A Deployment-Friendly Foundational Framework for Efficient Computational Pathology

2026-02-15

AI Summary

This study addresses the challenge of deploying pathology foundation models on clinical edge devices due to their parameter redundancy and high computational overhead. To this end, the authors propose LitePath, a framework that constructs a lightweight foundation model, LiteFM, via multi-model distillation from Virchow2, H-Optimus-1, and UNI2, and incorporates an Adaptive Patch Selector (APS) for efficient inference. The resulting model retains 99.71% of the original AUC performance while reducing parameters by 28× and FLOPs by 403.5×, enabling it to process 208 whole-slide images per hour on a Jetson Orin Nano Super with a 171× reduction in energy consumption. Furthermore, the work introduces the D-Score metric to quantify deployment-friendliness, demonstrating substantial improvements over existing models.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Other Foundations of Computer Vision
• Search and Optimization: Other Foundations of Search & Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Pathology foundation models (PFMs) have enabled robust generalization in computational pathology through large-scale datasets and expansive architectures, but their substantial computational cost, particularly for gigapixel whole slide images, limits clinical accessibility and scalability. Here, we present LitePath, a deployment-friendly foundational framework designed to mitigate model over-parameterization and patch level redundancy. LitePath integrates LiteFM, a compact model distilled from three large PFMs (Virchow2, H-Optimus-1 and UNI2) using 190 million patches, and the Adaptive Patch Selector (APS), a lightweight component for task-specific patch selection. The framework reduces model parameters by 28x and lowers FLOPs by 403.5x relative to Virchow2, enabling deployment on low-power edge hardware such as the NVIDIA Jetson Orin Nano Super. On this device, LitePath processes 208 slides per hour, 104.5x faster than Virchow2, and consumes 0.36 kWh per 3,000 slides, 171x lower than Virchow2 on an RTX3090 GPU. We validated accuracy using 37 cohorts across four organs and 26 tasks (26 internal, 9 external, and 2 prospective), comprising 15,672 slides from 9,808 patients disjoint from the pretraining data. LitePath ranks second among 19 evaluated models and outperforms larger models including H-Optimus-1, mSTAR, UNI2 and GPFM, while retaining 99.71% of the AUC of Virchow2 on average. To quantify the balance between accuracy and efficiency, we propose the Deployability Score (D-Score), defined as the weighted geometric mean of normalized AUC and normalized FLOP, where LitePath achieves the highest value, surpassing Virchow2 by 10.64%. These results demonstrate that LitePath enables rapid, cost-effective and energy-efficient pathology image analysis on accessible hardware while maintaining accuracy comparable to state-of-the-art PFMs and reducing the carbon footprint of AI deployment.

Problem

• computational pathology
• foundation models
• model over-parameterization
• patch redundancy
• deployment efficiency

Innovation

• LitePath
• knowledge distillation
• adaptive patch selection
• computational pathology
• deployability score