Large language models in methodological quality evaluation of radiomics research based on METRICS: ChatGPT vs NotebookLM vs radiologist

European Journal of Radiology, Journal Year: 2025, Volume and Issue: unknown, P. 111960 - 111960

Published: Jan. 1, 2025

Language: Английский

---

Radiomics for differentiating radiation-induced brain injury from recurrence in gliomas: systematic review, meta-analysis, and methodological quality evaluation using METRICS and RQS

European Radiology, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 12, 2025

Abstract Objective To systematically evaluate glioma radiomics literature on differentiating between radiation-induced brain injury and tumor recurrence. Methods Literature was searched PubMed Web of Science (end date: May 7, 2024). Quality eligible papers assessed using METhodological RadiomICs Score (METRICS) Radiomics (RQS). Reliability quality scoring tools were analyzed. Meta-analysis, meta-regression, subgroup analysis performed. Results Twenty-seven included in the qualitative assessment. Mean average METRICS score RQS percentage across three readers 57% (SD, 14%) 16% 12%), respectively. Score-wise inter-rater agreement for ranged from poor to excellent, while demonstrated moderate excellent agreement. Item-wise both tools. Meta-analysis 11 studies yielded an estimated area under receiver operating characteristic curve 0.832 (95% CI, 0.757–0.908), with significant heterogeneity ( I 2 = 91%) no statistical publication bias p 0.051). Meta-regression did not identify potential sources heterogeneity. Subgroup revealed high all subgroups, lowest at 68% proper validation higher scores. Statistical generally significant, except 0.044). However, most (26/27; 96%) primary meta-analysis (10/11; reported positive effects radiomics, indicating non-statistical bias. Conclusion While a good performance noted results should be interpreted cautiously due heterogeneity, bias, issues thoroughly examined this study. Key Points Question Radiomic distinguishing recurrence lacks systematic reviews meta-analyses that assess methodological radiomics-specific . Findings are encouraging, there substantial toward findings, notable concerns regarding Clinical relevance need cautious interpretation problems detected during (e.g., suboptimal quality, bias), which may help explain why has yet been translated into clinical practice Graphical

Language: Английский

---

Reproducibility of methodological radiomics score (METRICS): an intra- and inter-rater reliability study endorsed by EuSoMII

European Radiology, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 19, 2025

Abstract Objectives To investigate the intra- and inter-rater reliability of total methodological radiomics score (METRICS) its items through a multi-reader analysis. Materials methods A 12 raters with different backgrounds experience levels were recruited for study. Based on their level expertise, randomly assigned to following groups: two groups, intra-rater where each group included one without preliminary training session use METRICS. Inter-rater groups assessed all 34 papers, while completed assessment 17 papers twice within 21 days time, “wash out” period 60 in between. Results was poor moderate between 1 (without training; ICC = 0.393; 95% CI 0.115–0.630; p 0.002), 2 (with 0.433; 0.127–0.671; 0.002). The analysis excellent 9 12, good 8 10, rater 7, 11. Conclusion METRICS relatively good, low. This highlights need further efforts achieve common understanding items, as well resources consisting explanations, elaborations, examples improve reproducibility enhance usability robustness. Key Points Questions Guidelines scoring tools are necessary quality research; however, application these is challenging less experienced . Findings Intra-rater high across regardless or previous training, generally Clinical relevance proper reporting research closing gap clinical implementation. There offering robustness guidelines

Language: Английский

---

Prediction of molecular subtypes of endometrial cancer patients on the basis of intratumoral and peritumoral radiomic features from multiparametric MR images

European Journal of Radiology, Journal Year: 2025, Volume and Issue: unknown, P. 112110 - 112110

Published: April 1, 2025

Language: Английский

---