| 摘要: |
| 胰腺神经内分泌肿瘤是一组临床表现多变的、具有显著异质性和复杂性的肿瘤，发病率低，具有恶性潜能。它的组织学分级对治疗方案的选择和预后评估具有重要意义，若能在手术前评估患者的病理分级，将有助于治疗策略的选择和预后评估。随着人工智能的蓬勃发展，以深度学习、影像组学、纹理分析技术为代表的人工智能精准影像分析技术可对图像表征信息进行更为深入的分析和阐释，这些图像表征信息与病理分级具有相关性。本文针对人工智能医学影像技术在神经内分泌肿瘤分级中的应用现状和进展作一综述。 |
| 关键词: 胰腺肿瘤 神经内分泌肿瘤 病理分期 人工智能 深度学习 影像组学 纹理分析 |
| DOI：10.16781/j.0258-879x.2020.04.0433 |
| 投稿时间：2020-02-18修订日期：2020-03-12 |
| 基金项目: |
|
| Application of artificial intelligence medical imaging technology in grading of pancreatic neuroendocrine neoplasms |
| SONG Tao,LU Jian-ping,ZHANG Qian-wen* |
(Department of Radiology, Changhai Hospital, Naval Medical University (Second Military Medical University), Shanghai 200433, China
*Corresponding author) |
| Abstract: |
| Pancreatic neuroendocrine neoplasms (pNENs) are a group of tumors with heterogeneous clinical characteristics and complex nature. Though the incidence of pNENs is low, all pNENs are considered to have a potential to be malignant. The histological grading is important to the selection of treatment and prognosis evaluation. If the pathological grading can be evaluated before operation, it will be helpful to the selection of treatment strategy and prognosis evaluation. With the rapid development of artificial intelligence (AI), the precision image analysis technology represented by deep learning, image group and texture analysis technology can analyze and interpret the image representation information more deeply, which is related to pathological grade. This paper reviews the current status and recent advances of the application of AI medical imaging technology in grading pNENs. |
| Key words: pancreatic neoplasms neuroendocrine neoplasms pathological stage artificial intelligence deep learning radiomics texture analysis |
