Abstract A040: Digital Controls: An Efficient, Robust, and Privacy-Preserving Bayesian Tool for Adaptive Information Borrowing from External Data
Bibliographic record
Abstract
Abstract Randomized controlled trials (RCTs), also known as A/B testing, are widely regarded as the gold standard for evaluating average treatment effects by comparing treatment and control outcomes. However, large-scale RCTs in real-world applications, such as clinical trials, often face significant practical challenges like budgetary constraints and participant enrollment, leading to small sample sizes and inadequate statistical power. The abundance of available external data presents a promising opportunity to augment information and enhance the analysis of RCTs. However, inappropriate information borrowing or low-quality external data can lead to biased estimations or even incorrect trial conclusions; and privacy concerns may restrict access to individual-level external data in some applications. To address these issues, we propose a novel Bayesian framework that leverages machine learning (ML) predictive and generative models trained on RCT and external data to generate multiple sets of “digital control” (DC) samples that mimic real control outcomes. Using these DC samples, we develop a Bayesian hierarchical framework to adaptively incorporate external information and augment RCTs. Our approach eliminates the need for individual-level external data by enabling the decentralized training of ML models, thereby preserving privacy. To ensure robust inference, we introduce a debiasing term to mitigate potential bias from unmeasured confounders and employ a data-driven prior to discourage inappropriate information borrowing. Through extensive simulation studies, we empirically demonstrate the validity and advantages of the proposed method. Compared to existing approaches, our method significantly improves the precision of treatment effect estimates and enhances statistical power. Citation Format: Ruitao Lin, Ying Yuan, Xiaohan Chi. Digital Controls: An Efficient, Robust, and Privacy-Preserving Bayesian Tool for Adaptive Information Borrowing from External Data [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Artificial Intelligence and Machine Learning; 2025 Jul 10-12; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(13_Suppl):Abstract nr A040.
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 | 0.020 | 0.072 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.004 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.005 | 0.005 |
| Research integrity | 0.002 | 0.005 |
| Insufficient payload (model declined to judge) | 0.008 | 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".