Automated Detection of Bowel Preparation Scoring and Adequacy With Deep Convolutional Neural Networks
Bibliographic record
Abstract
Abstract Introduction Adequate bowel preparation is integral to effective colonoscopy. Inadequate bowel preparation has been associated with reduced adenoma detection rate and increased post-colonoscopy colorectal cancer (PCCRC). As a result, the USMSTF recommends early interval reevaluation for colonoscopies with inadequate bowel preparation. However, bowel preparation documentation is highly variable with subjective interpretation. In this study, we developed deep convolutional neural networks (DCNN) to objectively ascertain bowel preparation. Methods Bowel preparation scores were assigned using the Boston Bowel Preparation Scale (BBPS). Bowel preparation adequacy and inadequacy were defined as BBPS ≥2 and BBPS <2, respectively. A total of 38523 images were extracted from 28 colonoscopy videos and split into 26966 images for training, 7704 for validation, and 3853 for testing. Two DCNNs were created using a Densenet-169 backbone in PyTorch library evaluating BBPS score and bowel preparation adequacy. We used Adam optimiser with an initial learning rate of 3 × 10−4 and a scheduler to decay the learning rate of each parameter group by 0.1 every 7 epochs along with focal loss as our criterion for both classifiers. Results The overall accuracy for BBPS subclassification and determination of adequacy was 91% and 98%, respectively. The accuracy for BBPS 0, BBPS 1, BBPS 2, and BBPS 3 was 84%, 91%, 85%, and 96%, respectively. Conclusion We developed DCCNs capable of assessing bowel preparation adequacy and scoring with a high degree of accuracy. However, this algorithm will require further research to assess its efficacy in real-time colonoscopy.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".