Advances in Psychology
Vol. 09  No. 01 ( 2019 ), Article ID: 28429 , 7 pages
10.12677/AP.2019.91005

Research Progress on Predictive Models Based on Machine Learning for Depression

Bingbing Yang, Junnan Li, Hong He, Qiuyang Feng

Key Laboratory of Cognition and Personality, Ministry of Education, Department of Psychology, Southwest University (SWU), Chongqing

Received: Dec. 24th, 2018; accepted: Jan. 4th, 2019; published: Jan. 11th, 2019

ABSTRACT

With the rise of machine learning methods, more and more studies have incorporated predictive models into the field of neuroscience research, especially in the study of depression, but there is an interpretation of the differences in the stability of research results. The prediction and treatment of individual differences in depression are currently achieved through machine learning. This paper summarizes: 1) the construction of predictive models; 2) the current research status of depression prediction; 3) the problems and current conclusions in the prediction; 4) the diagnosis and significance of depression. The overall diagnosis of depression is achieved through predictive models, but more evidence is needed for the consistency of neural prediction studies or by meta-analysis. In the future, study should be combined with clinical to improvement the diagnosis and treatment of depression.

Keywords:Depression, Machine Learning, Predictive Model

基于机器学习的预测模型对抑郁症的 研究进展

杨兵兵,李俊男,何鸿,冯秋阳

西南大学心理学部,认知与人格教育部重点实验室,重庆

收稿日期:2018年12月24日;录用日期:2019年1月4日;发布日期:2019年1月11日

摘 要

随着机器学习方法的兴起,越来越多的研究将预测模型纳入神经领域的研究中,尤其在抑郁症的研究中做了大量工作,但是存在研究结果稳定性的差异。目前通过机器学习的方法实现对抑郁症个体差异的预测以及治疗。本文总结了:1) 预测模型的构建;2) 抑郁症预测的研究现状;3) 预测中存在的问题和当前的总结;4) 对抑郁症的诊断和意义的展望。总体实现了通过预测模型实现对抑郁症的诊断,但是在神经预测的研究的一致性上需要更多的证据或者运用元分析的方法实现。未来结合临床提高抑郁症的诊断和治疗。

关键词 :抑郁,机器学习,预测模型

Copyright © 2019 by author(s) and Hans Publishers Inc.

This work is licensed under the Creative Commons Attribution International License (CC BY).

http://creativecommons.org/licenses/by/4.0/

1. 引言

抑郁症(Major Depressive Disorder)是常见的精神疾病,大量的神经影像研究揭示由于大脑结构存在异常造成的脑部疾病。抑郁症的发病率很高且终身患病率高达20% (Blazer, Kessler et al., 1994)。抑郁症主要表现为情绪低落,伴随反复而持久的悲伤感,负罪感和无用感,具有很高的自杀率(Jia, Huang et al., 2010),造成短暂的或者永久的情绪障碍,影响问题解决、注意、动机和睡眠(Blazer, Kessler et al., 1994)。世界范围内抑郁症是使个体衰退丧失生活能力第二大疾病(Ferrari, Charlson et al., 2013)。

核磁共振成像(MRI)的脑成像的研究发现抑郁症病人存在多个脑区的结构和功能异常。结构上发现抑郁症病人在杏仁核,右侧海马旁扣带回的灰质体积会更小(Sacher, Neumann et al., 2012)。静息态功能上发现抑郁症病人在左侧海马旁回,右侧小脑后叶,右侧额中回表现出明显的比率低频振幅Fractional Amplitude of Low-Frequency Fluctuation (fALFF)的下降(Liu, Guo et al., 2013),在左侧枕中皮层,右侧顶下小叶,右侧楔前叶表现出局部一致性(Regional Homogeneity, ReHo)的上升(Liang, Zhou et al., 2013)。情绪加工的任务态(Task-based fMRI)发现抑郁症病人在杏仁核和膝下前扣带回存在异常(Langenecker, Kennedy et al., 2007; Costafreda, Khanna et al., 2009; Pizzagalli, 2011; Dichter, Gibbs et al., 2015)。脑网络发现抑郁症病人在凸显网络和情绪加工网络存在异常(Dichter, Gibbs et al., 2015)。虽然研究证实了一些异常的脑区,但是这些脑区结果往往分布广泛而且不够精确,为了从整体上理解和探讨抑郁症病人的病理性特征,通过机器学习的方法实现对抑郁症个体差异的预测。

