DiffKD-DCIS: Predicting Upgrade of Ductal Carcinoma In Situ with Diffusion Augmentation and Knowledge Distillation

2026-01-04

arXiv.org

AI Summary

Accurate prediction of upgrade from ductal carcinoma in situ (DCIS) to invasive ductal carcinoma (IDC) is critical for preoperative decision-making but remains challenging due to limited ultrasound data and poor model generalizability. To address this, this work proposes DiffKD-DCIS, a novel framework that first employs a multimodal conditional diffusion model to synthesize high-quality ultrasound images for data augmentation. A robust teacher network is then trained on the enhanced dataset, and its knowledge is distilled into a lightweight student network to improve inference efficiency without compromising generalization. This study presents the first integration of conditional diffusion-based image generation with knowledge distillation for DCIS upgrade prediction. Evaluated on a multicenter cohort of 1,435 cases, the student model achieves superior external test performance compared to existing baselines, with fewer parameters, faster inference, and accuracy comparable to that of experienced radiologists.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurately predicting the upgrade of ductal carcinoma in situ (DCIS) to invasive ductal carcinoma (IDC) is crucial for surgical planning. However, traditional deep learning methods face challenges due to limited ultrasound data and poor generalization ability. This study proposes the DiffKD-DCIS framework, integrating conditional diffusion modeling with teacher-student knowledge distillation. The framework operates in three stages: First, a conditional diffusion model generates high-fidelity ultrasound images using multimodal conditions for data augmentation. Then, a deep teacher network extracts robust features from both original and synthetic data. Finally, a compact student network learns from the teacher via knowledge distillation, balancing generalization and computational efficiency. Evaluated on a multi-center dataset of 1,435 cases, the synthetic images were of good quality. The student network had fewer parameters and faster inference. On external test sets, it outperformed partial combinations, and its accuracy was comparable to senior radiologists and superior to junior ones, showing significant clinical potential.

Problem

• DCIS upgrade prediction
• invasive ductal carcinoma
• ultrasound data limitation
• generalization ability

Innovation

• conditional diffusion model
• knowledge distillation
• data augmentation
• ultrasound image synthesis
• DCIS upgrade prediction