Automated and Interpretable Detection of Hippocampal Sclerosis in temporal lobe epilepsy: AID-HS
Bibliographic record
Abstract
Abstract Hippocampal Sclerosis (HS) can elude visual detection on MRI scans of patients with temporal lobe epilepsy (TLE), causing delays in surgical treatment and reducing the likelihood of postsurgical seizure-freedom. We developed an open-source software that (1) detects HS from structural MRI scans, (2) generalises across a heterogeneous multicentre cohort of children and adults, and (3) generates individualised predictions for clinical evaluation. We included a cohort of 363 participants (152 patients with HS, 90 disease controls with focal cortical dysplasia, and 121 healthy controls) from four epilepsy centres in the UK, North America, and China. We used the open-source software HippUnfold to extract morphological surface-based features and volumes of the hippocampus from T1w MRI scans. We compared pathological hippocampal morphology in patients with HS to normative growth charts generated from healthy controls, and to the contralateral hippocampi in patients with HS. HS was characterised by decreased volume, thickness and gyrification as well as increased mean and intrinsic curvature. A logistic regression classifier trained on these features detected 90.1% of HS patients, and accurately lateralised 97.4% of the HS cohort. Crucially, in patients with MRI-negative histopathologically confirmed HS, the classifier detected HS in 79.2% (19/24) and accurately lateralised the lesions in 91.7% (22/24). The Automated and Interpretable Detection of Hippocampal Sclerosis classifier (AID-HS) was packaged into an open-source pipeline, which detects and lateralises HS and generates individualised patient reports that characterise hippocampal developmental abnormalities. AID-HS is capable of accurately detecting and lateralising HS in a large, heterogeneous, multi-centre, cohort of paediatric and adult patients with diagnostically challenging HS. Moreover, by offering transparent, robust and interpretable results, AID-HS can support the presurgical evaluation of patients with suspected TLE.
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 | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 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".