目前研究抑郁症和涉及的风险因素面临很多的挑战,神经学家设计了一些适合抑郁症病人反应的实验,目的都是想探讨神经机制和调节机制通过神经反应去预测个体的行为反应。大部分的研究通过使用机器学习的方法基于神经影像和歌词行为的构建预测模型实现对抑郁症的神经机制探索,从而实现对抑郁的治疗和诊断。事实上,越来越多的证据表明通过大脑活动去预测抑郁症个体的治疗效果是优于标准的临床措施(Hahn, Kircher et al., 2015; Williams, Korgaonkar et al., 2015; Ball, Goldsteinpiekarski et al., 2017),但是目前的研究存在的研究结果差异,为实现准确预测还需要进一步研究总结(Crane, Jenkins et al., 2017)。

2. 预测模型

随着机器学习在神经领域的兴起,基于神经影像的预测研究也越来多,就目前研究使用机器学习的方法构建预测模型的过程是基本保持统一的,但主要是四个步骤:特征的选择、建模、测试模型和预测效果。通常特征选择和建模都是在训练集数据,测试模型是用在之前未见过的数据。预测模型是描述在数据特征和结果间相互学习的过程中形成预测模型预测新的数据,而不是去描述大脑-行为的相关(Gabrieli, Ghosh et al., 2015)。

2.1. 特征选择

特征选择主是鉴别模型预测的参数,方法主要有两大类:假设驱动和数据驱动。假设驱动是根据假设的依据去选择特征用于测试模型的预测。数据驱动是依赖于统计技术去鉴别在行为上最有可能和个体差异相关的特征。假设驱动和数据驱动都有过滤法(filter methods) (基于与行为的关系选择特征)包装法(wrapper methods) (不同特征的整合提升预测效率,比如通过系统地消除模型中最小的预测特征)和嵌入法(选择特征匹配到模型中,比如拉索、弹性网络和回归) (Guyon & Elisseeff, 2003)。

假设和数据驱动方法都可以结合来自多个领域的预测因子,包括遗传学,大脑结构和功能以及行为(Whelan, Watts et al., 2014)。对于测量的描述通常用坡度,截距或拐点。虽然模型特征的数量没有理论上的限制,但最好的做法是预测值不超过观测值,避免造成噪音(过度拟合) (Guyon & Elisseeff, 2003)。

此外,需要考虑在广泛性(generalizability)、可解释性(interpretability)和变量的解释(variance explained.)间存在的局限性。虽然具有较少特征的模型可能更易于解释,但具有更多特征的模型在行为上有额外差异,能够更好地表征复杂的模式神经表型(Kievit, Brandmaier et al., 2017)。

2.2. 建模

经过选择特征后,使用分类模型(classifier model)或者回归模型(regression mode)建立预测变量和行为的关系。分类是预测模型中常用的方法,比如用支持向量机(support vector machine, SVM),逻辑回归(logistic regression),目的是区分预测变量。分类多变量模型通过影像区分控制组和患者的准确率达75% (Woo, Chang et al., 2017)。回归模型是连续预测不用于分类预测,比如用线性回归,支持向量回归算法。分类器和回归模型都可以应用于横断面数据或纵向数据,回归模型可以包含诸如生长曲线建模等技术来预测过去或未来的变化。促进精神病理学和跨诊断的发展。

2.3. 模型测试

模型测试是通过测试新的数据判断是否具有普遍性。常用的检测方法就是交叉验证,比如(k-fold or leave-one-subject-out cross-validation在过度拟合和假阳性结果的问题上还是如要谨慎(Woo, Chang et al. 2017, Yarkoni & Westfall, 2017)。最优模型的选择和统计的严谨性来说交叉验证是必要的,但是通过置换检验的方法还是会存在估计误差的。评判一个预测模型是否具有更广泛性是要通过外部验证的,比如需要一个独立的样本来验证。目前,交叉验证在神经影像的预测模型是必不可少的。

