Adaptive Bitrate Selection for Medical Video Compression Balancing Bandwidth Efficiency and Segmentation Quality
Bibliographic record
Abstract
Medical video compression is essential for reducing storage and bandwidth requirements while ensuring the preservation of diagnostic and segmentation quality, enabling efficient transmission and analysis in healthcare applications. The primary challenge in medical video compression lies in achieving significant data reduction without compromising critical visual details required for accurate diagnosis and segmentation tasks. In this work, we propose a framework for selecting the proper bitrate for medical video compression, focusing on maintaining segmentation performance with minimal impact on accuracy. By leveraging the SALI segmentation model and a diverse set of video-based and frame-based features, the framework predicts the Dice score for videos compressed at different bitrates, enabling informed decision-making. To evaluate our work, SUN-SEG dataset is used. Two approaches were evaluated: one prioritizing high compression ratios while maintaining an acceptable Dice score threshold, and the other optimizing segmentation accuracy. In the first approach, the framework achieved an average Dice score of 77.66 with a compression ratio of 37.06 and in the second approach, the Dice score improved to 81.23 with a more conservative compression ratio of 8.16. These results highlight the adaptability of the framework to varying application requirements, such as telemedicine and medical video storage. This framework ensures efficient medical video compression while preserving segmentation accuracy, optimizing bandwidth for telemedicine, storage, remote diagnostics, and surgical video analysis, ultimately enhancing AI-assisted clinical decision-making.
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 | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".