应用深度学习模型用于精确分类急性死性胰腺炎的液体采集在计算机断层扫描:一个多中心研究
Pankaj Gupta1, Ruby Siddiqui2, Shravya Singh2
1Department of Radiodiagnosis and Imaging, Postgraduate Institute of Medical Education and Research, Chandigarh, India. pankajgupta959@gmail.com.
Abdominal radiology (New York)
|September 30, 2024
概括
深度学习模型在基于使用CT扫描的固体碎片含量对胰腺液收集的分类中显示了中度的准确性. 这些模型有助于区分隔壁的死亡和更大的收藏.
科学领域:
- 放射学 放射学是一门学科.
- 人工智能的人工智能
- 医疗成像医学成像
背景情况:
- 胰腺液收集 (PFC) 在急性胰腺炎 (AP) 中很常见.
- 准确的PFC分类,特别是固体残渣含量,对于指导处理至关重要.
- 当前的分类方法可能是主观的,耗时的.
研究的目的:
- 评估深度学习 (DL) 模型的诊断性能,用于AP中基于固体碎片的PFC分类.
- 为了完成这个任务,比较不同的DL架构 (ResNet 50,视觉变压器 (ViT),MedViT).
- 用CT成像来评估模型的准确性.
主要方法:
- 追溯研究涉及152名来自四家三级医院的AP和PFC患者.
- 轴心CT图像用于DL模型的训练和验证.
- 固体碎片含量 (≤30%对比>30%) 通过MRI或EUS确定,并用于地面真相标签.
- 模型在单中心数据上进行训练,并在外部多中心测试队列上进行验证.
主要成果:
- 在培训和验证队列中,ViT和MedViT模型表现出高性能 (AUC 0.908-0.980).
- 在外部测试队列中,MedViT获得了最佳性能,灵敏度为75.2%,特异性为75.3%,AUC为0.753.
- MedViT在隔墙性瘤和大于5厘米的集体中表现相似.
结论:
- 深度学习模型提供了适度的诊断性能,用于根据使用CT的固体碎片含量来分类隔墙性和大型PFC.
- 进一步完善DL模型可能会提高AP管理中的临床决策的准确性.
相关概念视频
Acute Pancreatitis II: Clinical Manifestations and Management
95
Acute pancreatitis presents a complex medical emergency characterized by rapid onset inflammation of the pancreas, demanding timely diagnosis and management to prevent complications. The condition primarily manifests through severe upper abdominal pain that often radiates to the back. This pain intensifies following the consumption of fatty foods. Accompanying symptoms such as nausea, vomiting, abdominal distention, fever, dyspnea, cyanosis, and jaundice can vary in intensity but significantly...
95
Computed Tomography
4.3K
Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...
4.3K