2.4. 预测评估

预测评估的方法是依赖预测变量是离散的还是连续的。分类预测结果有准确率、敏感性(正确辨别出病人的比率)、特异性(正确辨别出控制组的比率)、阳性预测值(真正鉴别出病人的百分比)和负性预测值(真正鉴别出控制组的百分比),这些指标还是取决于病人和控制组有没有明显的疾病差异。回归预测用均方误差(mean-squareerror, MSE)或者相关进行测量评估(Shen, Finn et al., 2017)。然后用可视化的数据评估行为和预测得分或者分类标签的关系。

3. 抑郁症病人的预测研究现状

3.1. 功能激活预测的研究

以往关于抑郁症大脑激活的研究主要是通过抑郁症以健康被试的对比发现激活差异的脑区。比如早期通正电子断层扫描(PET)结果发现rACC (Rostral anterior cingulate)和rACCsubgenual cingulate activity 在抑郁症和控制组上是显著不同的,表明扣带异常代谢可能是抑郁症的独特脑区(Mayberg, Brannan et al., 1997; Pizzagalli, Pascualmarqui et al., 2014)。结构上发现抑郁症病人在杏仁核,右侧海马旁扣带回的灰质体积会更小(Sacher, Neumann et al., 2012)。静息态功能上发现抑郁症病人在左侧海马旁回,右侧小脑后叶,右侧额中回表现出明显的比率低频振幅Fractional Amplitude of Low-Frequency Fluctuation (fALFF)的下降(Liu, Guo et al., 2013),在左侧枕中皮层,右侧顶下小叶,右侧楔前叶表现出局部一致性(Regional Homogeneity, ReHo)的上升(Liang, Zhou et al., 2013)。情绪加工的任务态(Task-based fMRI)发现抑郁症病人在杏仁核和膝下前扣带回存在异常(Langenecker, Kennedy et al., 2007; Costafreda, Khanna et al., 2009; Pizzagalli 2011; Dichter, Gibbs et al., 2015)。不同模态的神经影像用机器学习的预测发现方法发现在结构像上扣带和海马的灰质体积在抑郁症病人对比控制组是能够预测治疗的效果(Costafreda, Chu et al., 2009)。或者通过多模态结合的方法,通过跨模态和多任务预测发现抑郁症病人和控制组的激活差异比如rACC,杏仁核,纹状体,前扣带。(Pizzagalli, 2011; Dichter, Gibbs et al., 2015; Young, Drevets et al., 2016; Crane, Jenkins et al. 2017)。有趣的是,以抑郁症病人进行情绪相关的任务态的研究中视觉区域和小脑与抑郁没有关系的脑区也会出现在结果中,这些结果往往没有的充分的讨论和明确的解释,对于这些大脑区域在抑郁症的潜在归因目前还是缺少理论支持或者假设(Langenecker, Kennedy et al., 2007; López-Solà, Pujol et al., 2010; Pizzagalli, 2011; Dichter, Gibbs et al., 2015)。关于抑郁症激活模式的预测研究较少,目前情绪加工的预测研究中发现杏仁核和膝下前扣带回的异常的预测特是稳定的。

3.2. 脑网络的现状研究和预测研究

静息态影像数据结合脑网络拓扑属性的研究发现抑郁症存在神经网络的异常协调,并不是单个脑区调控的异常。异常的网络包括默认网络、凸显网络、执行控制网络和情感网络,还有部分边缘系统的参与。凸显网络是注意环境中的刺激,由于抑郁症凸显网络的异常管理所以抑郁症病人对比控制组在对负性刺激的解读普遍存在误差(Menon & Uddin, 2010)。情感网络的组成包括膝下扣带膝,前扣带(subgenual and pregenual cingulate)和杏仁核组成,抑郁症病人对比控制组的过度激活会更容易嗜睡(Sheline, Price et al., 2010)。在抑郁病人发现凸显网络和情感网络在对涉及自我相关和强情感刺激信息的加工和处理的优先性是不清楚的,选取两个网络重叠的节点整合成一个更大的网络特征进行跨模态预测,在预测情绪情感的任务中是凸显网络的异常(Langenecker, Kennedy et al., 2007; López-Solà, Pujol et al., 2010; Pizzagalli, 2011; Dichter, Gibbs et al., 2015; Young, Drevets et al., 2016; Crane, Jenkins et al., 2017)。在执行控制网络上的关键节点是背外侧前额叶、顶下叶和北侧前扣带,通常选择背外侧前额叶作为抑郁症治疗的特征(Langenecker, Kennedy et al., 2007; Gyurak, Patenaude et al., 2016)。抑郁症病人存在默认网络激活与任务相反的模式,涉及的脑区大多在中间皮层的区域为特征去预测治疗的效果(Dichter, Felder et al., 2010, Miller, Schneck et al., 2013, Rizvi, Salomons et al., 2013)。目前研究选取的预测特征主要是抑郁症病人存在异常的脑网络中的关键节点实现对治疗效果的预测,但是对于特征稳定性选择上没有一致的结论,未来还需要更为细致的实验世纪或者方法实现对治疗的预测。

