《计算机应用》唯一官方网站 ›› 2024, Vol. 44 ›› Issue (12): 3949-3957.DOI: 10.11772/j.issn.1001-9081.2023121764
收稿日期:
2023-12-21
修回日期:
2024-04-03
接受日期:
2024-04-07
发布日期:
2024-04-28
出版日期:
2024-12-10
通讯作者:
高保禄
作者简介:
沈嫣然(1997—),女,山西长治人,硕士研究生,CCF会员,主要研究方向:脑科学、深度学习基金资助:
Yanran SHEN, Xin WEN, Jinhao ZHANG, Shuai ZHANG, Rui CAO, Baolu GAO()
Received:
2023-12-21
Revised:
2024-04-03
Accepted:
2024-04-07
Online:
2024-04-28
Published:
2024-12-10
Contact:
Baolu GAO
About author:
SHEN Yanran, born in 1997, M. S. candidate. Her research interests include brain science, deep learning.Supported by:
摘要:
针对功能磁共振成像(fMRI)脑龄预测精度较低以及该问题与深度学习结合研究较少的现状,提出一种轻量级多尺度卷积网络的fMRI脑龄预测模型(LMCN)。首先,通过计算fMRI中感兴趣区(ROI)的皮尔逊相关系数(R)得到ROI的功能连接(FC)矩阵,并将该矩阵作为输入;其次,在提升FC通道数保证特征数的同时缩小特征图尺寸,并采用具有人类视觉注意力特点的多尺度空洞卷积模块RFB(Receptive Field Block)提取年龄特征;最后,通过全连接层输出预测脑龄,并计算各脑区的消融预测结果,从而探索对脑龄预测结果产生影响的关键脑区。在E-NKI和Cam-CAN这2个公开数据集上评估,可知LMCN参数所需内存为2.30 MB,比MobileNetV3、ShuffleNetV2分别减少了60.3%、52.0%。预测结果方面,在E-NKI数据集上,LMCN的平均绝对误差(MAE)为5.16,R为0.947,均方根误差(RMSE)为6.40,与基于网络的特征选择结合最小角回归模型相比,MAE减小了1.34,R增加了0.037;在Cam-CAN数据集上,LMCN的MAE为5.97,R为0.904,RMSE为7.93,与基于连接组的机器学习模型相比,R提升了0.019,RMSE减小了0.64。结果表明,LMCN在参数量较小易于部署的同时,能够有效提高fMRI脑龄预测的精度,并为评估健康成人的脑部状态提供线索。
中图分类号:
沈嫣然, 温昕, 张瑾昊, 张帅, 曹锐, 高保禄. 轻量级多尺度卷积网络的功能磁共振成像脑龄预测模型[J]. 计算机应用, 2024, 44(12): 3949-3957.
Yanran SHEN, Xin WEN, Jinhao ZHANG, Shuai ZHANG, Rui CAO, Baolu GAO. fMRI brain age prediction model with lightweight multi-scale convolutional network[J]. Journal of Computer Applications, 2024, 44(12): 3949-3957.
数据集 | 人数 | 年龄范围 | 年龄均值±方差 | 性别(男/女) |
---|---|---|---|---|
E-NKI | 350 | 18~85 | 49.75±19.21 | 130/220 |
Cam-CAN | 640 | 18~87 | 54.29±18.52 | 315/325 |
表1 数据集统计情况
Tab. 1 Statistics of datasets
数据集 | 人数 | 年龄范围 | 年龄均值±方差 | 性别(男/女) |
---|---|---|---|---|
E-NKI | 350 | 18~85 | 49.75±19.21 | 130/220 |
Cam-CAN | 640 | 18~87 | 54.29±18.52 | 315/325 |
模型 | 参数文件 大小/MB | MAE | |
---|---|---|---|
E-NKI | Cam-CAN | ||
ResNet18[ | 42.61 | 8.70 | 13.56 |
GoogLeNet[ | 23.98 | 7.46 | 11.72 |
MobileNetV3[ | 5.79 | 12.26 | 16.56 |
ShuffleNetV2[ | 4.79 | 11.93 | 15.18 |
LMCN | 2.30 | 5.16 | 5.97 |
表2 不同模型的对比实验结果
Tab. 2 Comparison experimental results of different models
模型 | 参数文件 大小/MB | MAE | |
---|---|---|---|
E-NKI | Cam-CAN | ||
ResNet18[ | 42.61 | 8.70 | 13.56 |
GoogLeNet[ | 23.98 | 7.46 | 11.72 |
MobileNetV3[ | 5.79 | 12.26 | 16.56 |
ShuffleNetV2[ | 4.79 | 11.93 | 15.18 |
LMCN | 2.30 | 5.16 | 5.97 |
模型 | 数据集 | 人数 | 年龄范围 | MAE | R | RMSE |
---|---|---|---|---|---|---|
文献[ | ADNI | 471 | 51~95 | 5.92 | 7.56 | |
文献[ | NKI | 125 | 12~85 | 5.14 | 0.875 | |
文献[ | Cam-CAN | 567 | 19~89 | 0.885 | 8.57 | |
文献[ | E-NKI | 496 | 6~85 | 6.50 | 0.910 | |
文献[ | 391 | 18~89 | 8.59 | 0.826 | 10.44 | |
本文模型 | E-NKI | 350 | 18~85 | 5.16 | 0.947 | 6.40 |
Cam-CAN | 640 | 18~87 | 5.97 | 0.904 | 7.93 |
表3 相关研究对比
Tab. 3 Comparison of related studies
模型 | 数据集 | 人数 | 年龄范围 | MAE | R | RMSE |
---|---|---|---|---|---|---|
文献[ | ADNI | 471 | 51~95 | 5.92 | 7.56 | |
文献[ | NKI | 125 | 12~85 | 5.14 | 0.875 | |
文献[ | Cam-CAN | 567 | 19~89 | 0.885 | 8.57 | |
