CAMP: continuous and adaptive learning model in pathology

Abstract

Conventional computational pathology treats diagnostic tasks as independent and individual image classification problems, leading to inefficiencies and high costs. To address this, we introduce CAMP (Continuous and Adaptive learning Model in Pathology), a unified and universal framework for pathology image classification. CAMP is a generative and adaptive classification model that can continuously adapt to new tasks by leveraging pathology-specific prior knowledge and learning task-specific knowledge with minimal computational cost and without catastrophic forgetting. Evaluated CAMP on 22 datasets, including 1,171,526 patches and 11,811 pathology slides, across 17 classification tasks, CAMP achieves state-of-the-art classification performance at both patch- and slide-levels. It also reduces up to 94% of computation time and 85% of storage memory in comparison to the conventional classification models. Our results demonstrate that CAMP can offer a fundamental transformation in pathology image classification, paving the way for the fully digitized and computerized pathology practice.

Ähnliche Arbeiten