迁移学习和空间协方差模型在研究脑萎缩和白质高信号关系中的应用
|
摘要
目的:为了明确白质高信号的危险因素和发病机制,基于迁移学习和空间协方差模型研究脑萎缩和白质高信号的相互关系及高血压对这种关系的影响。方法:选取美国亚利桑那大学附属医院41名健康中老年人(健康对照组)和41名高血压中老年患者(高血压组)MRI数据。首先,采用基于种子点的模糊聚类算法测量白质高信号的体积。其次,基于上百万张图片预训练得到的卷积神经网络AlexNet采用迁移学习的方法从灰质密度图中提取大脑形态特征。最后,采用基于主成分分析方法的尺度子配置模型来获取与脑白质病变相关的形态学空间协方差模式。结果:与灰度密度图相比,深度学习模型对影像特征具有更好的表达,通过迁移学习提取的特征具有更高的模型拟合优度。与健康对照组相比,高血压组患者的脑部影像对这种特征模式具有更强的表达。结论:基于迁移学习和空间协方差方法可以较好地应用于脑萎缩和白质高信号间关系的研究中。
Objective To examine the relationship between brain atrophy and cerebral small vessel disease and the influence of hypertension on this relationship with transfer learning and spatial covariance method. Methods Totally 41 healty milldeaged and elderly people(enrolled into a control group) and 41 middle-aged and elderly hypertension patients(hypertension group) from the affiliated hospital of The University of Arizona had their MRI data involved into the study. First, a fuzzy clustering method based on seed points was used to measure the volume of white matter hyperintensity(WMH). Second, a convolutional neural network, AlexNet, pretrained on around one million images was used to extract morphometric features from gray matter map. Finally, Scaled Subprofile Model(SSM), a principal components analysis(PCA)-based method was applied to characteristic spatial covariance pattern associated with white matter lesion. Results Compared with gray matter map, deep learning model had a better representation of image features; the features extracted from transfer learning had a higher goodness of fit. It was also found that the pattern was expressed more in the hypertension group than the normotensives. Conclusion The proposed method based on transfer learning and spatial covariance modeling can be used to study the relationship between brain atrophy and white matter hyperintensity.
Objective To examine the relationship between brain atrophy and cerebral small vessel disease and the influence of hypertension on this relationship with transfer learning and spatial covariance method. Methods Totally 41 healty milldeaged and elderly people(enrolled into a control group) and 41 middle-aged and elderly hypertension patients(hypertension group) from the affiliated hospital of The University of Arizona had their MRI data involved into the study. First, a fuzzy clustering method based on seed points was used to measure the volume of white matter hyperintensity(WMH). Second, a convolutional neural network, AlexNet, pretrained on around one million images was used to extract morphometric features from gray matter map. Finally, Scaled Subprofile Model(SSM), a principal components analysis(PCA)-based method was applied to characteristic spatial covariance pattern associated with white matter lesion. Results Compared with gray matter map, deep learning model had a better representation of image features; the features extracted from transfer learning had a higher goodness of fit. It was also found that the pattern was expressed more in the hypertension group than the normotensives. Conclusion The proposed method based on transfer learning and spatial covariance modeling can be used to study the relationship between brain atrophy and white matter hyperintensity.
引文
[1]THOMAS A J,PERRY R,BARBER R,et al.Pathologies and pathological mechanisms for white matter hyperintensities in depression[J].Ann N Y Acad Sci,2002,977:333-339.
[2]YLIKOSKI A,ERKINJUNTTI T,RAIN-INKO R,et al.White matter hyperintensities on MRI in the neurologically nondiseased elderly.Analysis of cohorts of consecutive subjects aged 55 to 85 years living at home[J].Stroke,1995,26(7):1 171-1 177.
[3]GARDE E,MORTENSEN E L,KRABBE K,et al.Relation between age-related decline in intelligence and cerebral whitematter hyperintensities in healthy octogenarians:a longitudinal study[J].Lancet,2000,356(9 230):628-634.
[4]KEMPTON M J,GEDDES J R,ETTINGER U,et al.Metaanalysis,database,and meta-regression of 98 structural imaging studies in bipolar disorder[J].Arch Gen Psychiatry,2008,65(9):1 017-1 032.
[5]VIDEBECH P.MRI findings in patients with affective disorder:a meta-analysis[J].Acta Psychiatr Scand,1997,96(3):157-168.
[6]CAPIZZANO A A,ACION L,BEKINSCHTEIN T,et al.White matter hyperintensities are significantly associated with cortical atrophy in Alzheimer's disease[J].J Neurol Neurosurg Psychiatry,2004,75(6):822-827.
[7]O'SULLIVAN M,LYTHGOE D J,PEREIRA A C,et al.Patterns of cerebral blood flow reduction in patients with ischemic leukoaraiosis[J].Neurology,2002,59(3):321-326.
[8]SCHMIDT R,SCHELTENS P,ERKINJUNTTI T,et al.Whitematter lesion progression:a surrogate endpoint for trials in cerebral small-vessel disease[J].Neurology,2004,63(1):139-144.
[9]DECARLI C,FLETCHER E,RAMEY V,et al.Anatomical mapping of white matter hyperintensities(WMH):exploring the relationships between periventricular WMH,deep WMH,and total WMH burden[J].Stroke,2005,36(1):50-55.
