GlauCUTU: Time Until Perceived Virtual Reality Perimetry With Humphrey Field Analyzer Prediction-Based Artificial Intelligence
The increasing prevalence of glaucomatous optic neuropathy, which can result in permanent blindness (visual impairment), accentuates the importance of screening and early diagnosis for prevention of blindness. GlauCUTU, a novel time until perceived (TUP) visual field (VF) testing, utilizes a portabl...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9745592/ |
| _version_ | 1811272767922765824 |
|---|---|
| author | Patthapol Kunumpol Nichapa Lerthirunvibul Phongphan Phienphanich Adirek Munthuli Kanjapat Temahivong Visanee Tantisevi Anita Manassakorn Sunee Chansangpetch Rath Itthipanichpong Kitiya Ratanawongphaibul Prin Rojanapongpun Charturong Tantibundhit |
| author_facet | Patthapol Kunumpol Nichapa Lerthirunvibul Phongphan Phienphanich Adirek Munthuli Kanjapat Temahivong Visanee Tantisevi Anita Manassakorn Sunee Chansangpetch Rath Itthipanichpong Kitiya Ratanawongphaibul Prin Rojanapongpun Charturong Tantibundhit |
| author_sort | Patthapol Kunumpol |
| collection | DOAJ |
| description | The increasing prevalence of glaucomatous optic neuropathy, which can result in permanent blindness (visual impairment), accentuates the importance of screening and early diagnosis for prevention of blindness. GlauCUTU, a novel time until perceived (TUP) visual field (VF) testing, utilizes a portable virtual reality (VR) headset with visual stimulus that progressively increases in intensity to detect VF defects. GlauCUTU was evaluated on participants with normal visual fields and those with early, moderate, and severe glaucoma. Responses were collected in terms of time until response (TUR). TUR was used to calculate TUP and reported in terms of GlauCUTU sensitivity. False positives were detected with pretest and latency analysis using reaction time (RT). In addition, a novel automated transformation was developed to convert GlauCUTU sensitivity into HFA sensitivity using machine learning (ML) and deep learning (DL) algorithms. Visual field index (VFI) was generated from HFA sensitivity to determine severity of glaucoma. The VFI results were evaluated using post-hoc analysis from two-way analysis of variance (ANOVA). Results demonstrate no significant difference (p=0.073) between Humphrey visual field analyzer (HFA) and GlauCUTU with machine learning transformation (GlauCUTU-ML) in all glaucoma stages. However, there was a significant difference between HFA and GlauCUTU with deep learning transformation (GlauCUTU-DL) in severe glaucoma (p<0.050). GlauCUTU-ML generates the lowest root mean square error (RMSE) of 4.92. Meanwhile, GlauCUTU-DL yields the highest Pearson’s <inline-formula> <tex-math notation="LaTeX">$r$ </tex-math></inline-formula> correlation coefficient with HFA of 0.74, but produces the highest RMSE of 6.31. Comparison between three expert ophthalmologists’ grading of glaucomatous eyes on GlauCUTU-ML and HFA aligns with the majority voting with an average agreement of 0.83, which is highly reliable. All in all, the portable and inexpensive GlauCUTU perimetry system introduces the use of TUP for VF evaluation with results comparable to HFA. GlauCUTU proves to be a promising method to increase accessibility to glaucoma screening, particularly in low-resource setting countries. |
| first_indexed | 2024-04-12T22:46:15Z |
| format | Article |
| id | doaj.art-a4374be80348420389cb68479be0e735 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-12T22:46:15Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-a4374be80348420389cb68479be0e7352022-12-22T03:13:31ZengIEEEIEEE Access2169-35362022-01-0110369493696210.1109/ACCESS.2022.31638459745592GlauCUTU: Time Until Perceived Virtual Reality Perimetry With Humphrey Field Analyzer Prediction-Based Artificial IntelligencePatthapol Kunumpol0https://orcid.org/0000-0002-3925-1566Nichapa Lerthirunvibul1https://orcid.org/0000-0003-4766-1650Phongphan Phienphanich2Adirek Munthuli3Kanjapat Temahivong4Visanee Tantisevi5Anita Manassakorn6Sunee Chansangpetch7Rath Itthipanichpong8Kitiya Ratanawongphaibul9Prin Rojanapongpun10Charturong Tantibundhit11https://orcid.org/0000-0002-3889-7314Center of Excellence in Intelligent Informatics, Speech and Language Technology, and Service Innovation (CILS), Thammasat University, Rangsit Campus, Khlong Luang, Bangkok, Pathum Thani, ThailandCenter of Excellence in Intelligent Informatics, Speech and Language Technology, and Service Innovation (CILS), Thammasat University, Rangsit Campus, Khlong Luang, Bangkok, Pathum Thani, ThailandCenter of Excellence in Intelligent Informatics, Speech and Language Technology, and Service Innovation (CILS), Thammasat University, Rangsit Campus, Khlong Luang, Bangkok, Pathum Thani, ThailandCenter of Excellence in Intelligent Informatics, Speech and Language Technology, and Service Innovation (CILS), Thammasat University, Rangsit Campus, Khlong Luang, Bangkok, Pathum Thani, ThailandCenter of Excellence in Intelligent Informatics, Speech and Language Technology, and Service Innovation (CILS), Thammasat University, Rangsit Campus, Khlong Luang, Bangkok, Pathum Thani, ThailandDepartment of Ophthalmology, Faculty of Medicine, Glaucoma Research Unit, Chulalongkorn University, Bangkok, ThailandDepartment of Ophthalmology, Faculty of Medicine, Glaucoma Research Unit, Chulalongkorn University, Bangkok, ThailandDepartment of Ophthalmology, Faculty of Medicine, Glaucoma Research Unit, Chulalongkorn