Abstract

2D single-slice abdominal computed tomography (CT) en- ables the assessment of body habitus and organ health with low ra- diation exposure. However, single-slice data necessitates the use of 2D networks for segmentation, but these networks often struggle to capture contextual information effectively. Consequently, even when trained on identical datasets, 3D networks typically achieve superior segmentation results. In this work, we propose a novel 3D-to-2D distillation framework, leveraging pre-trained 3D models to enhance 2D single-slice segmenta- tion. Specifically, we extract the prediction distribution centroid from the 3D representations, to guide the 2D student by learning intra- and inter-class correlation. Unlike traditional knowledge distillation methods that require the same data input, our approach employs unpaired 3D CT scans with any contrast to guide the 2D student model. Experiments conducted on 707 participants from the single-slice Baltimore Longitu- dinal Study of Aging (BLSA) dataset demonstrate that state-of-the-art 2D multi-organ segmentation methods can benefit from the 3D teacher model, achieving enhanced performance in single-slice multi-organ seg- mentation. Notably, our approach demonstrates considerable efficacy in low-data regimes, outperforming the model trained with all available training subjects even when utilizing only 200 training subjects. Thus, this work underscores the potential to alleviate manual annotation bur- dens.

Links to Paper and Supplementary Materials

Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/ADSMI2024_129.pdf

SharedIt Link: Not yet available

SpringerLink (DOI): Not yet available

Supplementary Material: Not Submitted

Link to Open Review

Open Review Page: Not Available

BibTex

@InProceedings{YuXin_Enhancing_MICCAISAT2026,
        author = { Yu, Xin AND Yang, Qi AND Liu, Han AND Lee, Ho Hin AND Tang, Yucheng AND Remedios, Lucas W. AND Kim, Michael E. AND Zhang, Rendong AND Bao, Shunxing AND Huo, Yuankai AND Moore, Ann Zenobia AND Ferrucci, Luigi AND Landman, Bennett A.},
        title = { { Enhancing Single-Slice Segmentation with 3D-to-2D Unpaired Scan Distillation } },
        booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
        year = {2026},
        publisher = {Springer Nature Switzerland},
        volume = {LNCS 17276},
        month = {pending},
        page = {pending}
}


back to top