文献速递腹部CTMRUS

基于CT的深度学习模型在急性坏死性胰腺炎积液准确分类中的应用:一项多中心研究

Application of deep learning models for accurate classification of fluid collections in acute necrotizing pancreatitis on computed tomography: a multicenter study.
PubMed 文献 · Abdom Radiol (NY) · 2024-09-30

这项回顾性多中心研究纳入152例急性胰腺炎伴胰腺积液且引流前有CT的患者,以MRI或EUS为参考标准,将CT图像按固体碎屑≤30%或>30%标注,用ResNet 50、ViT和MedViT建模。训练验证队列中ViT和MedViT敏感度92.4-98.7%、特异度89.7-98.4%、AUC 0.908-0.980,外部测试队列中MedViT表现最佳,敏感度75.2%、特异度75.3%、AUC 0.753。

要点

  • 深度学习模型可基于CT对急性坏死性胰腺炎积液按固体碎屑≤30%或>30%分类。
  • 训练验证队列中ViT和MedViT的AUC为0.908-0.980,外部测试队列中MedViT的AUC为0.753。
  • 外部测试队列中MedViT对包裹性坏死和>5 cm积液的AUC分别为0.748和0.767。

核心结论

  • 深度学习模型基于CT对包裹性坏死和大于5 cm积液的固体碎屑分类具有中等诊断性能。

研究设计与样本

  • 回顾性研究,涵盖四家三级医院,共纳入152例患者(训练/验证129例,测试23例),训练、验证和测试队列分别有1334、334和512幅图像;以MRI或EUS为参考标准,对轴位CT图像进行标注。

诊断标准与关键数值

  • 以MRI或EUS为参考,将CT图像按固体碎屑≤30%和>30%标注;训练验证队列中ViT和MedViT敏感度92.4-98.7%、特异度89.7-98.4%、AUC 0.908-0.980;外部测试队列中MedViT敏感度75.2%、特异度75.3%、AUC 0.753;MedViT在包裹性坏死中敏感度75.2%、特异度74.3%、AUC 0.748,在大于5 cm积液中敏感度79%、特异度74.2%、AUC 0.767(原文此处写作「79%, 74.2%, 75.3%, and 0.767」,按上下文应为敏感度79%、特异度74.2%、AUC 0.767)。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

PURPOSE: To apply CT-based deep learning (DL) models for accurate solid debris-based classification of pancreatic fluid collections (PFC) in acute pancreatitis (AP).

MATERIAL AND METHODS: This retrospective study comprised four tertiary care hospitals. Consecutive patients with AP and PFCs who had computed tomography (CT) prior to drainage were screened. Those who had magnetic resonance imaging (MRI) or endoscopic ultrasound (EUS) within 20 days of CT were considered for inclusion. Axial CT images were utilized for model training. Images were labelled as those with≤30% solid debris and >30% solid debris based on MRI or EUS. Single center data was used for model training and validation. Data from other three centers comprised the held out external test cohort. We experimented with ResNet 50, Vision transformer (ViT), and MedViT architectures.

RESULTS: Overall, we recruited 152 patients (129 training/validation and 23 testing). There were 1334, 334 and 512 images in the training, validation, and test cohorts, respectively. In the overall training and validation cohorts, ViT and MedVit models had high diagnostic performance (sensitivity 92.4-98.7%, specificity 89.7-98.4%, and AUC 0.908-0.980). The sensitivity (85.3-98.6%), specificity (69.4-99.4%), and AUC (0.779-0.984) of all the models was high in all the subgroups in the training and validation cohorts. In the overall external test cohort, MedViT had the best diagnostic performance (sensitivity 75.2%, specificity 75.3%, and AUC 0.753). MedVit had sensitivity, specificity, and AUC of 75.2%, 74.3%, and 0.748, in walled off necrosis and 79%, 74.2%, 75.3%, and 0.767 for collections >5 cm.

CONCLUSION: DL-models have moderate diagnostic performance for solid-debris based classification of WON and collections greater than 5 cm on CT.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Gupta P, Siddiqui R, Singh S, et al. Application of deep learning models for accurate classification of fluid collections in acute necrotizing pancreatitis on computed tomography: a multicenter study. Abdom Radiol (NY). 2024. PMID: 39347977. doi:10.1007/s00261-024-04607-y

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。