Is Uncertainty Quantification a Viable Alternative to Learned Deferral?
This work addresses the problem of AI safety in clinical settings, such as ophthalmology, by proposing a more robust deferral strategy to handle out-of-distribution data, though it appears incremental as it builds on existing uncertainty quantification techniques.
The study investigated whether uncertainty quantification methods could serve as a robust alternative to learned deferral models for AI decision-making in medical imaging, particularly under data shift conditions, finding that uncertainty quantification methods show promise for accurate glaucoma classification and deferral.
Artificial Intelligence (AI) holds the potential to dramatically improve patient care. However, it is not infallible, necessitating human-AI-collaboration to ensure safe implementation. One aspect of AI safety is the models' ability to defer decisions to a human expert when they are likely to misclassify autonomously. Recent research has focused on methods that learn to defer by optimising a surrogate loss function that finds the optimal trade-off between predicting a class label or deferring. However, during clinical translation, models often face challenges such as data shift. Uncertainty quantification methods aim to estimate a model's confidence in its predictions. However, they may also be used as a deferral strategy which does not rely on learning from specific training distribution. We hypothesise that models developed to quantify uncertainty are more robust to out-of-distribution (OOD) input than learned deferral models that have been trained in a supervised fashion. To investigate this hypothesis, we constructed an extensive evaluation study on a large ophthalmology dataset, examining both learned deferral models and established uncertainty quantification methods, assessing their performance in- and out-of-distribution. Specifically, we evaluate their ability to accurately classify glaucoma from fundus images while deferring cases with a high likelihood of error. We find that uncertainty quantification methods may be a promising choice for AI deferral.