UniCAD: Efficient and Extendable Architecture for Multi-Task Computer-Aided Diagnosis System

2025-05-14

AI Summary

Medical imaging multi-task computer-aided diagnosis (CAD) systems face challenges including complex pretraining requirements and insufficient open-source platform support. To address these, we propose a unified multi-task diagnostic architecture for both 2D and 3D medical images: it freezes a vision foundation model and employs low-rank adaptation (LoRA)—introducing only 0.17% trainable parameters—for efficient fine-tuning; additionally, it incorporates plug-and-play task-specific expert modules to enable flexible functional expansion. This design substantially reduces task adaptation overhead while maintaining high diagnostic accuracy and deployment efficiency. Evaluated on 12 mainstream medical imaging datasets, our method consistently outperforms existing state-of-the-art approaches. Furthermore, we publicly release the complete codebase, a lightweight expert model library, and an integrated platform—thereby fostering reproducible, extensible, and clinically translatable medical AI research and deployment.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Machine Learning: Deep Neural Architectures and Foundation Models
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

The growing complexity and scale of visual model pre-training have made developing and deploying multi-task computer-aided diagnosis (CAD) systems increasingly challenging and resource-intensive. Furthermore, the medical imaging community lacks an open-source CAD platform to enable the rapid creation of efficient and extendable diagnostic models. To address these issues, we propose UniCAD, a unified architecture that leverages the robust capabilities of pre-trained vision foundation models to seamlessly handle both 2D and 3D medical images while requiring only minimal task-specific parameters. UniCAD introduces two key innovations: (1) Efficiency: A low-rank adaptation strategy is employed to adapt a pre-trained visual model to the medical image domain, achieving performance on par with fully fine-tuned counterparts while introducing only 0.17% trainable parameters. (2) Plug-and-Play: A modular architecture that combines a frozen foundation model with multiple plug-and-play experts, enabling diverse tasks and seamless functionality expansion. Building on this unified CAD architecture, we establish an open-source platform where researchers can share and access lightweight CAD experts, fostering a more equitable and efficient research ecosystem. Comprehensive experiments across 12 diverse medical datasets demonstrate that UniCAD consistently outperforms existing methods in both accuracy and deployment efficiency. The source code and project page are available at https://mii-laboratory.github.io/UniCAD/.

Problem

• Addresses resource-intensive multi-task CAD system development
• Lacks open-source platform for extendable diagnostic models
• Unifies 2D/3D medical image handling with minimal parameters

Innovation

• Leverages pre-trained vision foundation models
• Uses low-rank adaptation for efficiency
• Modular plug-and-play architecture for expandability