Comparing code-free and bespoke deep learning approaches in ophthalmology.

AIM: Code-free deep learning (CFDL) allows clinicians without coding expertise to build high-quality artificial intelligence (AI) models without writing code. In this review, we comprehensively review the advantages that CFDL offers over bespoke expert-designed deep learning (DL). As exemplars, we use the following tasks: (1) diabetic retinopathy screening, (2) retinal multi-disease classification, (3) surgical video classification, (4) oculomics and (5) resource management. METHODS: We performed a search for studies reporting CFDL applications in ophthalmology in MEDLINE (through PubMed) from inception to June 25, 2023, using the keywords 'autoML' AND 'ophthalmology'. After identifying 5 CFDL studies looking at our target tasks, we performed a subsequent search to find corresponding bespoke DL studies focused on the same tasks. Only English-written articles with full text available were included. Reviews, editorials, protocols and case reports or case series were excluded. We identified ten relevant studies for this review. RESULTS: Overall, studies were optimistic towards CFDL's advantages over bespoke DL in the five ophthalmological tasks. However, much of such discussions were identified to be mono-dimensional and had wide applicability gaps. High-quality assessment of better CFDL applicability over bespoke DL warrants a context-specific, weighted assessment of clinician intent, patient acceptance and cost-effectiveness. We conclude that CFDL and bespoke DL are unique in their own assets and are irreplaceable with each other. Their benefits are differentially valued on a case-to-case basis. Future studies are warranted to perform a multidimensional analysis of both techniques and to improve limitations of suboptimal dataset quality, poor applicability implications and non-regulated study designs. CONCLUSION: For clinicians without DL expertise and easy access to AI experts, CFDL allows the prototyping of novel clinical AI systems. CFDL models concert with bespoke models, depending on the task at hand. A multidimensional, weighted evaluation of the factors involved in the implementation of those models for a designated task is warranted.

Association of retinal neurodegeneration with the progression of cognitive decline in Parkinson's disease.

Retinal thickness may serve as a biomarker in Parkinson's disease (PD). In this prospective longitudinal study, we aimed to determine if PD patients present accelerated thinning rate in the parafoveal ganglion cell-inner plexiform layer (pfGCIPL) and peripapillary retinal nerve fiber layer (pRNFL) compared to controls. Additionally, we evaluated the relationship between retinal neurodegeneration and clinical progression in PD. A cohort of 156 PD patients and 72 controls underwent retinal optical coherence tomography, visual, and cognitive assessments between February 2015 and December 2021 in two Spanish tertiary hospitals. The pfGCIPL thinning rate was twice as high in PD (ß [SE] = -0.58 [0.06]) than in controls (ß [SE] = -0.29 [0.06], p < 0.001). In PD, the progression pattern of pfGCIPL atrophy depended on baseline thickness, with slower thinning rates observed in PD patients with pfGCIPL below 89.8 µm. This result was validated with an external dataset from Moorfields Eye Hospital NHS Foundation Trust (AlzEye study). Slow pfGCIPL progressors, characterized by older at baseline, longer disease duration, and worse cognitive and disease stage scores, showed a threefold increase in the rate of cognitive decline (ß [SE] = -0.45 [0.19] points/year, p = 0.021) compared to faster progressors. Furthermore, temporal sector pRNFL thinning was accelerated in PD (ßtime x group [SE] = -0.67 [0.26] µm/year, p = 0.009), demonstrating a close association with cognitive score changes (ß [SE] = 0.11 [0.05], p = 0.052). This study suggests that a slower pattern of pfGCIPL tissue loss in PD is linked to more rapid cognitive decline, whereas changes in temporal pRNFL could track cognitive deterioration.

Fundus topographical distribution patterns of ocular toxoplasmosis.

BACKGROUND: To establish topographic maps and determine fundus distribution patterns of ocular toxoplasmosis (OT) lesions. METHODS: In this retrospective study, patients who presented with OT to ophthalmology clinics from four countries (Argentina, Turkey, UK, USA) were included. Size, shape and location of primary (1°)/recurrent (2°) and active/inactive lesions were converted into a two-dimensional retinal chart by a retinal drawing software. A final contour map of the merged image charts was then created using a custom Matlab programme. Descriptive analyses were performed. RESULTS: 984 lesions in 514 eyes of 464 subjects (53% women) were included. Mean area of all 1° and 2° lesions was 5.96±12.26 and 5.21±12.77 mm2, respectively. For the subset group lesions (eyes with both 1° and 2° lesions), 1° lesions were significantly larger than 2° lesions (5.52±6.04 mm2 vs 4.09±8.90 mm2, p=0.038). Mean distances from foveola to 1° and 2° lesion centres were 6336±4267 and 5763±3491 µm, respectively. The majority of lesions were found in temporal quadrant (p<0.001). Maximum overlap of all lesions was at 278 µm inferotemporal to foveola. CONCLUSION: The 1° lesions were larger than 2° lesions. The 2° lesions were not significantly closer to fovea than 1° lesions. Temporal quadrant and macular region were found to be densely affected underlining the vision threatening nature of the disease.

