Semi-Supervised Multi-Sequence Glioblastoma MRI Radiogenomics for Prediction of IDH Mutation Status: Improved Robustness to Limited Labels and SHAP Interpretations
Bibliographic record
Abstract
Background: Glioblastoma (GBM) is an aggressive brain tumor, with IDH mutation status as a key prognostic biomarker.Traditional IDH testing requires invasive biopsies, highlighting the need for non-invasive alternatives.MRI-based radiogenomics features coupled with machine learning show promise, but past studies were mostly single-center-based and rarely used semisupervised learning (SSL) to exploit unlabeled data. Methods:We analyzed MRI sequences T1, T2-weighted, contrast-enhanced T1 (T1CE) and FLAIR from 1,329 patients across eight centers, with IDH labels available for 1,061 cases.Radiomics features (n=1,223 per case) were extracted using PyRadiomics with Laplacian of Gaussian and wavelet filters.Both supervised learning (SL) and SSL (via pseudo-labeling) were implemented, incorporating 38 feature selection/attribute extraction and 24 classifiers.Five-fold cross-validation was performed on UCSF-PDGM and UPENN datasets, with external validation on IvyGAP, TCGA-LGG and TCGA-GBM.SHAP analysis quantified feature importance.Results: Multimodal MRI (T1+T2+T1CE+FLAIR) provided the strongest performance, outperforming single-sequence models.The best SSL model (involving Recursive Feature Elimination (RFE) + SVM) achieved 0.93±0.01cross-validation and 0.75±0.02external accuracy, while the best SL model (RFE+Complement Naive Bayes (CNB)) reached 0.90±0.02and 0.80±0.006,respectively.SSL also demonstrated greater stability with lower sensitivity to dataset size compared to SL, maintaining robust performance in data-limited conditions.SHAP analysis showed SSL amplified the discriminative value of first-order statistics of Root Mean Square (FO_RMS) (T1CE) and wavelet-based metrics, strengthening biomarker interpretability.Conclusion: SSL improves accuracy, efficiency and interpretability in MRI-based IDH prediction, remaining less sensitive to data size while reinforcing multimodal fusion as the most reliable, scalable strategy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.007 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".