Implementing a large-scale multicentric study for evaluation of lossy JPEG and JPEG2000 medical image compression: challenges and rewards
Bibliographic record
Abstract
At modest compression ratios, lossy compression schemes allow substantial image size reduction without a significant loss in visual information. This is a consequence of the coding engines' transformation (such as the Discrete Cosine Transfomation (DCT) and the Discrete Wavelet Transform (DWT) in combination with quantization and truncation operations which all exploit the characteristics of the human visual system to achieve file-size reduction. The objective of our study was to determine levels of lossy compression that can be confidently used in diagnostic imaging. We conducted an extensive clinical evaluation using a standardized methodology incorporating two recognized evaluation techniques: Diagnostic Accuracy with Receiver Operating Characteristic (ROC) Analysis and Original-Revealed Forced Choice. Images covering 5 modalities and 7 anatomical regions were compressed at 3 different levels using JPEG and JPEG 2000 compression algorithms. To enable radiologists across Canada to evaluate images for our study, we developed a dedicated software application that was synchronized to a centralized server; which allowed results were reported, in real-time, to the central database via the Internet. In order to obtain findings that were relevant to everyday clinical evaluation, images were not viewed under a strict laboratory environment, but rather they were read under typical viewing conditions that comply with current standards of practice. We present here the methodology and specific technology developed for the purpose of this study, we explain the specific problems that we have encountered during the implementation and we give preliminary results. Our preliminary findings suggest that the most appropriate compression algorithm and compression ratios are largely dependent on the image specifics including the type/ modality and anatomical region studied.
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 | 0.035 | 0.026 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 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".