Almeida 2022.

Study characteristics
Patient Sampling	Multicentre, case‐control study. All participants were examined at the Visum Eye Center and Rio Claro Eye Institute between January 2012 and January 2019. Exclusion criteria were a history of ocular trauma, corneal scarring, and neurotrophic keratopathy. Normal Eyes: 2296 people who underwent LASIK or photorefractive keratectomy and were stable after at least 18 months of follow‐up. Very Asymmetric Eyes With Normal Topography and Tomography Group: 187 eyes of 187 people with very asymmetric eyes with normal topography (VAE‐NT) in 1 eye and frank ectasia (VAE‐E) in the fellow eye Ectasia Group: 410 people (1 eye each) with bilateral clinical keratoconus
Patient characteristics and setting	410 eyes with keratoconus 2296 healthy corneas 187 very asymmetric eyes with a normal topography of the cornea.
Index tests	Multiple logistic regression analysis (MLRA) is based on the logistic function that bounds its output within the range of 0 to 1. To build the algorithm extracted from MLRA, 22 variables were used.
Target condition and reference standard(s)	All eyes were examined by rotating Scheimpflug corneal and anterior segment tomography (Pentacam HR, Oculus Optikgerate GmbH). Image quality was checked to ensure that only cases with acceptable‐quality images were included. All cases were reviewed by an experienced fellowship‐trained corneal specialist (G.C.A.J.) for correct classification into keratoconus and VAE‐NT groups. Objective criteria for considering normal topography included objective front surface curvature metrics derived from Pentacam. Normal topography criteria were rigorously considered based on the objective criteria of a maximum keratometry curvature (Kmax; steepest front keratometry) of < 47.2 D, a paracentral I‐S asymmetry value at 6 mm (3 mm radii) of < 1.45, and a keratoconus percentage index score of < 60. An objective criterion for normal tomography criteria was adopted for the control group and the VAE‐NT group, and the maximum values were 3.8 mm for anterior chamber depth, 4 mm for front apical elevation, 5 mm for front corneal elevation at the thinnest point, and 12 mm for front corneal elevation in the central 4.0 mm. The corresponding posterior elevation values were 7 mm, 13 mm, and 25 mm.
Flow and timing	All cases were included in the reference standard and index test. All data were included in a 2 × 2 table.
Comparative	Not applicable
Notes	This work was supported by the Sao Paulo State Research Support Foundation (FAPESP, grant nos: 2015/17226‐7 and 2019/04475‐0) and the National Council for Scientific and Technological Development (CNPq, grant no: 306808/2018‐8.). The funding organizations had no role in design or conduct of this research, and they have no related commercial interests.
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient selection
Was a consecutive or random sample of patients enrolled?	Unclear
Was a case‐control design avoided?	No
Did the study avoid inappropriate exclusions?	No
Could the selection of patients have introduced bias?		High risk
Are there concerns that the included patients and setting do not match the review question?			High
DOMAIN 2: Index test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Unclear
If a threshold was used, was it pre‐specified?	Unclear
Was the model designed in an appropriate manner?	Unclear
Could the conduct or interpretation of the index test have introduced bias?		Unclear risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Low concern
DOMAIN 3: Reference standard
Is the reference standard likely to correctly classify the target condition?	No
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		High risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and timing
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk
DOMAIN 5: Comparative
Were different AI tests were developed and interpreted without knowledge of each other.
Are the proportions and reasons for missing data similar for all index tests?