3.3. 预测的研究导向

目前在对抑郁症病人基础临床诊断上是不够明确比如常用的DSM-IV,但是利用预测的方法可以使研究者在抑郁的发病,发展和复发的关键因素这些关键的临床问题也能做出更明确的判断,达到在预测抑郁症的治疗的效果减少治疗的次数。最新的研究通过不同神经影像的方法的整合或者对比治疗高准确率的预测治疗效果。研究运用立成分分析技术结合认知控制的实验范式实现90%预测抑郁症病人治疗效果(Crane, Jenkins et al., 2017)。未来研究应该是以预测模型和临床的结合为抑郁症的诊断和治疗的方向发展。

4. 预测的局限

基于神经影像的预测不可避免的要面临数据的预处理,比如头动,处理的不当就会造成结构和功能像的质量存在问题(Pardoe, Kucharsky et al., 2016; Ciric, Wolf et al., 2017),最新的数据采集和和预处理的技术对于准确预测是必要的。除此之外还要面临在训练数据集上预测过拟合,可能是数据样本的噪声引起的。在防止过拟合交叉验证是必要的,通过多站点或新样本验证测试模型。此外,要根据预测的目标选择合适的预测方法,比如是考虑统计预测的显著性还是特征的可解释性,因此在方法的选择上要慎重。抑郁症样本的数量的问题,比如样本足够大微小的效应也可能达到统计显著,数量大的抑郁症样本比小样本更具有普遍性(Swanson, 2012)。但是这些问题都会随着方法学的改进得到改善。虽然预测提升临床和诊断,但是虽然可能实现比较好的分类效果,但是对分类过程中的解释生理心理特点和过程的影响是存在难度。

5. 未来的研究方向

首先基于机器学的构建预测模型的方法是未来神经领域的不可或缺的一部分,尤其是应用在抑郁的诊断和治疗上,这方面的研究较少。抑郁症的发病机制目前认不清楚,从而在个体的治疗上人存在问题,在过去50年抑郁症的药物治疗中,大约有70%的病人是症状改善,还有30%的抑郁症病人是药物不起作用(Stimpson, Agrawal et al., 2002)。使用机器学习的方法对于抑郁中的诊断和治疗是未来的研究方向。其次是预测方法的改进。比如使用独立成分分析和实验范式联系在一起将预测模型结合实现90%准确预测抑郁症病人治疗效果(Crane, Jenkins et al., 2017)。但是预测个体水平实现准确预测还需要进一步研究(Crane, Jenkins et al., 2017)。最后,目前的基于抑郁症神经影像的预测研究的结果呈现多样性,未来元分析的方法能够提供更为有效的预测特征。

文章引用

杨兵兵,李俊男,何 鸿,冯秋阳. 基于机器学习的预测模型对抑郁症的研究进展
Research Progress on Predictive Models Based on Machine Learning for Depression[J]. 心理学进展, 2019, 09(01): 34-40. https://doi.org/10.12677/AP.2019.91005

参考文献

  1. 1. Ball, T. M., Goldsteinpiekarski, A. N., Gatt, J. M., & Williams, L. M. (2017). Quantifying Person-Level Brain Network Functioning to Facilitate Clinical Translation. Translational Psychiatry, 7, e1248.
    https://doi.org/10.1038/tp.2017.204

