Development and validation of LightGBM algorithm for optimizing of Helicobacter pylori antibody during the minimum living guarantee crowd based gastric cancer screening program in Taizhou, China
Bibliographic record
Abstract
Gastric cancer continues to be a significant health concern in China, with a high incidence rate. To mitigate its impact, early detection and treatment is key. However, conducting large-scale endoscopic gastric cancer screening is not feasible in China. Instead, a more appropriate approach would be to initially screen high-risk groups and follow up with endoscopic testing as needed. We conducted a study on 25,622 asymptomatic participants aged 45-70 years from a free gastric cancer screening program in the Taizhou city government's Minimum Living Guarantee Crowd (MLGC) initiative. Participants completed questionnaires, blood tests, and underwent gastrin-17 (G-17), pepsinogen I and II (PGI and PGII), and H. pylori IgG antibody (IgG) assessments. Using the light gradient boosting machine (lightGBM) algorithm, we developed a predictive model for gastric cancer risk. In the full model, F1 score was 2.66%, precision was 1.36%, and recall was 58.14%. In the high-risk model, F1 score was 2.51%, precision was 1.27%, and recall was 94.55%. Excluding IgG, the F1 score was 2.73%, precision was 1.40%, and recall was 68.62%. We conclude that H. pylori IgG appears to be able to be excluded from the prediction model without significantly affecting its performance, which is important from a health economic point of view. It suggests that screening indicators can be optimized, and expenditures reduced. These findings can have important implications for policymakers, as we can focus resources on other important aspects of gastric cancer prevention and control.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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".