摘要
机器人与机器学习的新闻编辑每日新闻-机器学习的新研究是一篇报道的主题。根据News Rx记者在德国慕尼黑的新闻报道，研究表明，“功能连接作为包括边缘人格障碍(BPD)在内的精神疾病的潜在生物标志物已经引起了人们的兴趣。然而，样本量小和缺乏研究内复制导致了没有明确空间焦点的不同发现。”新闻记者从Max-Planck-Contute的精神病学研究中获得了一段引文，“评估功能性连接标记物对BPD的判别性能和泛化能力。在116名BPD患者和72名对照个体的匹配子样本中，采用三种分组策略进行全脑功能MRI静息状态功能连接。我们使用基于感兴趣区域内和之间多尺度功能连接的分类器预测BPD状态(ROIs)，覆盖全脑-全局ROI网络、种子ROI连接、功能一致性和功能一致性。”体素到Voxe L连接-并评估非匹配数据L输出部分分类的泛化能力。全脑连接允许在匹配的内部交叉中对BPD患者与对照进行分类(70%)。当应用于不匹配的外部F样本数据(61-70%)时，分类仍然显著。基于种子的最高准确度在与全局准确度(70-75%)的S极小范围内。在多次比较校正后，单变量连接值不能预测BPD，但弱局部效应与最具辨别力的SEED-ROI一致，在完整的临床访谈中获得了最高的准确度，而自我报告的结果仍然处于偶然水平。全球信号和协变量限制了实际适用性。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Machine Learning is th e subject of a report. According to news reporting from Munich, Germany, by News Rx journalists, research stated, “Functional connectivity has garnered interest as a potential biomarker of psychiatric disorders including borderline personali ty disorder (BPD). However, small sample sizes and lack of within-study replicat ions have led to divergent findings with no clear spatial foci.” The news correspondents obtained a quote from the research from Max-Planck-Insti tute for Psychiatry, “Evaluate discriminative performance and generalizability o f functional connectivity markers for BPD. Whole-brain fMRI resting state functi onal connectivity in matched subsamples of 116 BPD and 72 control individuals de fined by three grouping strategies. We predicted BPD status using classifiers wi th repeated cross-validation based on multiscale functional connectivity within and between regions of interest (ROIs) covering the whole brain-global ROI-based network, seed-based ROI-connectivity, functional consistency,and voxel-to-voxe l connectivity-and evaluated the generalizability of the classification in the l eft-out portion of non-matched data. Full-brain connectivity allowed classificat ion ( 70 %) of BPD patients vs. controls in matched inner cross-val idation. The classification remained significant when applied to unmatched out-o f-sample data ( 61-70 %). Highest seed-based accuracies were in a s imilar range to global accuracies ( 70-75 %), but spatially more sp ecific. The most discriminative seed regions included midline, temporal and soma tomotor regions. Univariate connectivity values were not predictive of BPD after multiple comparison corrections, but weak local effects coincided with the most discriminative seed- ROIs. Highest accuracies were achieved with a full clinical interview while self-report results remained at chance level. The accuracies va ry considerably between random sub-samples of the population, global signal and covariates limiting the practical applicability.”
NETL
NSTL