Rare disease diagnosis: A review of web search, social media and large-scale data-mining approaches
Bibliographic record
Abstract
Physicians and the general public are increasingly using web-based tools to find answers to medical questions. The field of rare diseases is especially challenging and important as shown by the long delay and many mistakes associated with diagnoses. In this paper we review recent initiatives on the use of web search, social media and data mining in data repositories for medical diagnosis. We compare the retrieval accuracy on 56 rare disease cases with known diagnosis for the web search tools google.com, pubmed.gov, omim.org and our own search tool findzebra.com. We give a detailed description of IBM's Watson system and make a rough comparison between findzebra.com and Watson on subsets of the Doctor's dilemma dataset. The recall@10 and recall@20 (fraction of cases where the correct result appears in top 10 and top 20) for the 56 cases are found to be be 29%, 16%, 27% and 59% and 32%, 18%, 34% and 64%, respectively. Thus, FindZebra has a significantly (p < 0.01) higher recall than the other 3 search engines. When tested under the same conditions, Watson and FindZebra showed similar recall@10 accuracy. However, the tests were performed on different subsets of Doctors dilemma questions. Advances in technology and access to high quality data have opened new possibilities for aiding the diagnostic process. Specialized search engines, data mining tools and social media are some of the areas that hold promise.
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 | 0.008 | 0.020 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.021 | 0.020 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".