Chandapura 2019.

Study characteristics
Patient Sampling	Retrospective case‐control study, involving 439 eyes from 2 centres (India and Brazil). Comparison between 4 AI models.
Patient characteristics and setting	Keratoconus Form fruste keratoconus Very asymmetric eyes with normal topography (VAE‐NT)
Index tests	Random forest models based on Pentacam (Oculus GmbH, Wetzlar, Germany) or OCT parameters (OCT topography of the Bowman's layer)
Target condition and reference standard(s)	Keratoconus group: corneas with significant inferior steepening, asymmetric astigmatism, and corneal thinning on both OCT (RTVue, Optovue Inc., Irvine) and Scheimpflug (Pentacam, OCULUS Optikgerate). Form fruste keratoconus: very mild localized steepening and suspicious anterior corneal surface topography on both devices. Very asymmetric eyes with normal topography (VAE‐NT): fellow eyes of people with highly asymmetric keratoconus. Examination of topographies of the anterior surface was performed by only 1 experienced refractive surgeon, who was masked to the information (disease present in 1 or both eyes) about the participants and the eyes. Classification was performed before the index tests.
Flow and timing	All cases were included in the reference standard and index test. All data were included in a 2 × 2 table.
Comparative	It is unclear whether different AI tests were developed and interpreted blind or independently and without knowledge of the results of each other. Missing data and their causes were similar for each AI test.
Notes	Indo‐German Science and Technology Center, Grant/Award Number: SIBAC
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient selection
Was a consecutive or random sample of patients enrolled?	No
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?	Yes
If a threshold was used, was it pre‐specified?	Unclear
Was the model designed in an appropriate manner?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low 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.	Unclear
Are the proportions and reasons for missing data similar for all index tests?	Yes
		Unclear risk