  2. 2. Blazer, D. G., Kessler, R. C., Mcgonagle, K. A., & Swartz, M. S. (1994). The Prevalence and Distribution of Major Depression in a National Community Sample: The National Comorbidity Survey. American Journal of Psychiatry, 151, 979.
    https://doi.org/10.1176/ajp.151.7.979

  3. 3. Ciric, R., Wolf, D. H., Power, J. D., Roalf, D. R., Baum, G., Ruparel, K., Shinohara, R. T., Elliott, M. A., Eickhoff, S. B., & Davatzikos, C. (2017). Benchmarking of Participant-Level Confound Regression Strategies for the Control of Motion Artifact in Studies of Functional Connectivity. Neuroimage, 154.
    https://doi.org/10.1016/j.neuroimage.2017.03.020

  4. 4. Costafreda, S. G., Chu, C., Ashburner, J., & Fu, C. H. Y. (2009). Prognostic and Diagnostic Potential of the Structural Neuroanatomy of Depression. Plos One, 4, e6353.
    https://doi.org/10.1371/journal.pone.0006353

  5. 5. Costafreda, S. G., Khanna, A., Mourao-Miranda, J., & Fu, C. H. (2009). Neural Correlates of Sad Faces Predict Clinical Remission to Cognitive Behavioural Therapy in Depression. Neuroreport, 20, 637-641.
    https://doi.org/10.1097/WNR.0b013e3283294159

  6. 6. Crane, N. A., Jenkins, L. M., Bhaumik, R., Dion, C., Gowins, J. R., Mickey, B. J., Zubieta, J. K., & Langenecker, S. A. (2017). Multidimensional Prediction of Treatment Response to Antidepressants with Cognitive Control and Functional MRI. Brain, 140, 472-486.
    https://doi.org/10.1093/brain/aww326

  7. 7. Dichter, G. S., Felder, J. N., & Smoski, M. J. (2010). The Effects of Brief Behavioral Activation Therapy for Depression on Cognitive Control in Affective Contexts: An fMRI Investigation. Journal of Affective Disorders, 126, 236.
    https://doi.org/10.1016/j.jad.2010.03.022

  8. 8. Dichter, G. S., Gibbs, D., & Smoski, M. J. (2015). A Systematic Review of Relations between Resting-State Functional-MRI and Treatment Response in Major Depressive Disorder. Journal of Affective Disorders, 172, 8-17.
    https://doi.org/10.1016/j.jad.2014.09.028

  9. 9. Ferrari, A. J., Charlson, F. J., Norman, R. E., Patten, S. B., Freedman, G., Murray, C. J., Vos, T., & Whiteford, H. A. (2013). Burden of Depressive Disorders by Country, Sex, Age, and Year: Findings from the Global Burden of Disease Study 2010. PLOS Medicine, 10, e1001547.
    https://doi.org/10.1371/journal.pmed.1001547

  10. 10. Gabrieli, J. D. E., Ghosh, S. S., & Whitfieldgabrieli, S. (2015). Prediction as a Humanitarian and Pragmatic Contribution from Human Cognitive Neuroscience. Neuron, 85, 11-26.

  11. 11. Guyon, I., & Elisseeff, A. (2003). An Introduction to Variable and Feature Selection. Journal of Machine Learning Research, 3, 1157-1182.

  12. 12. Gyurak, A., Patenaude, B., Korgaonkar, M. S., Grieve, S. M., Williams, L. M., & Etkin, A. (2016). Frontoparietal Activation during Response Inhibition Predicts Remission to Antidepressants in Patients with Major Depression. Biological Psychiatry, 79, 274.
    https://doi.org/10.1016/j.biopsych.2015.02.037

  13. 13. Hahn, T., Kircher, T., Straube, B., Wittchen, H. U., Konrad, C., Ströhle, A., Wittmann, A., Pfleiderer, B., Reif, A., & Arolt, V. (2015). Predicting Treatment Response to Cognitive Behavioral Therapy in Panic Disorder with Agoraphobia by Integrating Local Neural Information. JAMA Psychiatry, 72, 68.
    https://doi.org/10.1001/jamapsychiatry.2014.1741

