Rate of Agreement for Manual and Automated Techniques for Determination of New T2 Lesions in Children with Multiple Sclerosis and Acute Demyelination (P2.242)
Bibliographic record
Abstract
OBJECTIVE: To evaluate inter-rater variability in new T2 lesion determination based on a manual lesion identification method and an automated probabilistic segmentation technique. BACKGROUND: Determination of new lesions on serial magnetic resonance imaging (MRI) is important in multiple sclerosis (MS) diagnosis, monitoring and clinical trials. METHODS: Scans were acquired according to a standard protocol at incident demyelinating attack, 3, 6, and 12 months and annually under a national prospective study. New T2 lesions were manually identified independently by 3 experts (2 pediatric neuroradiologists, 1 imaging scientist) using a standard scoring tool. PDw, T2w, and FLAIR sequences were used. Readers were blinded to clinical information. Automated new T2 lesion segmentation was performed on the same cohort using an optimized, computer-assisted method in which a voxelwise Bayesian and lesion-level random forest classification is performed on reference, follow-up and difference images. Active scans were defined as having 蠅1 new T2 lesion. New T2 lesion counts were categorized into 0, 1, 2 and 蠅3 lesions. Rate of agreement was assessed for scan activity and new T2 lesion count. RESULTS: From 19 children included (10 MS, 9 monophasic ADS), 130 scans were analyzed; 12 scans not meeting quality specifications were excluded. A mean of 101 new lesions were identified on all scans (expert readers: R1=87, R2=90, R3=106; automated method: A1=122). Mean rate of agreement on scan activity between the automated method and expert readers was 87% (R1-A1=91%, R2-A1=83%, R3-A1=88%), compared to 91% between readers. Mean agreement on lesion count of the automated method with expert readers was 82% (R1-A1=85%, R2-A1=77%, R3-A1=84%) compared to 85% between readers; when zero counts were removed, the mean was was 53% (R1-A1=61%, R2-A1=39%, R3-A1=59%), compared to 61% between readers. CONCLUSIONS: Agreement is high between manual and automated methods for new T2 lesion determination. Overall, the automated method is more sensitive than the manual readers and identifies a higher number of lesions. Study Supported by: Canadian Multiple Sclerosis Scientific Research Foundation
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 | 0.049 | 0.093 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".