文献[ | E-NKI | 496 | 6~85 | 6.50 | 0.910 | |
文献[ | 391 | 18~89 | 8.59 | 0.826 | 10.44 | |
本文模型 | E-NKI | 350 | 18~85 | 5.16 | 0.947 | 6.40 |
Cam-CAN | 640 | 18~87 | 5.97 | 0.904 | 7.93 |
消融后 性能情况 | 中文名 | 英文名 |
---|---|---|
预测 性能 变差 | 颞级:颞中回 | Temporal pole: middle temporal gyrus |
楔前叶 | Precuneus | |
豆状苍白球 | Lenticular nucleus, pallidum | |
颞横回 | Heschl gyrus | |
嗅皮质 | Olfactory cortex | |
海马 | Hippocampus | |
角回 | Angular gyrus | |
岛盖部额下回 | Inferior frontal gyrus, opercular part | |
顶下缘角回 | Inferior parietal, but supramarginal and angular gyri | |
中央前回 | Precentral gyrus | |
杏仁核 | Amygdala | |
预测 性能 变好 | 缘上回 | Supramarginal gyrus |
内侧额上回 | Superior frontal gyrus, medial | |
枕中回 | Middle occipital gyrus |
表4 消融后预测性能变差和变好的脑区
Tab. 4 Brain regions with worse and better prediction performance after ablation
消融后 性能情况 | 中文名 | 英文名 |
---|---|---|
预测 性能 变差 | 颞级:颞中回 | Temporal pole: middle temporal gyrus |
楔前叶 | Precuneus | |
豆状苍白球 | Lenticular nucleus, pallidum | |
颞横回 | Heschl gyrus | |
嗅皮质 | Olfactory cortex | |
海马 | Hippocampus | |
角回 | Angular gyrus | |
岛盖部额下回 | Inferior frontal gyrus, opercular part | |
顶下缘角回 | Inferior parietal, but supramarginal and angular gyri | |
中央前回 | Precentral gyrus | |
杏仁核 | Amygdala | |
预测 性能 变好 | 缘上回 | Supramarginal gyrus |
内侧额上回 | Superior frontal gyrus, medial | |
枕中回 | Middle occipital gyrus |
1 | GRADY C. The cognitive neuroscience of ageing[J]. Nature Reviews Neuroscience, 2012, 13(7): 491-505. |
2 | DE LANGE A M G, ANATÜRK M, ROKICKI J, et al. Mind the gap: performance metric evaluation in brain-age prediction[J]. Human Brain Mapping, 2022, 43(10): 3113-3129. |
3 | SMITH S M, ELLIOTT L T, ALFARO-ALMAGRO F, et al. Brain aging comprises many modes of structural and functional change with distinct genetic and biophysical associations [J]. eLife, 2020, 9: No.e52677. |
4 | FRANKE K, GASER C. Ten years of BrainAGE as a neuroimaging biomarker of brain aging: what insights have we gained?[J]. Frontiers in Neurology, 2019, 10: No.789. |
5 | COLE J H, FRANKE K. Predicting age using neuroimaging: innovative brain ageing biomarkers [J]. Trends in Neurosciences, 2017, 40(12): 681-690. |
6 | GIANNAKOPOULOS P, MONTANDON M L, HERRMANN F R, et al. Alzheimer resemblance atrophy index, BrainAGE, and normal pressure hydrocephalus score in the prediction of subtle cognitive decline: added value compared to existing MR imaging markers [J]. European Radiology, 2022, 32(11): 7833-7842. |
7 | CHERUBINI A, CALIGIURI M E, PÉRAN P, et al. Importance of multimodal MRI in characterizing brain tissue and its potential application for individual age prediction[J]. IEEE Journal of Biomedical and Health Informatics, 2016, 20(5): 1232-1239. |
8 | LOGOTHETIS N K. The neural basis of the blood-oxygen-level-dependent functional magnetic resonance imaging signal[J]. Philosophical Transactions of the Royal Society of London Series B, 2002, 357: 1003-1037. |
9 | EDLOW B L, CLAASSEN J, SCHIFF N D, et al. Recovery from disorders of consciousness: mechanisms, prognosis and emerging therapies [J]. Nature Reviews Neurology, 2020, 17(3): 135-156. |
