Lavric 2021.

Study characteristics
Patient Sampling	Retrospective case‐control study. Pentacam data (Oculus GmbH, Wetzlar, Germany) obtained from people screened for keratoconus disease in Brazil. Elevation, topography, and pachymetry parameters were obtained from 5881 eyes of 2800 participants.
Patient characteristics and setting	Participant characteristics and setting are not clearly described in the study. Data seems to originate from a larger data set which included people with keratoconus and healthy controls.
Index tests	Support vector machine that uses elevation, topography or pachymetry parameters obtained from the raw data of the Pentacam to detect keratoconus. The workflow for the development of the algorithm was as follows: splitting the initial data set in elevation, topography and pachymetry data sets; data cleaning and elimination; feature selection; machine learning validation; and performance evaluation. It is unclear whether different data were used for testing and validating the model.
Target condition and reference standard(s)	The target condition was keratoconus; however, the article provided no definition. Tomography images of the Pentacam were used in this study. It is unclear who interpreted the images and made the diagnosis; however, the diagnosis was made before the algorithms analysed the images.
Flow and timing	The article did not describe the reference standard, nor did it describe whether all participants received the same reference standard. All data were included in a 2 × 2 table.
Comparative	In total, 6 algorithms were developed, tested, and compared: decision tree, discriminant naïve Bayes, support vector machine, k‐nearest neighbour, and ensemble.
Notes	This work was supported in part by a grant from the Romanian Ministry of Research and Innovation, CCCDI‐UEFISCDI, within PNCDI III, under Project PN‐III‐P2‐2.1‐PTE‐2019‐0642, and in part by the Romania National Council for Higher Education Funding, CNFIS, under Project CNFIS‐FDI‐2021‐0357.
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?	Unclear
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?	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?	Unclear
Were the reference standard results interpreted without knowledge of the results of the index tests?	Unclear
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?	Unclear
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