Necessity of Local Modification for Deep Learning Algorithms to Predict Diabetic Retinopathy

Ching-Yao Tsai, Chueh-Tan Chen, Guan-An Chen, Chun-Fu Yeh, Chin-Tzu Kuo, Ya-Chuan Hsiao, Hsiao-Yun Hu, I-Lun Tsai, Ching-Hui Wang, Jian-Ren Chen, Su-Chen Huang, Tzu-Chieh Lu and Lin-Chung Woung
Additional contact information
Ching-Yao Tsai: Department of Ophthalmology, Taipei City Hospital, Taipei 103, Taiwan
Chueh-Tan Chen: Department of Ophthalmology, Taipei City Hospital, Taipei 103, Taiwan
Guan-An Chen: Smart Medical and Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu County 310, Taiwan
Chun-Fu Yeh: Smart Medical and Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu County 310, Taiwan
Chin-Tzu Kuo: Department of Ophthalmology, Taipei City Hospital, Taipei 103, Taiwan
Ya-Chuan Hsiao: Department of Ophthalmology, Taipei City Hospital, Taipei 103, Taiwan
Hsiao-Yun Hu: Institute of Public Health, National Yang Ming Chiao Tung University, Taipei 112, Taiwan
I-Lun Tsai: Department of Ophthalmology, Taipei City Hospital, Taipei 103, Taiwan
Ching-Hui Wang: Department of Ophthalmology, Taipei City Hospital, Taipei 103, Taiwan
Jian-Ren Chen: Smart Medical and Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu County 310, Taiwan
Su-Chen Huang: Smart Medical and Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu County 310, Taiwan
Tzu-Chieh Lu: Smart Medical and Healthcare, Service Systems Technology Center, Industrial Technology Research Institute, Hsinchu County 310, Taiwan
Lin-Chung Woung: Department of Ophthalmology, Taipei City Hospital, Taipei 103, Taiwan

IJERPH, 2022, vol. 19, issue 3, 1-12

Abstract: Deep learning (DL) algorithms are used to diagnose diabetic retinopathy (DR). However, most of these algorithms have been trained using global data or data from patients of a single region. Using different model architectures (e.g., Inception-v3, ResNet101, and DenseNet121), we assessed the necessity of modifying the algorithms for universal society screening. We used the open-source dataset from the Kaggle Diabetic Retinopathy Detection competition to develop a model for the detection of DR severity. We used a local dataset from Taipei City Hospital to verify the necessity of model localization and validated the three aforementioned models with local datasets. The experimental results revealed that Inception-v3 outperformed ResNet101 and DenseNet121 with a foreign global dataset, whereas DenseNet121 outperformed Inception-v3 and ResNet101 with the local dataset. The quadratic weighted kappa score ( κ ) was used to evaluate the model performance. All models had 5–8% higher κ for the local dataset than for the foreign dataset. Confusion matrix analysis revealed that, compared with the local ophthalmologists’ diagnoses, the severity predicted by the three models was overestimated. Thus, DL algorithms using artificial intelligence based on global data must be locally modified to ensure the applicability of a well-trained model to make diagnoses in clinical environments.

Keywords: diabetic retinopathy; deep learning algorithms; model localised; Taiwan; predict (search for similar items in EconPapers)
JEL-codes: I I1 I3 Q Q5 (search for similar items in EconPapers)
Date: 2022
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1660-4601/19/3/1204/pdf (application/pdf)
https://www.mdpi.com/1660-4601/19/3/1204/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jijerp:v:19:y:2022:i:3:p:1204-:d:730583

Access Statistics for this article

IJERPH is currently edited by Ms. Jenna Liu

More articles in IJERPH from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().