Abstract A041: A Hybrid Active Learning (AL) Random Forest Model with KNN Imputation to Predict Recurrence in Ductal Carcinoma In Situ (DCIS) of the Breast
Bibliographic record
Abstract
Abstract Background: DCIS is a significant clinical burden that may benefit from machine learning. However, the data requirement of traditional Machine Learning (ML) is impractical. Herein, I explore the use of an emerging ML technique that can work with a small dataset. Methods: A dataset from a published DCIS study (Chan et al., 2001; DOI: 10.1002/1097-0142(2001010191:1<9::aid-cncr2>3.0.co;2-e)) was used with permission. Unlike the original analysis, patients with missing margin width data were retained. Little’s Missing Completely at Random (MCAR) test (P = 0.88) supported this assumption. Missing values were imputed using k-Nearest Neighbours (KNN), preserving statistical properties (pre-imputation vs post-imputation margin width mean: 4.62 mm vs 4.61 mm; SD: 5.67 mm vs 5.59 mm; t-test P = 0.986). A hybrid AL strategy, combining uncertainty-based and random sampling, was applied using a Random Forest classifier. Results: By round 3 of AL, the model achieved a mean F1-score of 0.95 (range: 0.92–1.00) and a mean Matthews Correlation Coefficient (MCC) of 0.94 (range: 0.91–1.00) across multiple random seeds. This level of performance was reached with approximately 80 labelled patient records, highlighting the data efficiency of AL. Conclusion: A hybrid AL approach combined with KNN imputation produced a high-performing recurrence prediction model using a small labelled dataset. This framework may support efficient clinical decision-making in DCIS and could be further extended with integration of a local large language model (LLM) for interpretability and clinical explanation. Citation Format: Kai Chan. Hybrid Active Learning (AL) Random Forest Model with KNN Imputation to Predict Recurrence in Ductal Carcinoma In Situ (DCIS) of the Breast [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 A041.
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 | 0.007 | 0.008 |
| 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 | 0.001 |
| Open science | 0.003 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| 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".