Artificial intelligence for detecting keratoconus.

BACKGROUND: Keratoconus remains difficult to diagnose, especially in the early stages. It is a progressive disorder of the cornea that starts at a young age. Diagnosis is based on clinical examination and corneal imaging; though in the early stages, when there are no clinical signs, diagnosis depends on the interpretation of corneal imaging (e.g. topography and tomography) by trained cornea specialists. Using artificial intelligence (AI) to analyse the corneal images and detect cases of keratoconus could help prevent visual acuity loss and even corneal transplantation. However, a missed diagnosis in people seeking refractive surgery could lead to weakening of the cornea and keratoconus-like ectasia. There is a need for a reliable overview of the accuracy of AI for detecting keratoconus and the applicability of this automated method to the clinical setting. OBJECTIVES: To assess the diagnostic accuracy of artificial intelligence (AI) algorithms for detecting keratoconus in people presenting with refractive errors, especially those whose vision can no longer be fully corrected with glasses, those seeking corneal refractive surgery, and those suspected of having keratoconus. AI could help ophthalmologists, optometrists, and other eye care professionals to make decisions on referral to cornea specialists. Secondary objectives To assess the following potential causes of heterogeneity in diagnostic performance across studies. • Different AI algorithms (e.g. neural networks, decision trees, support vector machines) • Index test methodology (preprocessing techniques, core AI method, and postprocessing techniques) • Sources of input to train algorithms (topography and tomography images from Placido disc system, Scheimpflug system, slit-scanning system, or optical coherence tomography (OCT); number of training and testing cases/images; label/endpoint variable used for training) • Study setting • Study design • Ethnicity, or geographic area as its proxy • Different index test positivity criteria provided by the topography or tomography device • Reference standard, topography or tomography, one or two cornea specialists • Definition of keratoconus • Mean age of participants • Recruitment of participants • Severity of keratoconus (clinically manifest or subclinical) SEARCH METHODS: We searched CENTRAL (which contains the Cochrane Eyes and Vision Trials Register), Ovid MEDLINE, Ovid Embase, OpenGrey, the ISRCTN registry, ClinicalTrials.gov, and the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP). There were no date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 29 November 2022. SELECTION CRITERIA: We included cross-sectional and diagnostic case-control studies that investigated AI for the diagnosis of keratoconus using topography, tomography, or both. We included studies that diagnosed manifest keratoconus, subclinical keratoconus, or both. The reference standard was the interpretation of topography or tomography images by at least two cornea specialists. DATA COLLECTION AND ANALYSIS: Two review authors independently extracted the study data and assessed the quality of studies using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. When an article contained multiple AI algorithms, we selected the algorithm with the highest Youden's index. We assessed the certainty of evidence using the GRADE approach. MAIN RESULTS: We included 63 studies, published between 1994 and 2022, that developed and investigated the accuracy of AI for the diagnosis of keratoconus. There were three different units of analysis in the studies: eyes, participants, and images. Forty-four studies analysed 23,771 eyes, four studies analysed 3843 participants, and 15 studies analysed 38,832 images. Fifty-four articles evaluated the detection of manifest keratoconus, defined as a cornea that showed any clinical sign of keratoconus. The accuracy of AI seems almost perfect, with a summary sensitivity of 98.6% (95% confidence interval (CI) 97.6% to 99.1%) and a summary specificity of 98.3% (95% CI 97.4% to 98.9%). However, accuracy varied across studies and the certainty of the evidence was low. Twenty-eight articles evaluated the detection of subclinical keratoconus, although the definition of subclinical varied. We grouped subclinical keratoconus, forme fruste, and very asymmetrical eyes together. The tests showed good accuracy, with a summary sensitivity of 90.0% (95% CI 84.5% to 93.8%) and a summary specificity of 95.5% (95% CI 91.9% to 97.5%). However, the certainty of the evidence was very low for sensitivity and low for specificity. In both groups, we graded most studies at high risk of bias, with high applicability concerns, in the domain of patient selection, since most were case-control studies. Moreover, we graded the certainty of evidence as low to very low due to selection bias, inconsistency, and imprecision. We could not explain the heterogeneity between the studies. The sensitivity analyses based on study design, AI algorithm, imaging technique (topography versus tomography), and data source (parameters versus images) showed no differences in the results. AUTHORS' CONCLUSIONS: AI appears to be a promising triage tool in ophthalmologic practice for diagnosing keratoconus. Test accuracy was very high for manifest keratoconus and slightly lower for subclinical keratoconus, indicating a higher chance of missing a diagnosis in people without clinical signs. This could lead to progression of keratoconus or an erroneous indication for refractive surgery, which would worsen the disease. We are unable to draw clear and reliable conclusions due to the high risk of bias, the unexplained heterogeneity of the results, and high applicability concerns, all of which reduced our confidence in the evidence. Greater standardization in future research would increase the quality of studies and improve comparability between studies.