  14. 14. Jia, Z., Huang, X., Wu, Q., Zhang, T., Lui, S., Zhang, J., Amatya, N., Kuang, W., Chan, R. C., & Kemp, G. J. (2010). High-Field Magnetic Resonance Imaging of Suicidality in Patients with Major Depressive Disorder. American Journal of Psychiatry, 167, 1381.
    https://doi.org/10.1176/appi.ajp.2010.09101513

  15. 15. Kievit, R. A., Brandmaier, A. M., Ziegler, G., van Harmelen, A. L., Smm, D. M., Moutoussis, M., Goodyer, I. M., Bullmore, E., Jones, P. B., & Fonagy, P. (2017). Developmental Cognitive Neuroscience Using Latent Change Score Models: A Tutorial and Applications. Developmental Cognitive Neuroscience, 33, 99-117.
    https://doi.org/10.1101/110429

  16. 16. Langenecker, S. A., Kennedy, S. E., Guidotti, L. M., Briceno, E. M., Own, L. S., Hooven, T., Young, E. A., Akil, H., Noll, D. C., & Zubieta, J. K. (2007). Frontal and Limbic Activation during Inhibitory Control Predicts Treatment Response in Major Depressive Disorder. Biological Psychiatry, 62, 1272-1280.
    https://doi.org/10.1016/j.biopsych.2007.02.019

  17. 17. Liang, M. J., Zhou, Q., Yang, K. R., Yang, X. L., Fang, J., Chen, W. L., & Huang, Z. (2013). Identify Changes of Brain Regional Homogeneity in Bipolar Disorder and Unipolar Depression Using Resting-State fMRI. PLoS ONE, 8, e79999.
    https://doi.org/10.1371/journal.pone.0079999

  18. 18. Liu, F., Guo, W., Liu, L., Long, Z., Ma, C., Xue, Z., Wang, Y., Li, J., Hu, M., & Zhang, J. (2013). Abnormal Amplitude Low-Frequency Oscillations in Medication-Naive, First-Episode Patients with Major Depressive Disorder: A Resting-State fMRI Study. Journal of Affective Disorders, 146, 401-406.
    https://doi.org/10.1016/j.jad.2012.10.001

  19. 19. López-Solà, M., Pujol, J., Hernández-Ribas, R., Harrison, B. J., Contreras-Rodríguez, O., Soriano-Mas, C., Deus, J., Ortiz, H., Menchón, J. M., & Vallejo, J. (2010). Effects of Duloxetine Treatment on Brain Response to Painful Stimulation in Major Depressive Disorder. Neuropsychopharmacology, 35, 2305.
    https://doi.org/10.1038/npp.2010.108

  20. 20. Mayberg, H. S., Brannan, S. K., Mahurin, R. K., Jerabek, P. A., Brickman, J. S., Tekell, J. L., Silva, J. A., Mcginnis, S., Glass, T. G., & Martin, C. C. (1997). Cingulate Function in Depression: A Potential Predictor of Treatment Response. Neuroreport, 8, 1057-1061.
    https://doi.org/10.1097/00001756-199703030-00048

  21. 21. Menon, V., & Uddin, L. Q. (2010). Saliency, Switching, Attention and Control: A Network Model of Insula Function. Brain Structure & Function, 214, 655-667.
    https://doi.org/10.1007/s00429-010-0262-0

  22. 22. Miller, J. M., Schneck, N., Siegle, G. J., Chen, Y., Ogden, R. T., Kikuchi, T., Oquendo, M. A., Mann, J. J., & Parsey, R. V. (2013). fMRI Response to Negative Words and SSRI Treatment Outcome in Major Depressive Disorder: A Preliminary Study. Psychiatry Research, 214, 296-305.
    https://doi.org/10.1016/j.pscychresns.2013.08.001

  23. 23. Pardoe, H. R., Kucharsky, H. R., & Kuzniecky, R. (2016). Motion and Morphometry in Clinical and Nonclinical Populations. Neuroimage, 135, 177.
    https://doi.org/10.1016/j.neuroimage.2016.05.005

  24. 24. Pizzagalli, D. A. (2011). Frontocingulate Dysfunction in Depression: Toward Biomarkers of Treatment Response. Neuropsychopharmacology Official Publication of the American College of Neuropsychopharmacology, 36, 183-206.
    https://doi.org/10.1038/npp.2010.166

