深度学习和影像组学在脑胶质瘤中的研究进展

Research progress of deep learning and radiomics in glioma

深度学习属于人工智能的一个属支,近几年在疾病检测、诊断、预后评估等方面快速发展,成为热门的研究方法,尤其在医学影像领域。影像组学方法自提出以来在脑胶质瘤方面应用极为可观。基于MRI的深度学习、影像组学能够在脑胶质瘤术前进行鉴别诊断和分级,亦可以预测基因型突变状态,并在术后评估治疗效果及预测无进展生存期、总生存期等,为临床治疗和患者术后随访提供了重要的基础,属于当前胶质瘤的研究热点。笔者就基于MRI的深度学习和影像组学在脑胶质瘤的鉴别诊断、术前分级、基因分型及预后评估方面的研究进展进行综述。

Deep learning is a branch of artificial intelligence.It has developed rapidly in disease detection and prognosis evaluation,and has become a popular research method,especially in the field of medical image in recent years.Radiomics is a very considerable method in the study of glioma.Deep learning and radiomics based on MRI can make differential diagnosis and classification of glioma,predict the genotype change status before operation,evaluate the treatment effect and predict the progression free survival and overall survival after operation,which provides a important basis for clinical treatment and postoperative follow-up.It is a research hotspot of glioma at present.This paper is to review the research progress of deep learning and radiomics based on MRI in the differential diagnosis,preoperative grading,genotyping and prognosis of glioma.