SynthesisFrontiers in oncology2026
Comparative diagnostic accuracy of radiomics, deep learning, and hybrid AI for invasiveness stratification of pulmonary ground-glass nodules: a systematic review and meta-analysis.
Synthesis in Frontiers in oncology, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
8 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
Background: Accurate preoperative stratification of ground-glass nodules (GGNs) into invasive versus non-invasive adenocarcinomas is essential for optimizing surgical management and preventing overtreatment. Radiomics, deep learning, and hybrid models have shown potential for noninvasive histopathologic prediction, but their comparative diagnostic performance and generalizability remain unclear. This study systematically compares the diagnostic accuracy, methodological quality, and clinical applicability of radiomics, deep learning, and hybrid AI approaches for differentiating invasive from non-invasive lung adenocarcinomas manifesting as GGNs. Methods: A comprehensive search of PubMed, Embase, Cochrane Library, and Web of Science was conducted through 24 September 2025 (PROSPERO CRD42025115051). Twenty-two studies (9,768 nodules) met inclusion criteria, evaluating AI models with exact segmentation or embedded nodule localization and histopathologic confirmation. Results: Hybrid models demonstrated higher pooled diagnostic performance (AUC 0.94, sensitivity 0.91 [95% CI 0.88-0.93], specificity 0.87 [0.83-0.90]) than deep learning (AUC 0.92) and radiomics (AUC 0.89). At a 20% pre-test probability, a positive hybrid AI result increased post-test probability of invasiveness to approximately 59%, while a negative result decreased it to approximately 5% (approximate pooled LR + 6.0; LR- 0.22). However, effect estimates showed substantial heterogeneity (I² > 60%) and publication bias suggested sensitivity inflation (3-5%). Most studies were single-center and retrospective, with 82% conducted in East Asia and limited external validation. Conclusion: Hybrid AI showed higher pooled accuracy compared with radiomics or deep learning alone, but the certainty of evidence is low-moderate due to inconsistency, publication bias, and restricted population diversity. AI-based GGN risk stratification should currently complement, rather than replace, multidisciplinary clinical decision-making until validated in prospective multicenter trials.
Indexed as
Identifiers
What Socratic holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the Socratic graph.