A novel data-driven algorithm to predict anomalous prescription based on patient's feature set

TL;DR

This paper presents a new data-driven algorithm that predicts anomalous radiotherapy prescriptions by analyzing patient features and historical data, aiming to enhance safety and efficiency in treatment plan reviews.

Contribution

The study introduces a novel anomaly detection method using dissimilarity metrics that outperforms traditional models and aids peer review in radiotherapy treatment planning.

Findings

F1 scores between 75% and 94% across groups

Lower false negative rate than manual peer review

Model explains flagged cases and handles class imbalance effectively

Abstract

Appropriate dosing of radiation is crucial to patient safety in radiotherapy. Current quality assurance depends heavily on a peer-review process, where the physicians' peer review on each patient's treatment plan, including dose and fractionation. However, such a process is manual and laborious. Physicians may not identify errors due to time constraints and caseload. We designed a novel prescription anomaly detection algorithm that utilizes historical data to predict anomalous cases. Such a tool can serve as an electronic peer who will assist the peer-review process providing extra safety to the patients. In our primary model, we created two dissimilarity metrics, R and F. R defining how far a new patient's prescription is from historical prescriptions. F represents how far away a patient's feature set is from the group with an identical or similar prescription. We flag prescription if either metric is greater than specific optimized cut-off values. We used thoracic cancer patients (n=2356) as an example and extracted seven features. Here, we report our testing f1 score, between 75%-94% for different treatment technique groups. We also independently validate our results by conducting a mock peer review with three thoracic specialists. Our model has a lower type 2 error rate compared to manual peer-review physicians. Our model has many advantages over traditional machine learning algorithms, particularly in that it does not suffer from class imbalance. It can also explain why it flags each case and separate prescription and non-prescription-related features without learning from the data.