A Parallel Network for LRCT Segmentation and Uncertainty Mitigation with Fuzzy Sets

Shiyi Wang, Yang Nan, Xiaodan Xing, Yingying Fang, Simon Lf Walsh, Guang Yang
Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence, PMLR 286:4443-4457, 2025.

Abstract

Accurate segmentation of airways in Low-Resolution CT (LRCT) scans is vital for diagnostics in scenarios such as reduced radiation exposure, emergency response, or limited resources. Yet manual annotation is labor-intensive and prone to variability, while existing automated methods often fail to capture small airway branches in lower-resolution 3D data. To address this, we introduce \textbf{FuzzySR}, a parallel framework that merges super-resolution (SR) and segmentation. By concurrently producing high-resolution reconstructions and precise airway masks, it enhances anatomic fidelity and captures delicate bronchi. FuzzySR employs a deep fuzzy set mechanism, leveraging learnable $t$-distribution and triangular membership functions via cross-attention. Through parameters $\mu$, $\sigma$, and $d_f$, it preserves uncertain features and mitigates boundary noise. Extensive evaluations on lung cancer, COVID-19, and pulmonary fibrosis datasets confirm FuzzySR’s superior segmentation accuracy on LRCT, surpassing even high-resolution baselines. By uniting fuzzy-logic-driven uncertainty handling with SR-based resolution enhancement, FuzzySR effectively bridges the gap for robust airway delineation from LRCT data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v286-wang25d, title = {A Parallel Network for LRCT Segmentation and Uncertainty Mitigation with Fuzzy Sets}, author = {Wang, Shiyi and Nan, Yang and Xing, Xiaodan and Fang, Yingying and Walsh, Simon Lf and Yang, Guang}, booktitle = {Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence}, pages = {4443--4457}, year = {2025}, editor = {Chiappa, Silvia and Magliacane, Sara}, volume = {286}, series = {Proceedings of Machine Learning Research}, month = {21--25 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v286/main/assets/wang25d/wang25d.pdf}, url = {https://proceedings.mlr.press/v286/wang25d.html}, abstract = {Accurate segmentation of airways in Low-Resolution CT (LRCT) scans is vital for diagnostics in scenarios such as reduced radiation exposure, emergency response, or limited resources. Yet manual annotation is labor-intensive and prone to variability, while existing automated methods often fail to capture small airway branches in lower-resolution 3D data. To address this, we introduce \textbf{FuzzySR}, a parallel framework that merges super-resolution (SR) and segmentation. By concurrently producing high-resolution reconstructions and precise airway masks, it enhances anatomic fidelity and captures delicate bronchi. FuzzySR employs a deep fuzzy set mechanism, leveraging learnable $t$-distribution and triangular membership functions via cross-attention. Through parameters $\mu$, $\sigma$, and $d_f$, it preserves uncertain features and mitigates boundary noise. Extensive evaluations on lung cancer, COVID-19, and pulmonary fibrosis datasets confirm FuzzySR’s superior segmentation accuracy on LRCT, surpassing even high-resolution baselines. By uniting fuzzy-logic-driven uncertainty handling with SR-based resolution enhancement, FuzzySR effectively bridges the gap for robust airway delineation from LRCT data.} }
Endnote
%0 Conference Paper %T A Parallel Network for LRCT Segmentation and Uncertainty Mitigation with Fuzzy Sets %A Shiyi Wang %A Yang Nan %A Xiaodan Xing %A Yingying Fang %A Simon Lf Walsh %A Guang Yang %B Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2025 %E Silvia Chiappa %E Sara Magliacane %F pmlr-v286-wang25d %I PMLR %P 4443--4457 %U https://proceedings.mlr.press/v286/wang25d.html %V 286 %X Accurate segmentation of airways in Low-Resolution CT (LRCT) scans is vital for diagnostics in scenarios such as reduced radiation exposure, emergency response, or limited resources. Yet manual annotation is labor-intensive and prone to variability, while existing automated methods often fail to capture small airway branches in lower-resolution 3D data. To address this, we introduce \textbf{FuzzySR}, a parallel framework that merges super-resolution (SR) and segmentation. By concurrently producing high-resolution reconstructions and precise airway masks, it enhances anatomic fidelity and captures delicate bronchi. FuzzySR employs a deep fuzzy set mechanism, leveraging learnable $t$-distribution and triangular membership functions via cross-attention. Through parameters $\mu$, $\sigma$, and $d_f$, it preserves uncertain features and mitigates boundary noise. Extensive evaluations on lung cancer, COVID-19, and pulmonary fibrosis datasets confirm FuzzySR’s superior segmentation accuracy on LRCT, surpassing even high-resolution baselines. By uniting fuzzy-logic-driven uncertainty handling with SR-based resolution enhancement, FuzzySR effectively bridges the gap for robust airway delineation from LRCT data.
APA
Wang, S., Nan, Y., Xing, X., Fang, Y., Walsh, S.L. & Yang, G.. (2025). A Parallel Network for LRCT Segmentation and Uncertainty Mitigation with Fuzzy Sets. Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 286:4443-4457 Available from https://proceedings.mlr.press/v286/wang25d.html.

Related Material