Immediate sequential bilateral surgery versus delayed sequential bilateral surgery for cataracts.

BACKGROUND: Age-related cataract affects both eyes in most cases. Most people undergo cataract surgery in both eyes on separate days, referred to as delayed sequential bilateral cataract surgery (DSBCS). An alternative procedure involves operating on both eyes on the same day, but as two separate procedures, known as immediate sequential bilateral cataract surgery (ISBCS). Potential advantages of ISBCS include fewer hospital visits for the patient, faster visual recovery, and lower healthcare costs. Nevertheless, concerns exist about possible bilateral, postoperative, sight-threatening adverse effects with ISBCS. Therefore, there is a clear need for evaluating evidence regarding the safety, effectiveness, and cost-effectiveness of ISBCS versus DSBCS. OBJECTIVES: To assess the safety of ISBCS compared to DSBCS in people with bilateral age-related cataracts and to summarise current evidence for the incremental resource use, utilities, costs, and cost-effectiveness associated with the use of ISBCS compared to DSBCS in people with bilateral age-related cataracts (primary objectives). The secondary objective was to assess visual and patient-reported outcomes of ISBCS compared to DSBCS in people with bilateral age-related cataracts. SEARCH METHODS: We searched CENTRAL (which contains the Cochrane Eyes and Vision Trials Register; 2021, Issue 5); Ovid MEDLINE; Ovid Embase; the ISRCTN registry; ClinicalTrials.gov; the WHO ICTRP; and DARE and NHS EED on the CRD Database on 11 May 2021. There were no language restrictions. We limited the searches to a date range of 2007 onwards. SELECTION CRITERIA: We included randomised controlled trials (RCTs) to assess complications, refractive outcomes, best-corrected distance visual acuity (BCDVA) and patient-reported outcome measures (PROMs) with ISBCS compared to DSBCS. We included non-randomised (NRSs), prospective, and retrospective cohort studies comparing ISBCS and DSBCS for safety assessment, because of the rare incidence of important adverse events. To assess cost-effectiveness of ISBCS compared to DSBCS, we included both full and partial economic evaluations, and both trial-based and model-based economic evaluations. DATA COLLECTION AND ANALYSIS: We used standard Cochrane methodological procedures and assessed risk of bias for NRSs using the ROBINS-I tool. For cost-evaluations, we used the CHEC-list, the CHEERS-checklist, and the NICE-checklist to investigate risk of bias. We assessed the certainty of evidence with the GRADE tool. We reported results for economic evaluations narratively. MAIN RESULTS: We included 14 studies in the review; two RCTs, seven NRSs, and six economic evaluations (one study was both an NRS and economic evaluation). The studies reported on 276,260 participants (7384 for ISBCS and 268,876 for DSBCS) and were conducted in Canada, the Czech Republic, Finland, Iran, (South) Korea, Spain (Canary Islands), Sweden, the UK, and the USA. Overall, we considered the included RCTs to be at 'high to some concerns' risk of bias for complications, 'some concerns' risk of bias for refractive outcomes and visual acuity, and 'high' risk of bias for PROMs. The overall risk of bias for NRSs was graded 'serious' regarding complications and 'serious to critical' regarding refractive outcomes.  With regard to endophthalmitis, we found that relative effects were estimated imprecisely and with low certainty, so that relative estimates were not reliable. Nonetheless, we found a very low risk of endophthalmitis in both ISBCS (1/14,076 participants) and DSBCS (55/556,246 participants) groups. Based on descriptive evidence and partially weak statistical evidence we found no evidence of an increased risk of endophthalmitis with ISBCS. Regarding refractive outcomes, we found moderate-certainty (RCTs) and low-certainty (NRSs) evidence there was no difference in the percentage of eyes that did not achieve refraction within 1.0 dioptre of target one to three months after surgery (RCTs: risk ratio (RR) 0.84, 95% confidence interval (CI) 0.57 to 1.26; NRSs: RR 1.02, 95% CI 0.60 to 1.75). Similarly, postoperative complications did not differ between groups (RCTs: RR 1.33, 95% CI 0.52 to 3.40; NRSs: 1.04, 95% CI 0.47 to 2.29), although the certainty of this evidence was very low for both RCTs and NRSs. Furthermore, we found low-certainty (RCTs) to very low-certainty (NRSs) evidence that total costs per participant were lower for ISBCS compared to DSBCS, although results of individual studies could not be pooled. Only one study reported on cost-effectiveness. This study found that ISBCS is cost-effective compared to DSBCS, but did not measure quality-adjusted life years using preferred methods and calculated costs erroneously. Finally, regarding secondary outcomes, we found limited evidence on BCDVA (data of two RCTs could not be pooled, although both studies individually found no difference between groups (very low-certainty evidence)). Regarding PROMs, we found moderate-certainty evidence (RCTs only) that there was no difference between groups one to three months after surgery (standardised mean difference -0.08, 95% CI -0.19 to 0.03). AUTHORS' CONCLUSIONS: Current evidence supports there are probably no clinically important differences in outcomes between ISBCS and DSBCS, but with lower costs for ISBCS. However, the amount of evidence is limited, and the certainty of the evidence was graded moderate to very low. In addition, there is a need for well-designed cost-effectiveness studies.

