Application of Convolutional Gated Recurrent Units U-Net for Distinguishing between Retinitis Pigmentosa and Cone–Rod Dystrophy

Skublewska-Paszkowska, Maria; Powroznik, Pawel; Rejdak, Robert; Nowomiejska, Katarzyna

doi:10.2478/ama-2024-0054

Abstract

Artificial Intelligence (AI) has gained a prominent role in the medical industry. The rapid development of the computer science field has caused AI to become a meaningful part of modern healthcare. Image-based analysis involving neural networks is a very important part of eye diagnoses. In this study, a new approach using Convolutional Gated Recurrent Units (GRU) U-Net was proposed for the classifying healthy cases and cases with retinitis pigmentosa (RP) and cone–rod dystrophy (CORD). The basis for the classification was the location of pigmentary changes within the retina and fundus autofluorescence (FAF) pattern, as the posterior pole or the periphery of the retina may be affected. The dataset, gathered in the Chair and Department of General and Pediatric Ophthalmology of Medical University in Lublin, consisted of 230 ultra-widefield pseudocolour (UWFP) and ultra-widefield FAF images, obtained using the Optos 200TX device (Optos PLC). The data were divided into three categories: healthy subjects (50 images), patients with CORD (48 images) and patients with RP (132 images). For applying deep learning classification, which rely on a large amount of data, the dataset was artificially enlarged using augmentation involving image manipulations. The final dataset contained 744 images. The proposed Convolutional GRU U-Net network was evaluated taking account of the following measures: accuracy, precision, sensitivity, specificity and F1. The proposed tool achieved high accuracy in a range of 91.00%–97.90%. The developed solution has a great potential in RP diagnoses as a supporting tool.

