Cross-Domain Semi-Supervised Organ Detection

Nian Li, MORTEZA GHAHREMANI, Bailiang Jian, Pascual Tejero Cervera, Benedikt Wiestler, Marcus Makowski, Christian Wachinger

Keywords: Cross-Domain;Semi-Supervised;Object Detection;Teacher-Student;Pseudo-label;Curriculum Learning

TL;DR: Cross-Domain Semi-Supervised Organ Detection

Abstract:

Accurate 3D organ detection in Computed Tomography (CT) imaging is crucial for various clinical applications. However, learning-based detection models rely on large, annotated datasets obtained from diverse imaging devices across multiple healthcare institutions, which are expensive and labor-intensive to acquire.
Traditional semi-supervised approaches often rely solely on unlabeled target data, neglecting the benefits of a few labeled target samples. To address this limitation, we introduce a novel cross-domain semi-supervised detection framework (CDSS-Det) built upon the Transformer-based Organ-DETR model. CDSS-Det synergistically integrates pseudo-labeling, curriculum learning, and domain adaptation to enable effective knowledge transfer from a well-annotated source domain to a target domain with limited labels. Experiments on multi-domain CT datasets demonstrate that incorporating a small number of labeled target samples significantly boosts detection performance over conventional domain adaptation and semi-supervised methods. CDSS-Det consistently achieves higher mean Average Precision (mAP), with notable improvements in detecting small organs, and surpasses a fully supervised model trained solely on the labeled target domain by over 10%. These results underscore the potential of CDSS-Det in efficiently leveraging both labeled and unlabeled target data in cross-domain organ detection, advancing annotation-efficient deep learning models in medical imaging.

Primary Subject Area: Transfer Learning and Domain Adaptation

Secondary Subject Area: Detection and Diagnosis

Registration Requirement: Yes

Reproducibility: https://github.com/CodeForAAAI2026/CDSS-Det

Visa & Travel: Yes

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 10