Multimodal Machine Learning Using Visual Fields and Peripapillary Circular OCT Scans in Detection of Glaucomatous Optic Neuropathy

Xiong, Jian; Li, Fēi; Song, Diping; Tang, Guangxian; He, Junjun; Gao, Xinbo; Zhang, Hengli; Cheng, Weiqiang; Song, Yunhe; Lin, Fengbin; Hu, Kun; Wang, Peiyuan; Li, Ji-Peng Olivia; Aung, Tin; Qiao, Yu; Zhang, Xiulan; Ting, Daniel

doi:10.1016/j.ophtha.2021.07.032

Cited by 38 publications

(15 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Ran et al for their valuable comments on our study. 1 First, we agree that a more convenient functional test for glaucoma is still needed, and algorithms based on single modality examinations such as fundus photography and OCT should be more feasible in large-scale glaucoma screening in the general population. However, visual field tests are essential and currently irreplaceable in glaucoma clinics, although they may take additional time.…”

mentioning

confidence: 82%

See 1 more Smart Citation

Reply

Xiong

Zhang

2022

Ophthalmology

Self Cite

View full text Add to dashboard Cite

We thank Dr. Ran et al for their valuable comments on our study. 1 First, we agree that a more convenient functional test for glaucoma is still needed, and algorithms based on single modality examinations such as fundus photography and OCT should be more feasible in large-scale glaucoma screening in the general population. However, visual field tests are essential and currently irreplaceable in glaucoma clinics, although they may take additional time. Visual field tests are helpful to differentiate glaucoma from high myopia and neuro-ophthalmic diseases. The best solution to the problem of visual field testing taking too much time is to design a simplified perimetry test, say using goggles to perform perimetry test, rather than giving up the test entirely. In many high-impact artificial intelligence studies, 2e4 OCT plus visual field data are used as the ground truth, demonstrating the importance and essence of visual field data in a glaucoma diagnosis.Second, diagnostic algorithms like FusionNet could, to a certain extent, bridge the imbalance of medical resources among regions in China. The difficulties and clinical needs of glaucoma diagnosis vary by region. There is a lack of glaucoma specialists in developed areas and we need a more precise clinical evaluation of glaucoma. Our algorithm could assist ophthalmologists who are not glaucoma subspecialists in diagnosing glaucoma based on visual field and OCT data, just as we do in clinical practice. In developing areas, there are fewer ophthalmologists and glaucoma screening is more important. OCT machines are rapidly spreading in these areas. OCTNet, the OCT processing module of FusionNet, could be used to diagnose glaucoma based on OCT images.Third, as the components of FusionNet, the VFNet and OCTNet can extract features of Pattern deviation probability plots from visual field data and peripapillary circular scans from OCT when only one kind of data is available, as we mentioned in the article. 1

show abstract

mentioning

confidence: 82%

“…OCTNet, the OCT processing module of FusionNet, could be used to diagnose glaucoma based on OCT images.Third, as the components of FusionNet, the VFNet and OCTNet can extract features of Pattern deviation probability plots from visual field data and peripapillary circular scans from OCT when only one kind of data is available, as we mentioned in the article. 1…”

mentioning

confidence: 99%

Reply

Xiong

Zhang

2022

Ophthalmology

Self Cite

View full text Add to dashboard Cite

show abstract

“…The ground truth of glaucoma grading task for each sample was determined based on mean deviation (MD) values from visual field reports following the criteria below: early-stage with ME value higher than -6 dB, intermediate stage with ME value between -6 and -12 dB, advanced stage with MD value worse than -12 dB. These visual field reports were generated on the same day as the OCT examination and were reliable with fixation losses of under 2/13 and false-positive rate under 15% and false-negative rate under 25% (Li et al (2020); Xiong et al (2021)).…”

Section: Glaucoma Gradingmentioning

confidence: 99%

GAMMA Challenge:Glaucoma grAding from Multi-Modality imAges

Wu¹,

Fang²,

Li³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

J. Wu et al. / Preprint (2022) development of fundus & OCT-based glaucoma grading. The primary task of the challenge is to grade glaucoma from both the 2D fundus images and 3D OCT scanning volumes. As part of GAMMA, we have publicly released a glaucoma annotated dataset with both 2D fundus color photography and 3D OCT volumes, which is the first multimodality dataset for glaucoma grading. In addition, an evaluation framework is also established to evaluate the performance of the submitted methods. During the challenge, 1272 results were submitted, and finally, top-10 teams were selected to the final stage. We analysis their results and summarize their methods in the paper. Since all these teams submitted their source code in the challenge, a detailed ablation study is also conducted to verify the effectiveness of the particular modules proposed. We find many of the proposed techniques are practical for the clinical diagnosis of glaucoma. As the first in-depth study of fundus & OCT multi-modality glaucoma grading, we believe the GAMMA Challenge will be an essential starting point for future research.

show abstract

“…Multimodal approaches have been applied in several areas, such as to detect glaucoma through combining clinical data with retinal imaging modalities [100] or using VF-OCT paired data [101 ▪ ,102,103], in predicting glaucoma progression based on VF and OCT data [104], or VF with clinical data [105]; in detecting AMD using fundus images with OCT, in predicting treatment response of neovascular AMD using a combination of OCT, fluorescein angiogram (FA), and indocyanine green angiography (ICGA) [108], and by also evaluating multiple factors (clinical, genetic and lifestyle) to predict AMD progression [96 ▪ ]; in detecting diabetic retinopathy by using OCT together with clinical and demographic data [106,107], and by identifying treatment-requiring retinal vascular diseases using OCT with FA, ICGA and fundus images [108].…”

Section: Text Of Reviewmentioning

confidence: 99%

Novel technical and privacy-preserving technology for artificial intelligence in ophthalmology

Lim

Hong²,

Lam

et al. 2022

Current Opinion in Ophthalmology

View full text Add to dashboard Cite

Purpose of reviewThe application of artificial intelligence (AI) in medicine and ophthalmology has experienced exponential breakthroughs in recent years in diagnosis, prognosis, and aiding clinical decision-making. The use of digital data has also heralded the need for privacy-preserving technology to protect patient confidentiality and to guard against threats such as adversarial attacks. Hence, this review aims to outline novel AI-based systems for ophthalmology use, privacy-preserving measures, potential challenges, and future directions of each. Recent findingsSeveral key AI algorithms used to improve disease detection and outcomes include: Data-driven, imagedriven, natural language processing (NLP)-driven, genomics-driven, and multimodality algorithms. However, deep learning systems are susceptible to adversarial attacks, and use of data for training models is associated with privacy concerns. Several data protection methods address these concerns in the form of blockchain technology, federated learning, and generative adversarial networks. SummaryAI-applications have vast potential to meet many eyecare needs, consequently reducing burden on scarce healthcare resources. A pertinent challenge would be to maintain data privacy and confidentiality while supporting AI endeavors, where data protection methods would need to rapidly evolve with AI technology needs. Ultimately, for AI to succeed in medicine and ophthalmology, a balance would need to be found between innovation and privacy.

show abstract

Multimodal Machine Learning Using Visual Fields and Peripapillary Circular OCT Scans in Detection of Glaucomatous Optic Neuropathy

Cited by 38 publications

References 29 publications

Reply

Reply

GAMMA Challenge:Glaucoma grAding from Multi-Modality imAges

Novel technical and privacy-preserving technology for artificial intelligence in ophthalmology

Contact Info

Product

Resources

About