Accardo 2002.

Study characteristics
Patient Sampling	Single‐centre, retrospective, case‐control study including 396 corneal topographic maps (396 eyes, 198 participants), obtained with a videokeratoscope (EyeSys, EyeSys Vision, Houston, Texas), selected from the cases of keratoconus or other conditions recorded over a 3‐year period at the centre, in Italy.
Patient characteristics and setting	Keratoconus cases (110 images, 110 eyes, 55 participants): mild and moderate keratoconus severity, with a sagittal cone apex power < 53 D, where no clinical sign was present (early keratoconus) or only the Vogt's striae were detected (mild or moderate keratoconus) Others (166 images of 166 eyes, 83 participants): various bilateral non‐keratoconus conditions, congenital astigmatism, contact lens corneal warpage
Index tests	A neural network using as input the parameters of both eyes of the same subject and as output the 3 categories of clinical classification (normal, keratoconus, other alterations) for each subject, a low number of neurons in the hidden layer (< 10), and a learning rate of 0.1.
Target condition and reference standard(s)	Keratoconus group comprised cases of mild and moderate severity with a sagittal cone apex power < 53 D, with no clinical sign (early keratoconus) or only the Vogt's striae, and keratoconus suspects that met one of the following criteria. The pathology was already present in the other eye with the same topographic pattern A family history of keratoconus was present The apex progression of the corneal protrusion was >1 D after 1 or 2 years of follow‐up The maps were classified before the analysis with the algorithm, the number of observers is unclear.
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 partially supported by the University of Trieste (MURST60%) and by Burlo Garofolo Hospital in Trieste, Italy.
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?	Yes
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?	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