  25. 25. Pizzagalli, D., Pascualmarqui, R. D., Nitschke, J. B., Oakes, T. R., Larson, C. L., Abercrombie, H. C., Schaefer, S. M., Koger, J. V., Benca, R. M., & Davidson, R. J. (2014). Anterior Cingulate Activity as a Predictor of Degree of Treatment Response in Major Depression: Evidence from Brain Electrical Tomography Analysis. American Journal of Psychiatry, 158, 405.
    https://doi.org/10.1176/appi.ajp.158.3.405

  26. 26. Rizvi, S. J., Salomons, T. V., Konarski, J. Z., Downar, J., Giacobbe, P., Mcintyre, R. S., & Kennedy, S. H. (2013). Neural Response to Emotional Stimuli Associated with Successful Antidepressant Treatment and Behavioral Activation. Journal of Affective Disorders, 151, 573-581.
    https://doi.org/10.1016/j.jad.2013.06.050

  27. 27. Sacher, J., Neumann, J., & Fünfstück, T. (2012). Mapping the Depressed Brain: A Meta-Analysis of Structural and Functional Alterations in Major Depressive Disorder. Journal of Affective Disorders, 140, 142-148.
    https://doi.org/10.1016/j.jad.2011.08.001

  28. 28. Sheline, Y. I., Price, J. L., Yan, Z., & Mintun, M. A. (2010). Resting-State Functional MRI in Depression Unmasks Increased Connectivity between Networks via the Dorsal Nexus. Proceedings of the National Academy of Sciences of the United States of America, 107, 11020.
    https://doi.org/10.1073/pnas.1000446107

  29. 29. Shen, X., Finn, E. S., Scheinost, D., Rosenberg, M. D., Chun, M. M., Papademetris, X., & Constable, R. T. (2017). Using Connectome-Based Predictive Modeling to Predict Individual Behavior from Brain Connectivity. Nature Protocols, 12, 506.
    https://doi.org/10.1038/nprot.2016.178

  30. 30. Stimpson, N., Agrawal, N., & Lewis, G. (2002). Randomised Controlled Trials Investigating Pharmacological and Psychological Interventions for Treatment-Refractory Depression. Systematic Review. British Journal of Psychiatry, 181, 284.
    https://doi.org/10.1192/bjp.181.4.284

  31. 31. Swanson, J. M. (2012). The UK Biobank and Selection Bias. The Lancet, 380, 110.

  32. 32. Whelan, R., Watts, R., Orr, C. A., Althoff, R. R., Artiges, E., Banaschewski, T., Barker, G. J., Bokde, A. L. W., Büchel, C., & Carvalho, F. M. (2014). Neuropsychosocial Profiles of Current and Future Adolescent Alcohol Misusers. Nature, 512, 185.
    https://doi.org/10.1038/nature13402

  33. 33. Williams, L. M., Korgaonkar, M. S., Song, Y. C., Paton, R., Eagles, S., Goldsteinpiekarski, A. N., Grieve, S. M., Harris, A., Usherwood, T., & Etkin, A. (2015). Amygdala Reactivity to Emotional Faces in the Prediction of General and Medication-Specific Responses to Antidepressant Treatment in the Randomized iSPOT-D Trial. Neuropsychopharmacology, 40, 2398-2408.
    https://doi.org/10.1038/npp.2015.89

  34. 34. Woo, C. W., Chang, L. J., Lindquist, M. A., & Wager, T. D. (2017). Building Better Biomarkers: Brain Models in Translational Neuroimaging. Nature Neuroscience, 20, 365-377.
    https://doi.org/10.1038/nn.4478

  35. 35. Yarkoni, T., & Westfall, J. (2017). Choosing Prediction over Explanation in Psychology: Lessons from Machine Learning. Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 12, 1745691617693393.

  36. 36. Young, K. D., Drevets, W. C., Bodurka, J., & Preskorn, S. S. (2016). Amygdala Activity during Autobiographical Memory Recall as a Biomarker for Residual Symptoms in Patients Remitted from Depression. Psychiatry Research, 248, 159-161.
    https://doi.org/10.1016/j.pscychresns.2016.01.017

期刊菜单