Toxicity Assessment in Preclinical Histopathology via Class-Aware Mahalanobis Distance for Known and Novel Anomalies
arXiv:2602.02124v2 Announce Type: replace-cross Abstract: Drug-induced toxicity is a leading cause of preclinical and early-clinical failure, making early detection critical. Histopathology is the gold standard for toxicity assessment but relies on expert pathologists, creating a bottleneck for large-scale screening. We introduce an AI-based anomaly detection framework for whole-slide images (WSIs) of rodent liver that identifies healthy tissue and known pathologies (anomalies) and flags samples without training data as out-of-distribution (OOD). We evaluate OOD detection on two held-out categ
Advances in AI, particularly in computer vision and anomaly detection, are enabling automated analysis of complex medical imagery, addressing bottlenecks in traditional labor-intensive processes.
This development can significantly accelerate preclinical drug development by automating a critical, time-consuming step, potentially leading to faster and safer drug discovery.
The reliance on human expert pathologists for preliminary toxicity assessment in preclinical trials can be reduced, shifting towards AI-assisted or AI-driven initial screening.
- · Pharmaceutical R&D
- · AI healthcare tech companies
- · Patients
- · Legacy histopathology service providers (without AI integration)
Reduced time and cost in preclinical drug development phases due to expedited toxicity assessment.
Increased throughput for drug candidates and potentially higher success rates in early-stage trials.
A shift in the role of pathologists towards AI supervision and more complex, nuanced diagnostic challenges rather than high-volume screening.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG