Scholarlink

Objectives <p>To assess a deep learning (DL) model using portal-venous phase CT for discriminating colorectal cancer liver metastasis (CRLMs) and hemangiomas (HMs).</p> Materials and methods <p>Colorectal cancer (CRC) patients diagnosed with CRLMs or HMs at two medical centers from January 2018 and April 2024 were retrospectively included. Lesions were automatically segmented using TotalSegmentator. DL models, DenseNet-201 and ResNet-152, were trained to classify CRLMs and HMs. Their performance, measured by AUC, was evaluated on validation and test sets. Subgroup analyses were conducted for lesions ≤ 10 mm (subcentimeter) and 10–30 mm. Radiologists’ diagnostic performance with and without DL assistance was compared using a multi-reader multi-case analysis.</p> Results <p>534 CRLMs (134 CRC-patients; median, 60 years) and 262 HMs (154 CRC-patients; median, 62 years) were divided into the training, validation and test set. The Dice coefficients of TotalSegmentor for automatically segmenting subcentimeter and 10–30 mm lesions were 0.692 ± 0.099 and 0.861 ± 0.033, respectively (<i>p</i> &lt; 0.01). ResNet-152 model achieved AUCs of 0.875 (95% CI: 0.838–0.912), 0.858 (95% CI: 0.781–0.935), 0.776 (95% CI: 0.703–0.848) for classifying CRLMs and HMs on the training, validation, and test sets, respectively. The AUCs for distinguishing between 10–30 mm CRLMs and HMs improved from 0.851 (95% CI: 0.821–0.880) to 0.879 (95% CI: 0.853–0.906) with DL assistance compared to without (<i>p</i> = 0.015). For subcentimeter CRLMs and HMs, the AUCs for the radiologists and the DL-assisted diagnosis were 0.742 (95% CI: 0.669–0.814) and 0.763 (95% CI: 0.681–0.845), respectively (<i>p</i> = 0.558).</p> Conclusion <p>DL can assist radiologists in distinguishing 10–30 mm CRLMs from HMs in CRC patients. The value of DL-assisted diagnosis is limited for subcentimetre CRLMs and HMs.</p> Critical relevance statement <p>Dynamic detection of hypoenhancing liver lesions in patients with CRC is exceptionally challenging. The DL tool we have developed can assist in evaluating CRLMs and HMs.</p> Key Points <p><UnorderedList Mark="Bullet"> <ItemContent> <p>TotalSegmentator can perform automatic segmentation of CRLMs and HMs, but demonstrates poorer segmentation consistency for subcentimeter lesions.</p> </ItemContent> <ItemContent> <p>This DL model assists radiologists in distinguishing 10–30 mm CRLMs from HMs in CRC patients.</p> </ItemContent> <ItemContent> <p>Subcentimeter CRLMs and HMs can require further MRI scanning.</p> </ItemContent> </UnorderedList></p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep learning in differentiating the colorectal cancer combined with hepatic enhancing nodules: liver metastases vs hemangiomas

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月26日

查看全文
  • Shenglin Li,
  • Shanshan Zhang,
  • Yuebo Wang,
  • Ting Lu,
  • Xinmei Yang,
  • Jialiang Ren,
  • Zhimei Jiao,
  • Yaqiong Ma,
  • Yuan Xu,
  • Yufeng Li,
  • Long Yuan,
  • Yu Guo,
  • Haisheng Wang,
  • Fengyu Zhou,
  • Qianqian Chen,
  • Jianqiang Liu,
  • Junlin Zhou,
  • Guojin Zhang

摘要

Objectives

To assess a deep learning (DL) model using portal-venous phase CT for discriminating colorectal cancer liver metastasis (CRLMs) and hemangiomas (HMs).

Materials and methods

Colorectal cancer (CRC) patients diagnosed with CRLMs or HMs at two medical centers from January 2018 and April 2024 were retrospectively included. Lesions were automatically segmented using TotalSegmentator. DL models, DenseNet-201 and ResNet-152, were trained to classify CRLMs and HMs. Their performance, measured by AUC, was evaluated on validation and test sets. Subgroup analyses were conducted for lesions ≤ 10 mm (subcentimeter) and 10–30 mm. Radiologists’ diagnostic performance with and without DL assistance was compared using a multi-reader multi-case analysis.

Results

534 CRLMs (134 CRC-patients; median, 60 years) and 262 HMs (154 CRC-patients; median, 62 years) were divided into the training, validation and test set. The Dice coefficients of TotalSegmentor for automatically segmenting subcentimeter and 10–30 mm lesions were 0.692 ± 0.099 and 0.861 ± 0.033, respectively (p < 0.01). ResNet-152 model achieved AUCs of 0.875 (95% CI: 0.838–0.912), 0.858 (95% CI: 0.781–0.935), 0.776 (95% CI: 0.703–0.848) for classifying CRLMs and HMs on the training, validation, and test sets, respectively. The AUCs for distinguishing between 10–30 mm CRLMs and HMs improved from 0.851 (95% CI: 0.821–0.880) to 0.879 (95% CI: 0.853–0.906) with DL assistance compared to without (p = 0.015). For subcentimeter CRLMs and HMs, the AUCs for the radiologists and the DL-assisted diagnosis were 0.742 (95% CI: 0.669–0.814) and 0.763 (95% CI: 0.681–0.845), respectively (p = 0.558).

Conclusion

DL can assist radiologists in distinguishing 10–30 mm CRLMs from HMs in CRC patients. The value of DL-assisted diagnosis is limited for subcentimetre CRLMs and HMs.

Critical relevance statement

Dynamic detection of hypoenhancing liver lesions in patients with CRC is exceptionally challenging. The DL tool we have developed can assist in evaluating CRLMs and HMs.

Key Points

TotalSegmentator can perform automatic segmentation of CRLMs and HMs, but demonstrates poorer segmentation consistency for subcentimeter lesions.

This DL model assists radiologists in distinguishing 10–30 mm CRLMs from HMs in CRC patients.

Subcentimeter CRLMs and HMs can require further MRI scanning.

Graphical Abstract