Morphologic and Functional Assessment of Photoreceptors in Laser-Induced Retinopathy Using Adaptive Optics Scanning Laser Ophthalmoscopy and Microperimetry.

PURPOSE: To assess the cone photoreceptors' morphology and associated retinal sensitivity in laser-induced retinopathy (LIR) using adaptive optics scanning laser ophthalmoscopy (AO-SLO) and microperimetry (MP). DESIGN: Cohort study. METHODS: This study included 13 patients (15 eyes) with LIR and 38 age-matched healthy volunteers (38 eyes). Participants underwent comprehensive evaluations including AO-SLO, MP, and spectral-domain OCT. Lesion morphology, cone density, dispersion, and regularity in AO-SLO were assessed and correlated with visual function. RESULTS: In AO-SLO images, LIR lesions were predominantly characterized by hyporeflective regions, suggesting potential cone loss at the fovea, accompanied by the presence of sizable clumps of hyperreflective material within these lesions. The average size of lesions in affected eyes was 97,128±107,478 µm², ranging from 6705 to 673,348 µm². Compared with the healthy contralateral eye and control group, LIR demonstrated significantly reduced cone density, increased cone dispersion, and notably decreased cone regularity in all 4 quadrants at 3° eccentricity (all P values < .05). Lesion morphology in AO-SLO correlated with ellipsoid zone defects observed in OCT, showing a positive correlation in size (r = 0.84, P < .001) but not with retinal sensitivities (P = .09). Similarly, cone density at 3° eccentricity did not correlate with retinal sensitivities (P = .13). CONCLUSIONS AND RELEVANCE: The study provides crucial insights into the morphologic and functional impacts of LIR on cone photoreceptors, revealing significant morphologic changes in cones that do not consistently align with functional outcomes. This research highlights the need for continued exploration into the relationship between retinal structure and function in LIR, and the importance of heightened public awareness and preventive strategies to mitigate the risk of LIR.

Automatic diagnosis of imbalanced ophthalmic images using a cost-sensitive deep convolutional neural network.

BACKGROUND: Ocular images play an essential role in ophthalmological diagnoses. Having an imbalanced dataset is an inevitable issue in automated ocular diseases diagnosis; the scarcity of positive samples always tends to result in the misdiagnosis of severe patients during the classification task. Exploring an effective computer-aided diagnostic method to deal with imbalanced ophthalmological dataset is crucial. METHODS: In this paper, we develop an effective cost-sensitive deep residual convolutional neural network (CS-ResCNN) classifier to diagnose ophthalmic diseases using retro-illumination images. First, the regions of interest (crystalline lens) are automatically identified via twice-applied Canny detection and Hough transformation. Then, the localized zones are fed into the CS-ResCNN to extract high-level features for subsequent use in automatic diagnosis. Second, the impacts of cost factors on the CS-ResCNN are further analyzed using a grid-search procedure to verify that our proposed system is robust and efficient. RESULTS: Qualitative analyses and quantitative experimental results demonstrate that our proposed method outperforms other conventional approaches and offers exceptional mean accuracy (92.24%), specificity (93.19%), sensitivity (89.66%) and AUC (97.11%) results. Moreover, the sensitivity of the CS-ResCNN is enhanced by over 13.6% compared to the native CNN method. CONCLUSION: Our study provides a practical strategy for addressing imbalanced ophthalmological datasets and has the potential to be applied to other medical images. The developed and deployed CS-ResCNN could serve as computer-aided diagnosis software for ophthalmologists in clinical application.