The quest for early detection of retinal disease: 3D CycleGAN-based translation of optical coherence tomography into confocal microscopy DOI Creative Commons
Xin Tian, Nantheera Anantrasirichai, Lindsay B. Nicholson

et al.

Biological Imaging, Journal Year: 2024, Volume and Issue: 4

Published: Jan. 1, 2024

Abstract Optical coherence tomography (OCT) and confocal microscopy are pivotal in retinal imaging, offering distinct advantages limitations. In vivo OCT offers rapid, noninvasive imaging but can suffer from clarity issues motion artifacts, while ex microscopy, providing high-resolution, cellular-detailed color images, is invasive raises ethical concerns. To bridge the benefits of both modalities, we propose a novel framework based on unsupervised 3D CycleGAN for translating unpaired to images. This marks first attempt exploit inherent information translate it into rich, detailed domain microscopy. We also introduce unique dataset, OCT2Confocal, comprising mouse facilitating development establishing benchmark cross-modal image translation research. Our model has been evaluated quantitatively qualitatively, achieving Fréchet inception distance (FID) scores 0.766 Kernel Inception Distance (KID) as low 0.153, leading subjective mean opinion (MOS). demonstrated superior fidelity quality with limited data over existing methods. approach effectively synthesizes closely approximating target outcomes suggesting enhanced potential diagnostic monitoring applications ophthalmology.

Language: Английский

The AI Revolution in Glaucoma: Bridging Challenges with Opportunities DOI
Fei Li, Biao Wang, Zefeng Yang

et al.

Progress in Retinal and Eye Research, Journal Year: 2024, Volume and Issue: 103, P. 101291 - 101291

Published: Aug. 25, 2024

Language: Английский

Citations

8
Artificial intelligence for the detection of glaucoma with SD-OCT images: a systematic review and Meta-analysis DOI Creative Commons
Nannan Shi, Guanghui Liu,

Ming-Fang Cao

et al.

International Journal of Ophthalmology, Journal Year: 2024, Volume and Issue: 17(3), P. 408 - 419

Published: Feb. 27, 2024

To quantify the performance of artificial intelligence (AI) in detecting glaucoma with spectral-domain optical coherence tomography (SD-OCT) images.

Language: Английский

Citations

4
Valuable insights into general practice staff's experiences and perspectives on AI-assisted diabetic retinopathy screening—An interview study DOI Creative Commons
M. Krogh,

Malene Hentze,

Morten Sig Ager Jensen

et al.

Frontiers in Medicine, Journal Year: 2025, Volume and Issue: 12

Published: March 11, 2025

Aim This study explores the hands-on experiences and perspectives of general practice staff regarding feasibility conducting artificial intelligence-assisted (AI-assisted) diabetic retinopathy screenings (DRS) in settings. Method The were tested 12 practices North Denmark Region conducted as part daily care routines over ~4 weeks. Subsequently, 21 members involved DRS interviewed. Results Thematic analysis generated four main themes: (1) Experiences with practice, (2) Effective implementation future, (3) Trust approval AI-assisted (4) Implications practice. findings suggest that recognise potential for to be integrated into their clinical workflows. However, they also emphasise importance addressing both practical systemic factors ensure successful within setting. Conclusion Focusing on staff, this lays groundwork future research aimed at optimising settings, while recognising insights gained may inform broader primary contexts.

Language: Английский

Citations

0
Glaucoma diagnosis using Gabor and entropy coded Sine Cosine integration in adaptive partial swarm optimization-based FAWT DOI
Rajneesh Kumar Patel, Nancy Kumari, Siddharth Singh Chouhan

et al.

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 107, P. 107832 - 107832

Published: March 26, 2025

Language: Английский

Citations

0
Automated Early-Stage Glaucoma Detection Using a Robust Concatenated AI Model DOI Creative Commons
Wheyming Tina Song, Ing‐Chou Lai

Bioengineering, Journal Year: 2025, Volume and Issue: 12(5), P. 516 - 516

Published: May 13, 2025

Glaucoma is a leading cause of irreversible blindness worldwide; therefore, detection this disease in its early stage crucial. However, previous efforts to identify early-stage glaucoma have faced challenges, including insufficient accuracy, sensitivity, and specificity. This study presents concatenated artificial intelligence model that combines two types input features: fundus images quantitative retinal thickness parameters derived from macular peri-papillary nerve fiber layer (RNFL) measurements. These features undergo an intelligent transformation, referred as "smart preprocessing", enhance their utility. The employs classification approaches: convolutional neural network approach for processing image analyzing parameters. To maximize performance, hyperparameters were fine-tuned using robust methodology the design experiments. proposed AI demonstrated outstanding performance detection, outperforming existing models; specificity, precision, F1-Score all exceeding 0.90.

