OP07.05: Artificial intelligence to detect pouch of Douglas obliteration from magnetic resonance imaging by combining unpaired transvaginal ultrasound videos
Bibliographic record
Abstract
Endometriosis is a common disorder that has many characteristics, including the pouch of Douglas (POD) obliteration, which can be diagnosed using transvaginal ultrasound (TVUS) and magnetic resonance imaging (MRI). TVUS and MRI are complementary, non-invasive endometriosis diagnosis imaging techniques, but patients are usually not scanned using both modalities. Furthermore, it is generally more challenging to detect POD obliteration from MRI than TVUS. To mitigate this imbalance, this research aimed to create a knowledge distillation training algorithm to improve the POD obliteration detection from MRI, by leveraging the detection results from unpaired TVUS data. A machine learning model was pretrained using 8,984 unlabelled MRIs of the female pelvis. The algorithm was then fine tuned with 89 labelled MRI, performed specifically for investigation of endometriosis. Following this, 749 unpaired labelled TVUS videos demonstrating the uterine ‘sliding sing’ were introduced, to distil the knowledge from the teacher TVUS POD obliteration detector to train the student MRI model. The inclusion of the unpaired TVUS videos increased detection of POD obliteration from MRI with our machine learning model. This multimodal analysis method improved the Area Under the Curve (AUC) from 65.0% to 90.6%. Pretraining using digital data from different imaging modalities can improve the diagnosis of POD obliteration. By using a multimodal approach to creating machine learning models, diagnostic accuracy of POD obliteration was able to be improved by utilising unpaired eTVUS videos to pretrain MRI models.
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
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".