Can Machine Learning of Magnetic Resonance Imaging Textural Features Differentiate Intra- and Extra-Axial Brain Tumours? A Feasibility Study
Bibliographic record
Abstract
Introduction: Determining the origin of intracranial lesions can be challenging.This study aimed to assess the feasibility of a machine learning model in distinguishing intra-axial (IA) from extra-axial (EA) brain tumours using magnetic resonance imaging (MRI).Methods: We retrospectively reviewed 92 consecutive adult patients (age >18 years) with newly diagnosed solitary brain lesions who underwent contrast-enhanced brain MRI at our institution from January 2017 to December 2018.Tumour volumes of interest (VOIs) were manually segmented on both T2-weighted (T2W) and T1-weighted (T1W) post-contrast images.An XGBoost machine learning algorithm was used to generate classification models based on textural features extracted from the segmented VOIs, with histopathology as the reference standard.Results: Among the 92 lesions analysed, 70 were IA and 22 were EA.The area under the receiver operating characteristic curve for identifying IA tumours was 0.91 (95% confidence interval [95% CI] = 0.89-0.93)for the T1W post-contrast model, 0.81 (95% CI = 0.78-0.84)based on T2WI model, and 0.92 (95% CI = 0.90-0.94)for the combined model.All models demonstrated high sensitivity (>90%) for identifying intra-axial tumours, though specificity was lower (39%-64%).Despite this, models achieved acceptable levels of accuracy (>80%) and precision (>88%).Conclusion: This preliminary study demonstrates the feasibility of a machine learning classification model for differentiating IA from EA tumours using MRI textual features.While sensitivity was high, specificity was limited, likely due to the class imbalance.Further studies with balanced datasets and external validation are warranted.
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 | 0.013 | 0.047 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".