Artificial Intelligence for the Estimation of Visual Acuity Using Multi-Source Anterior Segment Optical Coherence Tomographic Images in Senile Cataract
PurposeTo investigate an artificial intelligence (AI) model performance using multi-source anterior segment optical coherence tomographic (OCT) images in estimating the preoperative best-corrected visual acuity (BCVA) in patients with senile cataract.DesignRetrospective, cross-instrument validation...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-05-01
|
| Series: | Frontiers in Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fmed.2022.871382/full |
| _version_ | 1811235295401607168 |
|---|---|
| author | Hyunmin Ahn Ikhyun Jun Ikhyun Jun Kyoung Yul Seo Eung Kweon Kim Eung Kweon Kim Tae-im Kim Tae-im Kim |
| author_facet | Hyunmin Ahn Ikhyun Jun Ikhyun Jun Kyoung Yul Seo Eung Kweon Kim Eung Kweon Kim Tae-im Kim Tae-im Kim |
| author_sort | Hyunmin Ahn |
| collection | DOAJ |
| description | PurposeTo investigate an artificial intelligence (AI) model performance using multi-source anterior segment optical coherence tomographic (OCT) images in estimating the preoperative best-corrected visual acuity (BCVA) in patients with senile cataract.DesignRetrospective, cross-instrument validation study.SubjectsA total of 2,332 anterior segment images obtained using swept-source OCT, optical biometry for intraocular lens calculation, and a femtosecond laser platform in patients with senile cataract and postoperative BCVA ≥ 0.0 logMAR were included in the training/validation dataset. A total of 1,002 images obtained using optical biometry and another femtosecond laser platform in patients who underwent cataract surgery in 2021 were used for the test dataset.MethodsAI modeling was based on an ensemble model of Inception-v4 and ResNet. The BCVA training/validation dataset was used for model training. The model performance was evaluated using the test dataset. Analysis of absolute error (AE) was performed by comparing the difference between true preoperative BCVA and estimated preoperative BCVA, as ≥0.1 logMAR (AE≥0.1) or <0.1 logMAR (AE <0.1). AE≥0.1 was classified into underestimation and overestimation groups based on the logMAR scale.Outcome MeasurementsMean absolute error (MAE), root mean square error (RMSE), mean percentage error (MPE), and correlation coefficient between true preoperative BCVA and estimated preoperative BCVA.ResultsThe test dataset MAE, RMSE, and MPE were 0.050 ± 0.130 logMAR, 0.140 ± 0.134 logMAR, and 1.3 ± 13.9%, respectively. The correlation coefficient was 0.969 (p < 0.001). The percentage of cases with AE≥0.1 was 8.4%. The incidence of postoperative BCVA > 0.1 was 21.4% in the AE≥0.1 group, of which 88.9% were in the underestimation group. The incidence of vision-impairing disease in the underestimation group was 95.7%. Preoperative corneal astigmatism and lens thickness were higher, and nucleus cataract was more severe (p < 0.001, 0.007, and 0.024, respectively) in AE≥0.1 than that in AE <0.1. The longer the axial length and the more severe the cortical/posterior subcapsular opacity, the better the estimated BCVA than the true BCVA.ConclusionsThe AI model achieved high-level visual acuity estimation in patients with senile cataract. This quantification method encompassed both visual acuity and cataract severity of OCT image, which are the main indications for cataract surgery, showing the potential to objectively evaluate cataract severity. |
| first_indexed | 2024-04-12T11:48:25Z |
| format | Article |
| id | doaj.art-770d45c1ae9540578a604dbacb66f532 |
| institution | Directory Open Access Journal |
| issn | 2296-858X |
| language | English |
| last_indexed | 2024-04-12T11:48:25Z |
| publishDate | 2022-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Medicine |
| spelling | doaj.art-770d45c1ae9540578a604dbacb66f5322022-12-22T03:34:15ZengFrontiers Media S.A.Frontiers in Medicine2296-858X2022-05-01910.3389/fmed.2022.871382871382Artificial Intelligence for the Estimation of Visual Acuity Using Multi-Source Anterior Segment Optical Coherence Tomographic Images in Senile CataractHyunmin Ahn0Ikhyun Jun1Ikhyun Jun2Kyoung Yul Seo3Eung Kweon Kim4Eung Kweon Kim5Tae-im Kim6Tae-im Kim7Department of Ophthalmology, Institute of Vision Research, Yonsei University College of Medicine, Seoul, South KoreaDepartment of Ophthalmology, Institute of Vision Research, Yonsei University College of Medicine, Seoul, South KoreaCorneal Dystrophy Research Institute, Yonsei University College of Medicine, Seoul, South KoreaDepartment of Ophthalmology, Institute of Vision Research, Yonsei University College of Medicine, Seoul, South KoreaCorneal Dystrophy Research Institute, Yonsei University College of Medicine, Seoul, South KoreaSaevit Eye Hospital, Goyang, South KoreaDepartment of Ophthalmology, Institute of Vision Research, Yonsei University College of Medicine, Seoul, South KoreaCorneal Dystrophy Research Institute, Yonsei University College of Medicine, Seoul, South KoreaPurposeTo investigate an artificial intelligence (AI) model performance using multi-source anterior segment optical coherence tomographic (OCT) images in estimating the preoperative best-corrected visual acuity (BCVA) in patients with senile