Kovacs 2016.

Study characteristics
Patient Sampling	Retrospective, single‐centre (Hungary), case‐control study, which involved 135 eyes of people with bilateral keratoconus (keratoconus group), normal fellow eyes of people with unilateral keratoconus (fellow‐eye group), and eyes of refractive surgery candidates (control group).
Patient characteristics and setting	Keratoconus group: 60 eyes of 30 people with bilateral keratoconus Fellow‐eye group: 15 normal fellow eyes of people with unilateral keratoconus Control group: 60 eyes of 30 refractive surgery candidates
Index tests	Multilayer perception classifier (neural network) trained on bilateral data of index of height decentration.
Target condition and reference standard(s)	Keratoconus was diagnosed according to classic corneal biomicroscopic and topographic findings using the criteria of Rabinowitz: the existence of central protrusion of the cornea with Fleischer's ring, Vogt's striae, or both, by slit‐lamp examination in addition to the following topographic findings: a central keratometry (K) value > 47.2 D or an I‐S value > 1.4 D, or KISA% > 100%. Both eyes in the keratoconus group and the affected eye in the unilateral keratoconus group had abnormal keratoconus indices measured by a Scheimpflug camera (Pentacam HR). It is unclear who classified the cases. Classification was performed before the index test.
Flow and timing	All cases were included in the reference standard and index test. All data were included in a 2 × 2 table.
Comparative	Unclear whether different AI tests were developed and interpreted blind or independently and without knowledge of the results of each other. However, missing data and their causes were similar for each AI test.
Notes	Supported by OTKA NN106649 from the Hungarian Scientific Research Fund. The funder had no role in the study design, data collection, analysis, decision to publish, or preparation of the manuscript.
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?	Yes
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?	No
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?	Unclear
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?		Unclear 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