医学影像智能识别痴呆症：模式分类方法综述

105 浏览量更新于2024-08-28 收藏 634KB PDF 举报

本文是一篇针对医学影像自动识别痴呆症的研究论文，标题为“使用医学影像自动识别痴呆症：从模式分类角度的调查”。作者Chuanchuan Zheng、Yong Xia、Yongsheng Pan 和 Jinhu Chen 在2015年11月对文献中现有的自动化痴呆识别算法进行了综合概述。该研究特别关注了模式分类的角度，因为大多数这类算法都涉及特征提取和分类两个步骤。文章首先详细介绍了特征提取方法的三个主要类别：基于voxel（像素）、vertex（顶点）以及ROI（感兴趣区域）的方法。Voxel-based方法关注的是图像中的每个像素特性，Vertex-based方法则关注局部结构和形状信息，而ROI-based方法则是选择特定区域进行分析，以提高计算效率和准确性。这些特征提取技术对于理解痴呆症病变的关键模式至关重要。接着，论文探讨了四类常用的分类器：线性判别分析（Linear Discriminant Analysis, LDA）、贝叶斯分类器、支持向量机（Support Vector Machines, SVM）和人工神经网络（Artificial Neural Networks, ANNs）。这些分类器在处理医学图像数据时各具优势，如LDA强调特征之间的线性分离，贝叶斯分类器基于概率模型，SVM通过构建最优决策边界，而ANN能够模拟人脑的学习过程，对复杂模式进行学习和预测。论文进一步比较了近期发表的一些痴呆症识别算法的表现。结果表明，许多算法在区分阿尔茨海默病（Alzheimer's Disease, AD）与老年人正常情况上达到了相当高的准确度，显示出良好的性能。然而，在区分轻度认知障碍（Mild Cognitive Impairment, MCI）与AD或老年正常人群之间，由于病情的细微差别，识别挑战更大，需要更精细的特征提取和分类策略。总结来说，这篇论文提供了一个系统性的视角，帮助读者理解如何利用医学影像数据和模式分类方法来自动识别痴呆症，并指出了未来可能需要改进和研究的方向，尤其是在MCI与AD区分上的优化。随着技术的发展，这类自动化识别系统有望在临床诊断和早期干预中发挥重要作用。

based pattern recognition problem, which can be solved in

two successive steps: feature extraction and pattern clas-

siﬁcation. During the training stage, image features that can

characterize the patterns of various types or stages of

dementia are calculated based on the quantitative analysis

of medical images. Those features are usually selected and/

or combined to reduce their dimensionality before training

a classiﬁer with the supervised learning techniques [36].

The trained classiﬁer may be treated as a ‘‘black box,’’

which encapsulates the knowledge gleaned from the ima-

ges and is capable of producing the expected predictions

[37]. For each testing image, the features extracted,

selected, and combined in the same way are applied to the

trained classiﬁer to generate a predicted class label that

indicates to which type or stage the dementia case belongs.

The diagram of a typical automated dementia identiﬁcation

system is shown in Fig. 1.

Next, we will review the feature extraction methods and

classiﬁcation methods used in the state of the art dementia

identiﬁcation approaches, respectively.

2.1 Feature extraction methods

According to the types of features extracted from brain

images, feature extraction methods can be roughly grouped

into voxel-based, vertex-based, and ROI-based ones [38].

2.1.1 Voxel-based methods

Voxel-based methods can be traced back to the mid-1990s,

when Wright et al. [39] statistically analyzed the gray

matter and white matter voxel values for schizophrenia

diagnosis. Typically, voxel-based features consist of

statistics of voxel distributions on major brain tissues, such

as the gray matter, white matter, and cerebrospinal ﬂuid

(CSF) [37, 40–43]. Magnin et al. [44] counted the voxel

value histogram in major anatomical regions, which could

be obtained by either image segmentation or registering a

brain atlas onto the image [44–46]. However, the

anatomical parcellation of brain is not a trivial task and

may not be adaptive to the pathology. Fan et al. [42] pro-

posed an adaptive parcellati on approach, in which the

image space is divided into the most discriminative regions

[40, 41, 47–49]. The voxel-based morphometry (VBM)

method proposed by Ashburner et al. [50] allows investi-

gation of focal differences in brain anatomy using the

statistical parametric mapping (SPM), and hence greatly

facilitates the extraction of voxel-based features. Papa-

kostas et al. [51] successfully applied the VBM analysis to

feature extraction on MRI data. Recently, Liu et al. pro-

posed a simul ation method to predict the longitudinal brain

morphological changes in neurodegenerative brains based

on VBM [52].

Fig. 1 Diagram of computer-aided identiﬁcation of dementia

Automated identiﬁcation of dementia using medical imaging

123

剩余10页未读，继续阅读

weixin_38719635

粉丝: 3

医学影像智能识别痴呆症：模式分类方法综述

阿尔兹海默症核磁共振数据集

老年痴呆CT图像数据集

使用FDG-PET成像的转移学习识别跨队列痴呆

脑机接口与可穿戴设备在脑科学领域(老年痴呆，抑郁症，情绪状态识别，数字疗法四个中选一个）中的现状这块做一调研.

老年痴呆症早期记忆能力检测系统设计

关于MRI sequence，在哪里能下载到一些数据集

CHARLS痴呆提取

adni 脑功能数据集下载

机器学习阿尔茨海默classification

最新资源