Language: Английский

Citations

0
Glaucoma: challenges and opportunities DOI Open Access
Michael Kalloniatis, Bang V. Bui, Jack Phu

et al.

Clinical and Experimental Optometry, Journal Year: 2024, Volume and Issue: 107(2), P. 107 - 109

Published: Feb. 17, 2024

Language: Английский

Citations

2
Harnessing the power of artificial intelligence for glaucoma diagnosis and treatment DOI Creative Commons
JohnDavis Akkara

Kerala Journal of Ophthalmology, Journal Year: 2024, Volume and Issue: 36(2), P. 194 - 199

Published: May 1, 2024

Artificial intelligence (AI) has great potential for diagnosing and managing glaucoma, a disease that causes irreversible vision loss. Early detection is paramount to prevent visual field AI algorithms demonstrate promising capabilities in analyzing various glaucoma investigations. In retinal fundus photographs, achieves high accuracy detecting glaucomatous optic nerve cupping, hallmark feature. can also analyze optical coherence tomography (OCT) images of the fiber layer(RNFL) ganglion cell complex, identifying structural changes indicative Anterior Segment OCT(AS-OCT) angle closure disease. OCT interpretation may even be extended diagnose early features systemic neurodegenerative diseases such as Alzheimer’s Disease Parkinson’s Disease. Furthermore, assist interpreting (VF) tests, including predicting future VF loss patterns next 5 years. The ability integrate data from multiple modalities, Intra Ocular Pressure(IOP) measurements, RNFL OCT, AS-OCT, paves way more comprehensive assessment. This approach revolutionize ophthalmology by enabling teleophthalmology facilitating development personalized treatment plans. However, authors emphasize crucial role human judgement oversight AI-generated results. Ultimately, ophthalmologists must make final decisions regarding diagnosis strategies.

Language: Английский

Citations

0
Automatic Glaucoma Classification and Feature Detection for the Justraigs Challenge DOI

Kristhian Aguilar,

Victor F. Cavalcante, Celso B. Carvalho

et al.

Published: May 27, 2024

Language: Английский

Citations

0
Radial polarisation patterns identify macular damage: a machine learning approach DOI Creative Commons
Gary P. Misson, Stephen J. Anderson, Mark Dunne

et al.

Clinical and Experimental Optometry, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 8

Published: Oct. 7, 2024

Clinical relevance Identifying polarisation-modulated patterns may be an effective method for both detecting and monitoring macular damage.

Language: Английский

Citations

0
The quest for early detection of retinal disease: 3D CycleGAN-based translation of optical coherence tomography into confocal microscopy DOI Creative Commons
Xin Tian, Nantheera Anantrasirichai, Lindsay B. Nicholson

et al.

Biological Imaging, Journal Year: 2024, Volume and Issue: 4

Published: Jan. 1, 2024

Abstract Optical coherence tomography (OCT) and confocal microscopy are pivotal in retinal imaging, offering distinct advantages limitations. In vivo OCT offers rapid, noninvasive imaging but can suffer from clarity issues motion artifacts, while ex microscopy, providing high-resolution, cellular-detailed color images, is invasive raises ethical concerns. To bridge the benefits of both modalities, we propose a novel framework based on unsupervised 3D CycleGAN for translating unpaired to images. This marks first attempt exploit inherent information translate it into rich, detailed domain microscopy. We also introduce unique dataset, OCT2Confocal, comprising mouse facilitating development establishing benchmark cross-modal image translation research. Our model has been evaluated quantitatively qualitatively, achieving Fréchet inception distance (FID) scores 0.766 Kernel Inception Distance (KID) as low 0.153, leading subjective mean opinion (MOS). demonstrated superior fidelity quality with limited data over existing methods. approach effectively synthesizes closely approximating target outcomes suggesting enhanced potential diagnostic monitoring applications ophthalmology.

Language: Английский

Citations

0