基于全大脑休息状态功能连接特征的刺激性肠综合征患者的支持向量机分类
Lei Xie1,2, Zelin Zhuang1,2, Xiaona Lin1
1Department of Radiology, the First Affiliated Hospital of Shantou University Medical College, Shantou, China.
Quantitative imaging in medicine and surgery
|October 21, 2024
概括
使用大脑连接的机器学习有效地区分了患有刺激性肠综合征 (IBS) 的患者与健康个体. 这种方法有望改善IBS诊断和个性化治疗.
科学领域:
- 神经成像是一种神经成像.
- 机器学习 机器学习
- 胃肠病学 胃肠病学
背景情况:
- 刺激性肠综合征 (IBS) 涉及大脑-肠道信号干扰,导致胃肠道问题和情绪症状恶化.
- 目前的IBS诊断面临着挑战,尽管我们了解大脑-肠轴的作用.
- 大脑-肠轴功能障碍是IBS病变的关键因素.
研究的目的:
- 应用机器学习与静止状态功能磁共振成像 (rs-fMRI) 来区分IBS患者与健康对照 (HCs).
- 为了确定脑功能连接 (FC) 模式,表明IBS.
- 探索rs-fMRI FC在提高IBS诊断精度方面的潜力.
主要方法:
- 176名参与者 (88名IBS患者,88名HC患者) 接受了rs-fMRI和T1WI.
- 全脑功能连接 (FC) 矩阵是使用自动解剖标记 (AAL) 提取的.
- 被监督的机器学习模型被训练并使用FC特征进行分类评估.
主要成果:
- 支持矢量机 (SVM) 模型实现了最高的分类精度 (75%) 和AUC (0.7788).
- 影响分类的关键FC连接包括罗兰 operculum到前带带状回形和calcarine sulcus到下额头回形.
- 特定的大脑区域及其连接模式被确定为IBS分类的重要区域.
结论:
- 使用全脑静止状态FC的SVM是对IBS患者和HCs进行分类的最佳模型.
- 与情绪相关的大脑区域的连接特征显著影响模型性能.
- rs-fMRI FC分析为精准医学和改善IBS诊断疗效提供了潜力.
相关概念视频
01:17Irritable Bowel Syndrome I: Introduction
252
Irritable Bowel Syndrome (IBS) is characterized by functional disturbances in the gastrointestinal system, presenting a cluster of symptoms without evident structural or biochemical abnormalities. It primarily affects the large intestine and may cause abdominal pain, bloating, excessive gas, diarrhea, constipation, or both.
IBS is a chronic condition that can persist over a long period or recur frequently.
The pathogenesis of IBS involves a complex interplay of the following factors:
Altered...
IBS is a chronic condition that can persist over a long period or recur frequently.
The pathogenesis of IBS involves a complex interplay of the following factors:
Altered...
252
01:30Irritable Bowel Syndrome II: Clinical Features and Diagnostic Evaluation
145
Irritable Bowel Syndrome II: Clinical Features and Diagnostic Evaluation
Irritable Bowel Syndrome (IBS) is classified into subtypes based on the predominant bowel habits as determined by the Bristol Stool Form Scale (BSFS). The subtypes are:
Irritable Bowel Syndrome (IBS) is classified into subtypes based on the predominant bowel habits as determined by the Bristol Stool Form Scale (BSFS). The subtypes are:
145