University, Bangkok, ThailandCenter of Excellence in Intelligent Informatics, Speech and Language Technology, and Service Innovation (CILS), Thammasat University, Rangsit Campus, Khlong Luang, Bangkok, Pathum Thani, ThailandDepartment of Ophthalmology, Faculty of Medicine, Glaucoma Research Unit, Chulalongkorn University, Bangkok, ThailandDepartment of Ophthalmology, Faculty of Medicine, Glaucoma Research Unit, Chulalongkorn University, Bangkok, ThailandCenter of Excellence in Intelligent Informatics, Speech and Language Technology, and Service Innovation (CILS), Thammasat University, Rangsit Campus, Khlong Luang, Bangkok, Pathum Thani, ThailandThe increasing prevalence of glaucomatous optic neuropathy, which can result in permanent blindness (visual impairment), accentuates the importance of screening and early diagnosis for prevention of blindness. GlauCUTU, a novel time until perceived (TUP) visual field (VF) testing, utilizes a portable virtual reality (VR) headset with visual stimulus that progressively increases in intensity to detect VF defects. GlauCUTU was evaluated on participants with normal visual fields and those with early, moderate, and severe glaucoma. Responses were collected in terms of time until response (TUR). TUR was used to calculate TUP and reported in terms of GlauCUTU sensitivity. False positives were detected with pretest and latency analysis using reaction time (RT). In addition, a novel automated transformation was developed to convert GlauCUTU sensitivity into HFA sensitivity using machine learning (ML) and deep learning (DL) algorithms. Visual field index (VFI) was generated from HFA sensitivity to determine severity of glaucoma. The VFI results were evaluated using post-hoc analysis from two-way analysis of variance (ANOVA). Results demonstrate no significant difference (p=0.073) between Humphrey visual field analyzer (HFA) and GlauCUTU with machine learning transformation (GlauCUTU-ML) in all glaucoma stages. However, there was a significant difference between HFA and GlauCUTU with deep learning transformation (GlauCUTU-DL) in severe glaucoma (p<0.050). GlauCUTU-ML generates the lowest root mean square error (RMSE) of 4.92. Meanwhile, GlauCUTU-DL yields the highest Pearson’s <inline-formula> <tex-math notation="LaTeX">$r$ </tex-math></inline-formula> correlation coefficient with HFA of 0.74, but produces the highest RMSE of 6.31. Comparison between three expert ophthalmologists’ grading of glaucomatous eyes on GlauCUTU-ML and HFA aligns with the majority voting with an average agreement of 0.83, which is highly reliable. All in all, the portable and inexpensive GlauCUTU perimetry system introduces the use of TUP for VF evaluation with results comparable to HFA. GlauCUTU proves to be a promising method to increase accessibility to glaucoma screening, particularly in low-resource setting countries.https://ieeexplore.ieee.org/document/9745592/Virtual realityglaucomavisual defectvisual field testportable perimetry |
| spellingShingle | Patthapol Kunumpol Nichapa Lerthirunvibul Phongphan Phienphanich Adirek Munthuli Kanjapat Temahivong Visanee Tantisevi Anita Manassakorn Sunee Chansangpetch Rath Itthipanichpong Kitiya Ratanawongphaibul Prin Rojanapongpun Charturong Tantibundhit GlauCUTU: Time Until Perceived Virtual Reality Perimetry With Humphrey Field Analyzer Prediction-Based Artificial Intelligence IEEE Access Virtual reality glaucoma visual defect visual field test portable perimetry |
| title | GlauCUTU: Time Until Perceived Virtual Reality Perimetry With Humphrey Field Analyzer Prediction-Based Artificial Intelligence |
| title_full | GlauCUTU: Time Until Perceived Virtual Reality Perimetry With Humphrey Field Analyzer Prediction-Based Artificial Intelligence |
| title_fullStr | GlauCUTU: Time Until Perceived Virtual Reality Perimetry With Humphrey Field Analyzer Prediction-Based Artificial Intelligence |
| title_full_unstemmed | GlauCUTU: Time Until Perceived Virtual Reality Perimetry With Humphrey Field Analyzer Prediction-Based Artificial Intelligence |
| title_short | GlauCUTU: Time Until Perceived Virtual Reality Perimetry With Humphrey Field Analyzer Prediction-Based Artificial Intelligence |
| title_sort | glaucutu time until perceived virtual reality perimetry with humphrey field analyzer prediction based artificial intelligence |
| topic | Virtual reality glaucoma visual defect visual field test portable perimetry |
| url | https://ieeexplore.ieee.org/document/9745592/ |
| work_keys_str_mv | AT patthapolkunumpol glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT nichapalerthirunvibul glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT phongphanphienphanich glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT adirekmunthuli glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT kanjapattemahivong glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT visaneetantisevi glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT anitamanassakorn glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT suneechansangpetch glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT rathitthipanichpong glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT kitiyaratanawongphaibul glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT prinrojanapongpun glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence AT charturongtantibundhit glaucututimeuntilperceivedvirtualrealityperimetrywithhumphreyfieldanalyzerpredictionbasedartificialintelligence |