Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
173 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

OUCopula: Bi-Channel Multi-Label Copula-Enhanced Adapter-Based CNN for Myopia Screening Based on OU-UWF Images (2403.11974v1)

Published 18 Mar 2024 in eess.IV and cs.CV

Abstract: Myopia screening using cutting-edge ultra-widefield (UWF) fundus imaging is potentially significant for ophthalmic outcomes. Current multidisciplinary research between ophthalmology and deep learning (DL) concentrates primarily on disease classification and diagnosis using single-eye images, largely ignoring joint modeling and prediction for Oculus Uterque (OU, both eyes). Inspired by the complex relationships between OU and the high correlation between the (continuous) outcome labels (Spherical Equivalent and Axial Length), we propose a framework of copula-enhanced adapter convolutional neural network (CNN) learning with OU UWF fundus images (OUCopula) for joint prediction of multiple clinical scores. We design a novel bi-channel multi-label CNN that can (1) take bi-channel image inputs subject to both high correlation and heterogeneity (by sharing the same backbone network and employing adapters to parameterize the channel-wise discrepancy), and (2) incorporate correlation information between continuous output labels (using a copula). Solid experiments show that OUCopula achieves satisfactory performance in myopia score prediction compared to backbone models. Moreover, OUCopula can far exceed the performance of models constructed for single-eye inputs. Importantly, our study also hints at the potential extension of the bi-channel model to a multi-channel paradigm and the generalizability of OUCopula across various backbone CNNs.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (42)
  1. Ultra-wide field imaging in paradoxical worsening of tubercular multifocal serpiginoid choroiditis after the initiation of anti-tubercular therapy. Ocular Immunology and Inflammation, 2017.
  2. Fundus-deepnet: Multi-label deep learning classification system for enhanced detection of multiple ocular diseases through data fusion of fundus images. Information Fusion, 102:102059, 2024.
  3. Learning graph convolutional networks for multi-label recognition and applications. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021.
  4. Deep learning models for screening of high myopia using optical coherence tomography. Scientific reports, 11(1):21663, 2021.
  5. Jaydeep Devda and R Eswari. Pathological myopia image analysis using deep learning. Procedia Computer Science, 165:239–244, 2019.
  6. Artificial intelligence for screening of multiple retinal and optic nerve diseases. JAMA Network Open, 5(5):e229960–e229960, 2022.
  7. Detecting multiple retinal diseases in ultra-widefield fundus imaging and data-driven identification of informative regions with deep learning. Nature Machine Intelligence, 4(12):1143–1154, 2022.
  8. A simplified severity scale for age-related macular degeneration: Areds report no. 18. Archives of ophthalmology (Chicago, Ill.: 1960), 123(11):1570–1574, 2005.
  9. Multi-class multi-label ophthalmological disease detection using transfer learning based convolutional neural network. Biomedical Signal Processing and Control, 66:102329, 2021.
  10. Multi-label ocular disease classification with a dense correlation deep neural network. Biomedical Signal Processing and Control, 63:102167, 2021.
  11. Pathological myopia classification with simultaneous lesion segmentation using deep learning. Computer Methods and Programs in Biomedicine, 199:105920, 2021.
  12. Intereye asymmetry in eyes with keratoconus and high ammetropia: Scheimpflug imaging analysis. Cornea, 34:S57–S60, 2015.
  13. Global prevalence of myopia and high myopia and temporal trends from 2000 through 2050. Ophthalmology, 123(5):1036–1042, 2016.
  14. Asymmetry between right and left fundus images identified using convolutional neural networks. Scientific Reports, 12(1):1444, 2022.
  15. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7482–7491, 2018.
  16. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
