Explainable one-class feature extraction by adaptive resonance for anomaly detection in quality assurance
Bibliographic record
Abstract
In this study, we address the inherent challenges in radiotherapy (RT) plan quality assessment (QA). RT, a prevalent cancer treatment, utilizes high-energy beams to target tumors while sparing adjacent healthy tissues. Typically, an RT plan is refined through several QA cycles by experts to ensure it meets clinical and operational objectives before being considered safe for patient treatment. This iterative process tends to eliminate unacceptable plans, creating a significant class imbalance problem for machine learning efforts aimed at automating the classification of RT plans as either acceptable or not. The complexity of RT treatment plans, coupled with the aforementioned class imbalance issue, introduces a generalization problem that significantly hinders the efficacy of traditional binary classification approaches. We introduce a novel one-class classification framework, using an adaptive neural network architecture, that outperforms both traditional binary and standard one-class classification methods in this imbalanced and complex context, despite the inherent disadvantage of not learning from unacceptable plans. Unlike its predecessors, our method enhances anomaly detection for RT plan QA without compromising on interpretability-a critical feature in healthcare applications, where understanding and trust in automated decisions are paramount. By offering clear insights into decision-making processes, our method allows healthcare professionals to quickly identify and address specific deficiencies in RT plans deemed unacceptable, thereby streamlining the QA process and enhancing patient care efficiency and safety.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".