Deep Learning Using Preoperative Optical Coherence Tomography Images to Predict Visual Acuity Following Surgery for Idiopathic Full-Thickness Macular Holes
This study presents a fully automated image informatics framework. The framework is combined with a deep learning (DL) approach to automatically predict visual acuity outcomes for people undergoing surgery for idiopathic full-thickness macular holes using 3D spectral-domain optical coherence tomogra...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10445126/ |
| _version_ | 1797271546736148480 |
|---|---|
| author | Burak Kucukgoz Muhammed Mutlu Yapici Declan C. Murphy Emma Spowart David H. Steel Boguslaw Obara |
| author_facet | Burak Kucukgoz Muhammed Mutlu Yapici Declan C. Murphy Emma Spowart David H. Steel Boguslaw Obara |
| author_sort | Burak Kucukgoz |
| collection | DOAJ |
| description | This study presents a fully automated image informatics framework. The framework is combined with a deep learning (DL) approach to automatically predict visual acuity outcomes for people undergoing surgery for idiopathic full-thickness macular holes using 3D spectral-domain optical coherence tomography (SD-OCT) images. To overcome the impact of high variation in real-world image quality on the robustness of DL models, comprehensive imaging data pre-processing, quality assurance, and anomaly detection procedures were utilised. We then implemented, trained, and tested nine state-of-the-art DL predictive models through our designed loss function with multiple 2D input channels on the imaging dataset. Finally, we quantitatively compared the models using four evaluation metrics. Overall, the predictive model achieved a MAE of 6.47 ETDRS letters score, demonstrating high predictability. This confirms that our fully automated approach with input from seven central SD-OCT images from each patient can robustly predict visual acuity measurements. Further research will focus on adapting 3D DL-based predictive models and the uncertainty of 2D and 3D DL-based predictive models. |
| first_indexed | 2024-03-07T14:05:07Z |
| format | Article |
| id | doaj.art-94bbf36c34ca4bb28d45da8b572f4756 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-07T14:05:07Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-94bbf36c34ca4bb28d45da8b572f47562024-03-07T00:00:25ZengIEEEIEEE Access2169-35362024-01-0112329113292610.1109/ACCESS.2024.336967610445126Deep Learning Using Preoperative Optical Coherence Tomography Images to Predict Visual Acuity Following Surgery for Idiopathic Full-Thickness Macular HolesBurak Kucukgoz0https://orcid.org/0000-0002-5589-4430Muhammed Mutlu Yapici1https://orcid.org/0000-0001-6171-1226Declan C. Murphy2Emma Spowart3https://orcid.org/0000-0002-4205-4440David H. Steel4https://orcid.org/0000-0001-8734-3089Boguslaw Obara5https://orcid.org/0000-0003-4084-7778School of Computing, Newcastle University, Newcastle upon Tyne, U.K.Elmadag Vocational School, Ankara University, Ankara, TurkeyBioscience Institute, Newcastle University, Newcastle upon Tyne, U.K.Sunderland Eye Infirmary, Sunderland, U.K.Bioscience Institute, Newcastle University, Newcastle upon Tyne, U.K.School of Computing, Newcastle University, Newcastle upon Tyne, U.K.This study presents a fully automated image informatics framework. The framework is combined with a deep learning (DL) approach to automatically predict visual acuity outcomes for people undergoing surgery for idiopathic full-thickness macular holes using 3D spectral-domain optical coherence tomography (SD-OCT) images. To overcome the impact of high variation in real-world image quality on the robustness of DL models, comprehensive imaging data pre-processing, quality assurance, and anomaly detection procedures were utilised. We then implemented, trained, and tested nine state-of-the-art DL predictive models through our designed loss function with multiple 2D input channels on the imaging dataset. Finally, we quantitatively compared the models using four evaluation metrics. Overall, the predictive model achieved a MAE of 6.47 ETDRS letters score, demonstrating high predictability. This confirms that our fully automated approach with input from seven central SD-OCT images from each patient can robustly predict visual acuity measurements. Further research will focus on adapting 3D DL-based predictive models and the uncertainty of 2D and 3D DL-based predictive models.https://ieeexplore.ieee.org/document/10445126/Image analysismachine learningoptical coherence tomographyvisual acuity measurement |
| spellingShingle | Burak Kucukgoz Muhammed Mutlu Yapici Declan C. Murphy Emma Spowart David H. Steel Boguslaw Obara Deep Learning Using Preoperative Optical Coherence Tomography Images to Predict Visual Acuity Following Surgery for Idiopathic Full-Thickness Macular Holes IEEE Access Image analysis machine learning optical coherence tomography visual acuity measurement |
| title | Deep Learning Using Preoperative Optical Coherence Tomography Images to Predict Visual Acuity Following Surgery for Idiopathic Full-Thickness Macular Holes |
| title_full | Deep Learning Using Preoperative Optical Coherence Tomography Images to Predict Visual Acuity Following Surgery for Idiopathic Full-Thickness Macular Holes |
| title_fullStr | Deep Learning Using Preoperative Optical Coherence Tomography Images to Predict Visual Acuity Following Surgery for Idiopathic Full-Thickness Macular Holes |
| title_full_unstemmed | Deep Learning Using Preoperative Optical Coherence Tomography Images to Predict Visual Acuity Following Surgery for Idiopathic Full-Thickness Macular Holes |
| title_short | Deep Learning Using Preoperative Optical Coherence Tomography Images to Predict Visual Acuity Following Surgery for Idiopathic Full-Thickness Macular Holes |
| title_sort | deep learning using preoperative optical coherence tomography images to predict visual acuity following surgery for idiopathic full thickness macular holes |
| topic | Image analysis machine learning optical coherence tomography visual acuity measurement |
| url | https://ieeexplore.ieee.org/document/10445126/ |
| work_keys_str_mv | AT burakkucukgoz deeplearningusingpreoperativeopticalcoherencetomographyimagestopredictvisualacuityfollowingsurgeryforidiopathicfullthicknessmacularholes AT muhammedmutluyapici deeplearningusingpreoperativeopticalcoherencetomographyimagestopredictvisualacuityfollowingsurgeryforidiopathicfullthicknessmacularholes AT declancmurphy deeplearningusingpreoperativeopticalcoherencetomographyimagestopredictvisualacuityfollowingsurgeryforidiopathicfullthicknessmacularholes AT emmaspowart deeplearningusingpreoperativeopticalcoherencetomographyimagestopredictvisualacuityfollowingsurgeryforidiopathicfullthicknessmacularholes AT davidhsteel deeplearningusingpreoperativeopticalcoherencetomographyimagestopredictvisualacuityfollowingsurgeryforidiopathicfullthicknessmacularholes AT boguslawobara deeplearningusingpreoperativeopticalcoherencetomographyimagestopredictvisualacuityfollowingsurgeryforidiopathicfullthicknessmacularholes |