Altered ocular surface immune cell profile in patients with dry eye disease.

PURPOSE: Aberrant inflammation and immune dysregulation are known pathogenic contributors in dry eye disease (DED). Aim of the study was to determine the proportions of immune cell subsets on the ocular surface (OS) of DED patients. METHODS: 15 healthy controls (22 eyes) and 48 DED subjects (36 eyes with evaporative DED - EDED; 60 eyes with aqueous deficient DED - ADED) were included in the study. Tear break up time (TBUT), Schirmer's test 1 (ST1), corneal staining (CS) and ocular surface disease index (OSDI) scoring were recorded. OS wash was used to collect immune cells on the OS of study subjects. The cells immunophenotyped using flow cytometry include leukocytes, neutrophils, macrophages, natural killer-NK cells and T cell subsets (CD4; CD8; double positive-DP; gamma delta-γδ and NK T cells). RESULTS: Significantly higher proportions of leukocytes, neutrophils, CD4 T cells, CD8 T cells, DP T cells and CD4/CD8 T cells ratio were observed in EDED and/or ADED patients. Significantly higher proportions of neutrophils and lower proportions of NK cells were observed in ADED subjects with corneal staining compared to those without and controls. Neutrophils/NK cells ratio was significantly higher in EDED and ADED subjects compared to controls. Correlation analysis revealed pathological relationships between proportions of leukocytes, neutrophils, CD4 T cells and Neutrophil/NK cells ratio with DED clinical parameters. CONCLUSION: OS immune cell subset proportion changes in DED patients were associated with DED types and severity. The data suggests the potential for a new generation of therapies targeting immune cells on the ocular surface.

Laser Vision Correction with Q Factor Modification for Keratoconus Management.

PURPOSE: To evaluate the outcomes of corneal laser ablation with Q factor modification for vision correction in patients with progressive keratoconus. METHODS: In this prospective study, 50 eyes of 50 patients were divided into two groups based on Q factor (>-1 in Group I and ≤-1 in Group II). All patients underwent a detailed ophthalmic examination including uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), subjective acceptance and corneal topography using the Pentacam. The topolyzer was used to measure the corneal asphericity (Q). Ablation was performed based on the preoperative Q values and thinnest pachymetry to obtain a target of near normal Q. This was followed by corneal collagen crosslinking to stabilize the progression. RESULTS: Statistically significant improvement (p ≤ 0.05) was noticed in refractive, topographic, and Q values posttreatment in both groups. The improvement in higher-order aberrations and total aberrations were statistically significant in both groups; however, the spherical aberration showed statistically significant improvement only in Group II. CONCLUSION: Ablation based on the preoperative Q and pachymetry for a near normal postoperative Q value appears to be an effective method to improve the visual acuity and quality in patients with keratoconus.