijing Ophthalmology & Visual Science Key Laboratory, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityBeijing Ophthalmology & Visual Science Key Laboratory, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityBeijing Ophthalmology & Visual Science Key Laboratory, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityBeijing Ophthalmology & Visual Science Key Laboratory, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityBeijing Ophthalmology & Visual Science Key Laboratory, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityBeijing Ophthalmology & Visual Science Key Laboratory, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical UniversityAbstract Eyelid tumors accounts for 5–10% of skin tumors. It is important but difficult to identify malignant eyelid tumors from benign lesions in a cost-effective way. Traditional screening methods for malignancy in eyelid tumors require laborious and time-consuming histopathological process. Therefore, we aimed to develop a deep learning (DL)-based image analysis system for automatic identification of benign and malignant eyelid tumors. Using a common digital camera, we collected clinical images from patients who were histopathologically diagnosed with eyelid tumors. We trained 8 convolutional neural network (CNN) models to identify benign and malignant eyelid tumors, including ResNet-50, ResNet-101, InceptionV3, and InceptionResNetV2. Another group of patients with eyelid tumors were also collected as the prospective validation dataset. Performance of DL models and human clinicians in prospective validation dataset were evaluated and compared. A total of 309 images from 209 patients were used for training DL system, all eight models reached an average accuracy greater than 0.958 in the internal cross-validation. 36 images from 36 patients were included for the prospective validation, the models reached the best performance in accuracy, sensitivity, specificity, and area under curve (AUC) of 0.889 (95% CI 0.747–0.956), 0.933 (95% CI 0.702–0.988), 0.857 (95% CI 0.654–0.950), and 0.966 (95% CI 0.850–0.993), respectively. DL system had a similar performance as the senior ophthalmologists, and outreached the performance of junior ophthalmologists and medical students. DL system can identify benign and malignant tumors through common clinical images, with a better performance than most ophthalmologists. Combining DL system with smartphone may enable patients’ self-monitoring for eyelid tumors and assist in doctors’ clinical decision making.https://doi.org/10.1186/s40537-022-00634-yDeep learningEyelid tumorClinical image |
spellingShingle | Shiqi Hui Li Dong Kai Zhang Zihan Nie Xue Jiang Heyan Li Zhijia Hou Jingwen Ding Yue Wang Dongmei Li Noninvasive identification of Benign and malignant eyelid tumors using clinical images via deep learning system Journal of Big Data Deep learning Eyelid tumor Clinical image |
title | Noninvasive identification of Benign and malignant eyelid tumors using clinical images via deep learning system |
title_full | Noninvasive identification of Benign and malignant eyelid tumors using clinical images via deep learning system |
title_fullStr | Noninvasive identification of Benign and malignant eyelid tumors using clinical images via deep learning system |
title_full_unstemmed | Noninvasive identification of Benign and malignant eyelid tumors using clinical images via deep learning system |
title_short | Noninvasive identification of Benign and malignant eyelid tumors using clinical images via deep learning system |
title_sort | noninvasive identification of benign and malignant eyelid tumors using clinical images via deep learning system |
topic | Deep learning Eyelid tumor Clinical image |
url | https://doi.org/10.1186/s40537-022-00634-y |
work_keys_str_mv | AT shiqihui noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT lidong noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT kaizhang noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT zihannie noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT xuejiang noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT heyanli noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT zhijiahou noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT jingwending noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT yuewang noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem AT dongmeili noninvasiveidentificationofbenignandmalignanteyelidtumorsusingclinicalimagesviadeeplearningsystem |