[10]林岚,张柏雯,付振荣,等.运用空间协方差模型分析高血压对大鼠大脑老化的影响[J].中国医疗设备,2016,31(2):34-38.
[11]林岚,张柏雯,付振荣,等.高血压对大脑年龄估值差的影响[J].中国医疗设备,2015,30(6):7-11.
[12]田苗,林岚,赵寅,等.高血压对健康老年人大脑网络特征的影响[J].智慧健康,2017(2):9-15.
[13]RAJI C A,LOPEZ O L,KULLER L H,et al.White matter lesions and brain gray matter volume in cognitively normal elders[J].Neurobiol Aging,2012,33(4):834.e7-834.e16.
[14]ASHBURNER J.A fast diffeomorphic image registration algorithm[J].Neuroimage,2007,38(1):95-113.
[15]田苗,林岚,张柏雯,等.深度学习在神经影像中的应用研究[J].中国医疗设备,2016,31(12):4-9.
[16]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks[C]//The 25th International Conference on Neural Information Processing Systems,December 3-6,2012,Lake Tahoe,Nevada.New York:IEEE,2012:1 097-1 105.
[17]MOELLER J R,STROTHER S C,SIDTIS J J,et al.Scaled subprofile model:a statistical approach to the analysis of functional patterns in positron emission tomographic data[J].J Cereb Blood Flow Metab,1987,7(5):649-658.
[18]林岚,靳聪,付振荣,等.健康老年人脑年龄预测:基于尺度子配置模型的大脑连接组分析[J].北京工业大学学报,2015,41(6):955-960.
[19]BURNHAM K P,ANDERSON D R.Model selection and multimodel inference:a practical information-theoretic approach[M].Heidelberg:Springer-Verlag,2002.
[2]YLIKOSKI A,ERKINJUNTTI T,RAIN-INKO R,et al.White matter hyperintensities on MRI in the neurologically nondiseased elderly.Analysis of cohorts of consecutive subjects aged 55 to 85 years living at home[J].Stroke,1995,26(7):1 171-1 177.
[3]GARDE E,MORTENSEN E L,KRABBE K,et al.Relation between age-related decline in intelligence and cerebral whitematter hyperintensities in healthy octogenarians:a longitudinal study[J].Lancet,2000,356(9 230):628-634.
[4]KEMPTON M J,GEDDES J R,ETTINGER U,et al.Metaanalysis,database,and meta-regression of 98 structural imaging studies in bipolar disorder[J].Arch Gen Psychiatry,2008,65(9):1 017-1 032.
[5]VIDEBECH P.MRI findings in patients with affective disorder:a meta-analysis[J].Acta Psychiatr Scand,1997,96(3):157-168.
[6]CAPIZZANO A A,ACION L,BEKINSCHTEIN T,et al.White matter hyperintensities are significantly associated with cortical atrophy in Alzheimer's disease[J].J Neurol Neurosurg Psychiatry,2004,75(6):822-827.
[7]O'SULLIVAN M,LYTHGOE D J,PEREIRA A C,et al.Patterns of cerebral blood flow reduction in patients with ischemic leukoaraiosis[J].Neurology,2002,59(3):321-326.
[8]SCHMIDT R,SCHELTENS P,ERKINJUNTTI T,et al.Whitematter lesion progression:a surrogate endpoint for trials in cerebral small-vessel disease[J].Neurology,2004,63(1):139-144.
[9]DECARLI C,FLETCHER E,RAMEY V,et al.Anatomical mapping of white matter hyperintensities(WMH):exploring the relationships between periventricular WMH,deep WMH,and total WMH burden[J].Stroke,2005,36(1):50-55.
[10]林岚,张柏雯,付振荣,等.运用空间协方差模型分析高血压对大鼠大脑老化的影响[J].中国医疗设备,2016,31(2):34-38.
[11]林岚,张柏雯,付振荣,等.高血压对大脑年龄估值差的影响[J].中国医疗设备,2015,30(6):7-11.
[12]田苗,林岚,赵寅,等.高血压对健康老年人大脑网络特征的影响[J].智慧健康,2017(2):9-15.
[13]RAJI C A,LOPEZ O L,KULLER L H,et al.White matter lesions and brain gray matter volume in cognitively normal elders[J].Neurobiol Aging,2012,33(4):834.e7-834.e16.
[14]ASHBURNER J.A fast diffeomorphic image registration algorithm[J].Neuroimage,2007,38(1):95-113.
[15]田苗,林岚,张柏雯,等.深度学习在神经影像中的应用研究[J].中国医疗设备,2016,31(12):4-9.
[16]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks[C]//The 25th International Conference on Neural Information Processing Systems,December 3-6,2012,Lake Tahoe,Nevada.New York:IEEE,2012:1 097-1 105.
[17]MOELLER J R,STROTHER S C,SIDTIS J J,et al.Scaled subprofile model:a statistical approach to the analysis of functional patterns in positron emission tomographic data[J].J Cereb Blood Flow Metab,1987,7(5):649-658.
[18]林岚,靳聪,付振荣,等.健康老年人脑年龄预测:基于尺度子配置模型的大脑连接组分析[J].北京工业大学学报,2015,41(6):955-960.
[19]BURNHAM K P,ANDERSON D R.Model selection and multimodel inference:a practical information-theoretic approach[M].Heidelberg:Springer-Verlag,2002.