Early detection of non-small cell lung cancer: an electronic health record data-driven approach
Bibliographic record
Abstract
BACKGROUND: Specific patient characteristics increase the risk of cancer, necessitating personalized healthcare approaches. For high-risk individuals, tailored clinical management ensures proactive monitoring and timely interventions. Electronic health record (EHR) data are crucial for supporting these personalized approaches, improving cancer prevention and early diagnosis. We leverage EHR data and build a prediction model for early detection of non-small cell lung cancer (NSCLC). METHODS: We utilize data from Mass General Brigham's EHR and implement a three-stage ensemble learning approach. Initially, we generate risk scores using multivariate logistic regression in a self-control and case-control design to distinguish between cases and controls. Subsequently, these risk scores are integrated and calibrated using a prospective Cox model to develop the risk prediction model. RESULTS: We identified 127 EHR-derived features predictive for early detection of NSCLC. The highly predictive features include smoking, relevant lab test results, and chronic lung diseases. The predictive model reached area under the receiver operating characteristic (ROC) curve (area under the curve, AUC) of 0.801 (positive predictive value (PPV) 0.0173 with specificity 0.02) for predicting 1-year NSCLC risk in a population aged 18 and above, and AUC of 0.757 (PPV 0.0196 with specificity 0.02) in a population aged 40 and above. CONCLUSIONS: This study identified EHR-derived features which are predictive of early NSCLC diagnosis. The developed risk prediction model exhibits superior performance for early detection of NSCLC compared to a baseline model that only relies on demographic and smoking information, demonstrating the potential of incorporating EHR-derived features for personalized cancer screening recommendations and early detection.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.002 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".