Skip to main content

Table 3 Radiomics quality score and subgroup analysis according to journal type and biomarker study design

From: A systematic review reporting quality of radiomics research in neuro-oncology: toward clinical utility and quality improvement using high-dimensional imaging features

Radiomics quality scoreMedian score (Interquartile range)Clinical
(n = 20)
Imaging
(n = 31)
PDiagnostic
(n = 34)
Prognostic /Predictive† (n = 17)P
Total 36 points11.00 (3–12.5)11.5 (3–13)10 (3.5–11.5).5710 (3–11)12 (6–14).07
Domain 1: Protocol quality and stability in image and segmentation (0 to 5 points)1 (1–2)1.5 (1–2)1 (1–2) 1 (1–2)1 (1–2) 
Protocol quality (2)1 (1–1.5)1 (1–1.25)1 (1–1).951 (1–1)1 (1–1.75).662
Test-retest (1)0 (0–0)0 (0–0)0 (0–0).2280 (0–0)0 (0–0).169
Phantom study (1)0 (0–0)0 (0–0)0 (0–0)NA0 (0–0)0 (0–0)NA
Multiple segmentation (1)0 (0–1)0 (0–0.25)0 (0–1).760 (0–1)0 (0–1).277
Domain 2: Feature selection and validation (−8 to 8 points)5 (−2–6)5 (−2–6)5 (−2–5) 5 (− 2–6)5 (− 2–6) 
Feature reduction or adjustment of multiple testing (−3 or 3)3 (3–3)3 (3–3)3 (3–3).8493 (3–3)3 (3–3)1
Validation (−5, 2, 3, 4, or 5)2 (− 5–3)2 (− 5–3)2 (− 5–2).9752 (−5–3)2 (− 5–3).833
Domain 3: Biologic/clinical validation and utility (0 to 6 points)2 (1–3)2 (1–3)2 (1–2) 2 (1–2)3 (2–4) 
Non-radiomics features (1)1 (0–1)1 (0–1)1 (0–1).7991 (0–1)1 (0–1).159
Biologic correlates (1)1 (0–1)1 (0–1)1 (0–1)11 (1–1)1 (0–1).001
Comparison to ‘gold standard’ (2)0 (0–2)0 (0–2)0 (0–2).970 (0–0)2 (2–2)<.001
Potential clinical utility (2)0 (0–0)0 (0–0)0 (0–0).440 (0–0)0 (0–0).50
Domain 4: Model performance index (0 to 5 points)2 (1–2)2 (1–3)2 (1–3) 2 (1–2)2 (1–3) 
Discrimination statistics (2)2 (1–2)1.5 (1–2)2 (1–2).752 (1–2)2 (1–2).99
Calibration statistics (2)0 (0–0)0 (0–0)0 (0–0).840 (0–0)0 (0–1).02
Cut-off analysis (1)0 (0–0.5)0 (0–0)0 (0–1).480 (0–0)1 (0–1).001
Domain 5: High level of evidence (0 to 8 points)0 (0–0)0 (0–0)0 (0–0) 0 (0–0)0 (0–0) 
Prospective study (7)0 (0–0)0 (0–0)0 (0–0).080 (0–0)0 (0–0).634
Cost-effective analysis (1)0 (0–0)0 (0–0)0 (0–0)NA0 (0–0)0 (0–0)NA
Domain 6: Open science and data (0 to 4 points)0 (0–0)0 (0–0)0 (0–0).030 (0–0)0 (0–0)1
  1. † Two studies overlap in both diagnostic and prognostic marker were classified as prognostic use