Deep learning based joint segmentation and characterization of multi-class retinal fluid lesions on OCT scans for clinical use in anti-VEGF therapy

• Published in: Computers in biology and medicine Vol. 136; p. 104727
• Main Authors: Hassan, Bilal, Qin, Shiyin, Ahmed, Ramsha, Hassan, Taimur, Taguri, Abdel Hakeem, Hashmi, Shahrukh, Werghi, Naoufel
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.09.2021; Elsevier Limited
• Subjects: Algorithms; Automation; Coders; Computational fluid dynamics; Datasets; Deep learning; Diabetes; Diabetic retinopathy; Dosimeters; Dosimetry; Edema; Growth factors; Image analysis; Image processing; Image segmentation; Internal Medicine; Lesions; Lesions detection; Macular degeneration; Medical image analysis; Optical Coherence Tomography; Optical coherence tomography (OCT); Other; Radiomics; Retina; Retinal fluids segmentation; Standardization; Therapy; Vascular endothelial growth factor; Deep learning; Lesions detection; Retinal fluids segmentation; Optical coherence tomography (OCT); Medical image analysis; Radiomics
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2021.104727

In anti-vascular endothelial growth factor (anti-VEGF) therapy, an accurate estimation of multi-class retinal fluid (MRF) is required for the activity prescription and intravitreal dose. This study proposes an end-to-end deep learning-based retinal fluids segmentation network (RFS-Net) to segment and recognize three MRF lesion manifestations, namely, intraretinal fluid (IRF), subretinal fluid (SRF), and pigment epithelial detachment (PED), from multi-vendor optical coherence tomography (OCT) imagery. The proposed image analysis tool will optimize anti-VEGF therapy and contribute to reducing the inter- and intra-observer variability. The proposed RFS-Net architecture integrates the atrous spatial pyramid pooling (ASPP), residual, and inception modules in the encoder path to learn better features and conserve more global information for precise segmentation and characterization of MRF lesions. The RFS-Net model is trained and validated using OCT scans from multiple vendors (Topcon, Cirrus, Spectralis), collected from three publicly available datasets. The first dataset consisted of OCT volumes obtained from 112 subjects (a total of 11,334 B-scans) is used for both training and evaluation purposes. Moreover, the remaining two datasets are only used for evaluation purposes to check the trained RFS-Net's generalizability on unseen OCT scans. The two evaluation datasets contain a total of 1572 OCT B-scans from 1255 subjects. The performance of the proposed RFS-Net model is assessed through various evaluation metrics. The proposed RFS-Net model achieved the mean F1 scores of 0.762, 0.796, and 0.805 for segmenting IRF, SRF, and PED. Moreover, with the automated segmentation of the three retinal manifestations, the RFS-Net brings a considerable gain in efficiency compared to the tedious and demanding manual segmentation procedure of the MRF. Our proposed RFS-Net is a potential diagnostic tool for the automatic segmentation of MRF (IRF, SRF, and PED) lesions. It is expected to strengthen the inter-observer agreement, and standardization of dosimetry is envisaged as a result. [Display omitted] •A deep learning based retinal fluids segmentation network (RFS-Net) is proposed.•It segments multi-class retinal fluid (MRF) lesions (IRF, SRF and PED) in OCT scan.•RFS-Net integrates a comprehensive pre-processing mechanism to boost the performance.•Using recurring ASPP modules, RFS-Net learns inherent multiscale MRF features.•RFS-Net outperforms state-of-the-art schemes in segmentation of MRF lesions in OCT.