cataract.DesignRetrospective, cross-instrument validation study.SubjectsA total of 2,332 anterior segment images obtained using swept-source OCT, optical biometry for intraocular lens calculation, and a femtosecond laser platform in patients with senile cataract and postoperative BCVA ≥ 0.0 logMAR were included in the training/validation dataset. A total of 1,002 images obtained using optical biometry and another femtosecond laser platform in patients who underwent cataract surgery in 2021 were used for the test dataset.MethodsAI modeling was based on an ensemble model of Inception-v4 and ResNet. The BCVA training/validation dataset was used for model training. The model performance was evaluated using the test dataset. Analysis of absolute error (AE) was performed by comparing the difference between true preoperative BCVA and estimated preoperative BCVA, as ≥0.1 logMAR (AE≥0.1) or <0.1 logMAR (AE <0.1). AE≥0.1 was classified into underestimation and overestimation groups based on the logMAR scale.Outcome MeasurementsMean absolute error (MAE), root mean square error (RMSE), mean percentage error (MPE), and correlation coefficient between true preoperative BCVA and estimated preoperative BCVA.ResultsThe test dataset MAE, RMSE, and MPE were 0.050 ± 0.130 logMAR, 0.140 ± 0.134 logMAR, and 1.3 ± 13.9%, respectively. The correlation coefficient was 0.969 (p < 0.001). The percentage of cases with AE≥0.1 was 8.4%. The incidence of postoperative BCVA > 0.1 was 21.4% in the AE≥0.1 group, of which 88.9% were in the underestimation group. The incidence of vision-impairing disease in the underestimation group was 95.7%. Preoperative corneal astigmatism and lens thickness were higher, and nucleus cataract was more severe (p < 0.001, 0.007, and 0.024, respectively) in AE≥0.1 than that in AE <0.1. The longer the axial length and the more severe the cortical/posterior subcapsular opacity, the better the estimated BCVA than the true BCVA.ConclusionsThe AI model achieved high-level visual acuity estimation in patients with senile cataract. This quantification method encompassed both visual acuity and cataract severity of OCT image, which are the main indications for cataract surgery, showing the potential to objectively evaluate cataract severity.https://www.frontiersin.org/articles/10.3389/fmed.2022.871382/fullartificial intelligencecataractconvolutional neural networkoptical coherence tomographyvisual acuity |
| spellingShingle | Hyunmin Ahn Ikhyun Jun Ikhyun Jun Kyoung Yul Seo Eung Kweon Kim Eung Kweon Kim Tae-im Kim Tae-im Kim Artificial Intelligence for the Estimation of Visual Acuity Using Multi-Source Anterior Segment Optical Coherence Tomographic Images in Senile Cataract Frontiers in Medicine artificial intelligence cataract convolutional neural network optical coherence tomography visual acuity |
| title | Artificial Intelligence for the Estimation of Visual Acuity Using Multi-Source Anterior Segment Optical Coherence Tomographic Images in Senile Cataract |
| title_full | Artificial Intelligence for the Estimation of Visual Acuity Using Multi-Source Anterior Segment Optical Coherence Tomographic Images in Senile Cataract |
| title_fullStr | Artificial Intelligence for the Estimation of Visual Acuity Using Multi-Source Anterior Segment Optical Coherence Tomographic Images in Senile Cataract |
| title_full_unstemmed | Artificial Intelligence for the Estimation of Visual Acuity Using Multi-Source Anterior Segment Optical Coherence Tomographic Images in Senile Cataract |
| title_short | Artificial Intelligence for the Estimation of Visual Acuity Using Multi-Source Anterior Segment Optical Coherence Tomographic Images in Senile Cataract |
| title_sort | artificial intelligence for the estimation of visual acuity using multi source anterior segment optical coherence tomographic images in senile cataract |
| topic | artificial intelligence cataract convolutional neural network optical coherence tomography visual acuity |
| url | https://www.frontiersin.org/articles/10.3389/fmed.2022.871382/full |
| work_keys_str_mv | AT hyunminahn artificialintelligencefortheestimationofvisualacuityusingmultisourceanteriorsegmentopticalcoherencetomographicimagesinsenilecataract AT ikhyunjun artificialintelligencefortheestimationofvisualacuityusingmultisourceanteriorsegmentopticalcoherencetomographicimagesinsenilecataract AT ikhyunjun artificialintelligencefortheestimationofvisualacuityusingmultisourceanteriorsegmentopticalcoherencetomographicimagesinsenilecataract AT kyoungyulseo artificialintelligencefortheestimationofvisualacuityusingmultisourceanteriorsegmentopticalcoherencetomographicimagesinsenilecataract AT eungkweonkim artificialintelligencefortheestimationofvisualacuityusingmultisourceanteriorsegmentopticalcoherencetomographicimagesinsenilecataract AT eungkweonkim artificialintelligencefortheestimationofvisualacuityusingmultisourceanteriorsegmentopticalcoherencetomographicimagesinsenilecataract AT taeimkim artificialintelligencefortheestimationofvisualacuityusingmultisourceanteriorsegmentopticalcoherencetomographicimagesinsenilecataract AT taeimkim artificialintelligencefortheestimationofvisualacuityusingmultisourceanteriorsegmentopticalcoherencetomographicimagesinsenilecataract |