Exploring Patient Demographics and Presence of Retinal Vascular Disease in Paracentral Acute Middle Maculopathy.

PURPOSE: To investigate the sociodemographic profile, the association with retinal vascular diseases (RVD) and systemic comorbidities, and visual outcomes of patients with paracentral acute middle maculopathy (PAMM) in a large, ethnically diverse single-center cohort. DESIGN: Retrospective cohort study. METHODS: Electronic health record query for all patients presenting with PAMM at Moorfields Eye Hospital, London, was completed. Detailed demographic, clinical, and systemic information were collected and analyzed. RESULTS: A total of 78 eyes of 78 patients with confirmed PAMM were included in the study. Forty patients (51.3%) presented with no RVD, 20 patients (25.6%) with retinal vein occlusion (RVO), 16 patients (20.5%) with retinal artery occlusion (RAO), and 2 patients (2.6%) with concomitant RAO and RVO. Patients with PAMM+RAO were older than those with RVO (P = .02) and more likely to have a history of major adverse cardiovascular events (MACE) (P = .01), with a significantly worse presenting best corrected visual acuity (BCVA) (20/50) compared to patients with RVO (P = .02) and no RVD (P < .001). Individuals with isolated PAMM had a significantly higher prevalence of previous MACE (P = .04) and sickle cell disease (SCD) (P = .04) compared to those with RVO. At the last follow-up, 64 patients (85.3%) had a good BCVA (>20/32). CONCLUSIONS: The significant association of PAMM with RVD supports the hypothesis of an ischemic etiology. Individuals with isolated PAMM had a higher prevalence of MACE and SCD. Thus, it is important to prompt immediate referral for a comprehensive systemic evaluation. Across the whole cohort, PAMM was associated with good BCVA improvement during follow-up, indicating a good visual prognosis.

Retinal Optical Coherence Tomography Features Associated With Incident and Prevalent Parkinson Disease.

BACKGROUND AND OBJECTIVES: Cadaveric studies have shown disease-related neurodegeneration and other morphological abnormalities in the retina of individuals with Parkinson disease (PD); however, it remains unclear whether this can be reliably detected with in vivo imaging. We investigated inner retinal anatomy, measured using optical coherence tomography (OCT), in prevalent PD and subsequently assessed the association of these markers with the development of PD using a prospective research cohort. METHODS: This cross-sectional analysis used data from 2 studies. For the detection of retinal markers in prevalent PD, we used data from AlzEye, a retrospective cohort of 154,830 patients aged 40 years and older attending secondary care ophthalmic hospitals in London, United Kingdom, between 2008 and 2018. For the evaluation of retinal markers in incident PD, we used data from UK Biobank, a prospective population-based cohort where 67,311 volunteers aged 40-69 years were recruited between 2006 and 2010 and underwent retinal imaging. Macular retinal nerve fiber layer (mRNFL), ganglion cell-inner plexiform layer (GCIPL), and inner nuclear layer (INL) thicknesses were extracted from fovea-centered OCT. Linear mixed-effects models were fitted to examine the association between prevalent PD and retinal thicknesses. Hazard ratios for the association between time to PD diagnosis and retinal thicknesses were estimated using frailty models. RESULTS: Within the AlzEye cohort, there were 700 individuals with prevalent PD and 105,770 controls (mean age 65.5 ± 13.5 years, 51.7% female). Individuals with prevalent PD had thinner GCIPL (-2.12 µm, 95% CI -3.17 to -1.07, p = 8.2 × 10-5) and INL (-0.99 µm, 95% CI -1.52 to -0.47, p = 2.1 × 10-4). The UK Biobank included 50,405 participants (mean age 56.1 ± 8.2 years, 54.7% female), of whom 53 developed PD at a mean of 2,653 ± 851 days. Thinner GCIPL (hazard ratio [HR] 0.62 per SD increase, 95% CI 0.46-0.84, p = 0.002) and thinner INL (HR 0.70, 95% CI 0.51-0.96, p = 0.026) were also associated with incident PD. DISCUSSION: Individuals with PD have reduced thickness of the INL and GCIPL of the retina. Involvement of these layers several years before clinical presentation highlight a potential role for retinal imaging for at-risk stratification of PD.