多模态MRI脑肿瘤图像分割方法研究进展

Research progress of multimodal MRI brain tumor image segmentation methods

MRI是一种非侵入性多模态成像方法,广泛应用于脑肿瘤检测和诊断。多模态MRI脑肿瘤图像分割对脑肿瘤的诊断和治疗具有重要意义。目前大部分分割工作还是由医生手动完成,效率低且主观性强,因此寻求一种高效准确的脑肿瘤自动分割方法对临床应用至关重要。本文就基于多模态MRI图像的脑肿瘤分割的研究进展进行综述,对比分析了传统分割方法和基于深度学习的分割方法,并对现有的脑肿瘤图像分割方法存在的问题进行总结并做出展望,以便该领域的研究者更好地了解目前多模态MRI脑肿瘤图像分割方法的研究进展。

MRI is a non-invasive multimodal imaging method,which is widely used in the detection and diagnosis of brain tumors.Multimodal MRI brain tumor image segmentation has important significance for the diagnosis and treatment of brain tumors.At present,most of the segmentation work is still manually completed by doctors,with low efficiency and strong subjectivity.Therefore,seeking an efficient and accurate automatic segmentation method for brain tumors is crucial for clinical applications.We reviewed the research progress of brain tumor segmentation based on multimodal MRI images,compared and analyzed traditional segmentation methods and deep learning based segmentation methods in this paper,and then summarized the problems of existing brain tumor image segmentation methods and makes prospects,so that researchers in this field could better understand the current research progress of multimodal MRI brain tumor image segmentation methods.