Viability of machine learning to reduce workload in systematic review\n screenings in the health sciences: a working paper
Bibliographic record
Abstract
Systematic reviews, which summarize and synthesize all the current research\nin a specific topic, are a crucial component to academia. They are especially\nimportant in the biomedical and health sciences, where they synthesize the\nstate of medical evidence and conclude the best course of action for various\ndiseases, pathologies, and treatments. Due to the immense amount of literature\nthat exists, as well as the output rate of research, reviewing abstracts can be\na laborious process. Automation may be able to significantly reduce this\nworkload. Of course, such classifications are not easily automated due to the\npeculiar nature of written language. Machine learning may be able to help. This\npaper explored the viability and effectiveness of using machine learning\nmodelling to classify abstracts according to specific exclusion/inclusion\ncriteria, as would be done in the first stage of a systematic review. The\nspecific task was performing the classification of deciding whether an abstract\nis a randomized control trial (RCT) or not, a very common classification made\nin systematic reviews in the healthcare field. Random training/testing splits\nof an n=2042 dataset of labelled abstracts were repeatedly created (1000 times\nin total), with a model trained and tested on each of these instances. A Bayes\nclassifier as well as an SVM classifier were used, and compared to non-machine\nlearning, simplistic approaches to textual classification. An SVM classifier\nwas seen to be highly effective, yielding a 90% accuracy, as well as an F1\nscore of 0.84, and yielded a potential workload reduction of 70%. This shows\nthat machine learning has the potential to significantly revolutionize the\nabstract screening process in healthcare systematic reviews.\n
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 | 0.414 | 0.692 |
| Meta-epidemiology (narrow) | 0.004 | 0.004 |
| Meta-epidemiology (broad) | 0.008 | 0.009 |
| Bibliometrics | 0.007 | 0.010 |
| Science and technology studies | 0.002 | 0.003 |
| Scholarly communication | 0.008 | 0.019 |
| Open science | 0.004 | 0.005 |
| Research integrity | 0.004 | 0.005 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".