Uncertainty-Aware Image Classification In Biomedical Imaging Using Spectral-normalized Neural Gaussian Processes
This work addresses the need for reliable uncertainty-aware models in safety-critical medical imaging to support clinical adoption and trust among pathologists, though it is incremental as it builds on existing methods with lightweight modifications.
The paper tackled the problem of overconfident and poorly calibrated deep learning models in digital pathology by implementing Spectral-normalized Neural Gaussian Process (SNGP) to improve uncertainty estimation and out-of-distribution detection, achieving comparable in-distribution performance with significant enhancements in these areas across six biomedical datasets.
Accurate histopathologic interpretation is key for clinical decision-making; however, current deep learning models for digital pathology are often overconfident and poorly calibrated in out-of-distribution (OOD) settings, which limit trust and clinical adoption. Safety-critical medical imaging workflows benefit from intrinsic uncertainty-aware properties that can accurately reject OOD input. We implement the Spectral-normalized Neural Gaussian Process (SNGP), a set of lightweight modifications that apply spectral normalization and replace the final dense layer with a Gaussian process layer to improve single-model uncertainty estimation and OOD detection. We evaluate SNGP vs. deterministic and MonteCarlo dropout on six datasets across three biomedical classification tasks: white blood cells, amyloid plaques, and colorectal histopathology. SNGP has comparable in-distribution performance while significantly improving uncertainty estimation and OOD detection. Thus, SNGP or related models offer a useful framework for uncertainty-aware classification in digital pathology, supporting safe deployment and building trust with pathologists.