  17. Single-stage broad multi-instance multi-label learning (bmiml) with diverse inter-correlations and its application to medical image classification. IEEE Transactions on Emerging Topics in Computational Intelligence, 2023.
  18. Adaptable multi-domain language model for transformer asr. In ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 7358–7362. IEEE, 2021.
  19. Deep learning for automated glaucomatous optic neuropathy detection from ultra-widefield fundus images. British journal of ophthalmology, 105(11):1548–1554, 2021.
  20. Cross-domain few-shot learning with task-specific adapters. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7161–7170, 2022.
  21. Joint classification and regression via deep multi-task multi-channel learning for alzheimer’s disease diagnosis. IEEE Transactions on Biomedical Engineering, 66(5):1195–1206, 2018.
  22. Interocular asymmetry of choroidal thickness and vascularity index measurements in normal eyes assessed by swept-source optical coherence tomography. Quantitative imaging in medicine and surgery, 12(1):781, 2022.
  23. Multichannel retinal blood vessel segmentation based on the combination of matched filter and u-net network. BioMed research international, 2021:1–18, 2021.
  24. Vit-dr: Vision transformers in diabetic retinopathy grading using fundus images. In 2022 IEEE 10th Region 10 Humanitarian Technology Conference (R10-HTC), pages 167–172. IEEE, 2022.
  25. Refractive error, axial length, and relative peripheral refractive error before and after the onset of myopia. Investigative ophthalmology & visual science, 48(6):2510–2519, 2007.
  26. Ultra-widefield fundus imaging: a review of clinical applications and future trends. Retina, 36(4):660–678, 2016.
  27. Accuracy of deep learning, a machine-learning technology, using ultra–wide-field fundus ophthalmoscopy for detecting rhegmatogenous retinal detachment. Scientific reports, 7(1):9425, 2017.
  28. Automatic detection of pathological myopia using machine learning. Scientific Reports, 11(1):16570, 2021.
  29. Efficient parametrization of multi-domain deep neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 8119–8127, 2018.
  30. Correlation in myopia progression between right and left eyes in chinese children with myopia. Investigative Ophthalmology & Visual Science, 54(15):5712–5712, 2013.
  31. M Sklar. Fonctions de repartition an dimensions et leurs marges. Publ. inst. statist. univ. Paris, 8:229–231, 1959.
  32. A deep multi-modal cnn for multi-instance multi-label image classification. IEEE Transactions on Image Processing, 27(12):6025–6038, 2018.
  33. Multi-label classification of fundus images with graph convolutional network and lightgbm. Computers in Biology and Medicine, 149:105909, 2022.
  34. An automatic early screening system of eye diseases using uwf fundus images based on deep neural networks. Intelligence-Based Medicine, 6:100079, 2022.
  35. Ctranscnn: Combining transformer and cnn in multilabel medical image classification. Knowledge-Based Systems, 281:111030, 2023.
  36. Dual-channel asymmetric convolutional neural network for an efficient retinal blood vessel segmentation in eye fundus images. Biocybernetics and Biomedical Engineering, 42(2):695–706, 2022.
  37. Learning mutually local-global u-nets for high-resolution retinal lesion segmentation in fundus images. In 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), pages 597–600. IEEE, 2019.
  38. Automatic identification of myopia based on ocular appearance images using deep learning. Annals of Translational Medicine, 8(11), 2020.
  39. Retinal artery/vein classification by multi-channel multi-scale fusion network. Applied Intelligence, 53(22):26400–26417, 2023.
  40. Prediction of juvenile-onset myopia. JAMA ophthalmology, 133(6):683–689, 2015.
  41. Cecnn: Copula-enhanced convolutional neural networks in joint prediction of refraction error and axial length based on ultra-widefield fundus images. arXiv preprint arXiv:2311.03967, 2023.
  42. Yinlong Zhu and Yi Zhang. M3la: A novel approach based on encoder-decoder with attention framework for multi-modal multi-label learning. In 2020 International Joint Conference on Neural Networks (IJCNN), pages 1–8. IEEE, 2020.
Citations (2)

Summary

We haven't generated a summary for this paper yet.