Real World Imaging Data: Opportunities and Challenges (Preprint)
Bibliographic record
Abstract
Unlabelled: The amount of data generated in clinical practice is increasing substantially. This has benefited the use of real-world data for real-world evidence in biomedical research. While early real-world evidence efforts focused on structured electronic health record and insurance claims data, advances in analytical methods, such as AI, are expected to further enhance the ability to obtain deeper insights from real-world data. Most recently, real-world imaging data (RWiD) has emerged as a novel and valuable resource. RWiD is enabled by improvements in imaging infrastructure, data standardization, and deidentification technologies. Medical imaging is essential at multiple stages of clinical care, ranging from screening to posttreatment assessment and surveillance. Medical imaging has become a key component of patient management, as it augments clinical decision-making across many medical specialties. RWiD is the retrospective collection of routinely gathered clinical imaging data. Using only radiology reports results in limited information compared to datasets that contain the actual images. Radiology reports primarily focus on clinical decision-making rather than research purposes. Therefore, the actual images add value to real-world datasets. However, using RWiD is challenging due to complex deidentification and harmonization, as well as requirements for file storage, file transfer, and computation. This editorial paper provides an educational overview, including the background, challenges, and opportunities of RWiD, and offers examples of RWiD applications that benefit life sciences and biopharmaceutical use cases.
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 | 0.011 | 0.044 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.006 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.012 | 0.009 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.017 | 0.011 |
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".