References

Abeysinghe A, Tohmuang S, Davy J L, Fard M. Data augmentation on convolutional neural networks to classify mechanical noise. Appl. Acoust. 2023:203:109209.
Search in Google Scholar Back to article
Alomar K, Aysel H I, Cai X. Data augmentation in classification and segmentation: A survey and new strategies. J. Imaging. 2023; 9(2): 46.
Search in Google Scholar Back to article
Azad R, Asadi-Aghbolaghi M, Fathy M, Escalera S. Bi-directional ConvLSTM U-Net with densley connected convolutions. 2019. Proc - IEEE/CVF international conference on computer vision workshops.
Search in Google Scholar Back to article
Baratloo A, Hosseini M, Negida A, El Ashal, G. Part 1: Simple definition and calculation of accuracy, sensitivity and specificity. Emergency. 2015;3(2):48-49.
Search in Google Scholar Back to article
Berger W, Kloeckener-Gruissem B, Neidhardt J. The molecular basis of human retinal and vitreoretinal diseases. Prog Retin Eye Res. 2010;29(5):335–75.
Search in Google Scholar Back to article
Bonnici E, Arn P. The impact of Data Augmentation on classification accuracy and training time in Handwritten Character Recognition. Kth Royal Institute of Technology. 2021.
Search in Google Scholar Back to article
Brancati N, Frucci M, Gragnaniello D, Riccio D, Di Iorio V, Di Perna L. Automatic segmentation of pigment deposits in retinal fundus images of Retinitis Pigmentosa. Comput. Med. Imag. Graph. 2018;66:73-81.
Search in Google Scholar Back to article
Brancati N, Frucci M, Gragnaniello D, Riccio D, Di Iorio V, Di Perna L, Simonelli F. Learning-based approach to segment pigment signs in fundus images for retinitis pigmentosa analysis. Neurocomputing. 2018,308:159-171.
Search in Google Scholar Back to article
Chen JX, Jiang DM, Zhang YN, A hierarchical bidirectional GRU model with attention for EEG-based emotion classification, IEEE Access. 2019;7:118530-118540.
Search in Google Scholar Back to article
Das H, Saha A, Deb S. An expert system to distinguish a defective eye from a normal eye. Proc - 2014 International Conference on Issues and Challenges in Intelligent Computing Techniques (ICICT). IEEE. 2014:155-158.
Search in Google Scholar Back to article
Fahim AT, Daiger SP, Weleber RG. Nonsyndromic retinitis pigmentosa overview. 2017: Adam MP, Ardinger HH, Pagon RA, Wallace SE, Bean LJH, Stephens K, Amemiya A, eds. Gene Reviews. Seattle: University of Washington.
Search in Google Scholar Back to article
Gill JS, Georgiou M, Kalitzeos A, Moore AT, Michaelides M. Progressive cone and cone-rod dystrophies: Clinical features, molecular genetics and prospects for therapy. Br. J. Ophthalmol. 2019;103(5): 711-720.
Search in Google Scholar Back to article
Graves A, Mohamed AR, Hinton G. Speech recognition with deep recurrent neural networks. Proc - IEEE Int. Conf. Acoust., Speech Signal Process. 2013;38:6645–6649.
Search in Google Scholar Back to article
Guo C, Yu M, Li J. Prediction of different eye diseases based on fundus photography via deep transfer learning. J. Clin. Med. 2021;10(23):5481.
Search in Google Scholar Back to article
Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. Proc - IEEE conference on computer vision and pattern recognition. 2017:4700-4708.
Search in Google Scholar Back to article
Hartong DT, Berson EL, Dryja TP. Retinitis pigmentosa. Lancet. 200618;368(9549):1795-809. doi: 10.1016/S0140-6736(06)69740-7
Search in Google Scholar Back to article
Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift. Proc - International conference on machine learning. 2015:448-456.
Search in Google Scholar Back to article
Jain L, Murthy H S, Patel C, Bansal D. Retinal eye disease detection using deep learning. Proc - Fourteenth International Conference on Information Processing (ICINPRO). IEEE. 2018:1-6.
Search in Google Scholar Back to article
LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521(7553):436-444.
Search in Google Scholar Back to article
Liu T Y A, Ling C, Hahn L, Jones C K, Boon C J, Singh M S. Prediction of visual impairment in retinitis pigmentosa using deep learning and multimodal fundus images. Br. J. Ophthalmol. 2022.
Search in Google Scholar Back to article
Masumot H, Tabuchi H, Nakakura S, Ohsugi H, Enno H, Ishitobi, N., Ohsugi E, Mitamura Y. Accuracy of a deep convolutional neural network in detection of retinitis pigmentosa on ultrawide-field images. PeerJ. 2019;7:e6900.
Search in Google Scholar Back to article
Merin S, Auerbach E. Retinitis pigmentosa. Surv. of Ophthalmol. 1976; 20(5):303-46. doi: 10.1016/s0039-6257(96)90001-6
Search in Google Scholar Back to article
Monaghan T F, Rahman S N, Agudelo C W, Wein A J, Lazar J M, Everaert K, Dmochowski R R. Foundational statistical principles in medical research: sensitivity, specificity, positive predictive value, and negative predictive value. Medicina. 2021;57(5):503.
Search in Google Scholar Back to article
Oishi A, Miyata M, Numa S, Otsuka Y, Oishi M, Tsujikawa A. Wide-field fundus autofluorescence imaging in patients with hereditary retinal degeneration: a literature review. Int. J. of Retina Vitr. 2019;12(5) (Suppl 1):23. https://doi.org/10.1186/s40942-019-0173-z.
Search in Google Scholar Back to article
Piri N, Grodsky JD, Kaplan HJ. Gene therapy for retinitis pigmentosa. Taiwan J. Ophthalmol. 2021;11(4):348-351.
Search in Google Scholar Back to article
RetNet Retinal Information Network. https://sph.uth.edu/retnet/ [6.06.2023]
Search in Google Scholar Back to article
Robson AG, Egan CA, Luong VA, Bird AC, Holder GE, Fitzke FW. Comparison of FAF with photopic and scotopic fine-matrix mapping in patients with retinitis pigmentosa and normal visual acuity. Invest. Ophthalmol. Vis. Sci. 2004;45(11):4119-4125.
Search in Google Scholar Back to article
Romo-Bucheli D, Erfurth U S, Bogunović, H. End-to-end deep learning model for predicting treatment requirements in neovascular AMD from longitudinal retinal OCT imaging. IEEE J. Biomed. Health Inform. 2020;24(12):3456-3465.
Search in Google Scholar Back to article
Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. Proc - 18th International Conference, Munich, Germany. October 5-9. Proceedings. Part III. 2015; 18:234-241. Springer International Publishing.
Search in Google Scholar Back to article
Schmitz-Valckenberg S, Holz FG, Bird AC, Spaide RF. Fundus autofluorescence imaging: review and perspectives. Retina. 2008;28(3):385-409.
Search in Google Scholar Back to article
Shi X, Chen Z, Wang H, Yeung DY, Wong WK, Woo WC. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. Adv. Neural. Inf. Process. Syst. 2015;28.
Search in Google Scholar Back to article
Shorten C, Khoshgoftaar T M. A survey on image data augmentation for deep learning. J. Big Data. 2019;6(1):1-48.
Search in Google Scholar Back to article
Skublewska-Paszkowska M, Powroznik P. Temporal Pattern Attention for Multivariate Time Series of Tennis Strokes Classification. Sensors. 2023;23(5):2422.
Search in Google Scholar Back to article
Song H, Wang W, Zhao S, Shen J, Lam KM. Pyramid dilated deeper convlstm for video salient object detection. Proc - European conference on computer vision (ECCV). 2018:715-731.
Search in Google Scholar Back to article
Sun G, Wang X. Xu L. Li C. Wang W. Yi, Z., Luo H, Su Y, Zheng J, Li Z, Chen Z, Zheng H, Chen, C. Deep learning for the detection of multiple fundus diseases using ultra-widefield images. Ophthalmol. Ther. 2023;12(2):895-907. https://doi.org/10.1007/s40123-022-00627-3
Search in Google Scholar Back to article
Tee JJ, Smith AJ, Hardcastle AJ, Michaelides M. RPGR-associated retinopathy: clinical features, molecular genetics, animal models and therapeutic options. Br J Ophthalmolol. 2016;100(8):1022-7. doi: 10.1136/bjophthalmol-2015-307698
Search in Google Scholar Back to article
Wong S C, Gatt A, Stamatescu V, McDonnell M D. Understanding data augmentation for classification: when to warp? Proc - International conference on digital image computing: techniques and applications. IEEE. 2016:1-6.
Search in Google Scholar Back to article
Yang S, Xiao W, Zhang M, Guo S, Zhao J, Shen F. Image data augmentation for deep learning: A survey. 2022. arXiv preprint arXiv:2204.08610.
Search in Google Scholar Back to article

Application of Convolutional Gated Recurrent Units U-Net for Distinguishing between Retinitis Pigmentosa and Cone–Rod Dystrophy

Abstract

Paradigm

My account