10 | 黄嘉爽,接标,丁卫平,等. 脑网络分析方法及其应用[J]. 数据采集与处理, 2021, 36(4): 648-663. |
HUANG J S, JIE B, DING W P, et al. Brain network analysis: method and application[J]. Journal of Data Acquisition and Processing, 2021, 36(4): 648-663. | |
11 | GREENE A S, GAO S, SCHEINOST D, et al. Task-induced brain state manipulation improves prediction of individual traits[J]. Nature Communications, 2018, 9: No.2807. |
12 | ANDERSON N D. Cognitive neuroscience of aging[J]. The Journals of Gerontology: Series B, 2019, 74(7): 1083-1085. |
13 | 范晨雨,李浩正,谢鸿宇,等. 基于功能性近红外光谱技术的健康青年人、老年人皮层脑网络静息态功能连接的特征研究 [J]. 中国康复医学杂志, 2021, 36(8):931-937, 942. |
FAN C Y, LI H Z, XIE H Y, et al. Healthy young and old adults functional connectivity among cortical networks: a resting-state functional near-infrared spectroscopy study [J]. Chinese Journal of Rehabilitation Medicine, 2021, 36(8):931-937, 942. | |
14 | GEERLIGS L, RENKEN R J, SALIASI E, et al. A brain-wide study of age-related changes in functional connectivity[J]. Cerebral Cortex, 2015, 25(7): 1987-1999. |
15 | 刘涵慧,李会杰. 老化认知神经科学:研究现状与未来展望[J]. 中国科学:生命科学, 2021, 51(6):743-763. |
LIU H H, LI H J. Cognitive neuroscience of aging: present research status and future prospect [J]. SCIENTIA SINICA Vitae, 2021, 51(6):743-763. | |
16 | ZONNEVELD H I, PRUIM R H R, BOS D, et al. Patterns of functional connectivity in an aging population: the Rotterdam Study[J]. NeuroImage, 2019, 189: 432-444. |
17 | MOWINCKEL A M, ESPESETH T, WESTLYE L T. Network-specific effects of age and in-scanner subject motion: a resting-state fMRI study of 238 healthy adults [J]. NeuroImage, 2012, 63(3): 1364-1373. |
18 | YIN W, LI L, WU F X. Deep learning for brain disorder diagnosis based on fMRI images[J]. Neurocomputing, 2022, 469: 332-345. |
19 | COLE J H, POUDEL R P K, TSAGKRASOULIS D, et al. Predicting brain age with deep learning from raw imaging data results in a reliable and heritable biomarker [J]. NeuroImage, 2017, 163: 115-124. |
20 | LUNDERVOLD A S, LUNDERVOLD A. An overview of deep learning in medical imaging focusing on MRI [J]. Zeitschrift für Medizinische Physik, 2019, 29(2): 102-127. |
21 | GONG W, BECKMANN C F, VEDALDI A, et al. Optimising a simple fully convolutional network for accurate brain age prediction in the PAC 2019 challenge [J]. Frontiers in Psychiatry, 2021, 12: No.627996. |
22 | KHOSLA M, JAMISON K, KUCEYESKI A, et al. Ensemble learning with 3D convolutional neural networks for functional connectome-based prediction[J]. NeuroImage, 2019, 199: 651-662. |
23 | GALAZZO I B, CRUCIANI F, BRUSINI L, et al. Explainable artificial intelligence for magnetic resonance imaging aging brainprints: grounds and challenges [J]. IEEE Signal Processing Magazine, 2022, 39(2): 99-116. |
24 | GAO J, LIU J, XU Y, et al. Brain age prediction using the graph neural network based on resting-state functional MRI in Alzheimer's disease [J]. Frontiers in Neuroscience, 2023, 17: No.1222751. |
25 | FISCH L, LEENINGS R, WINTER N R, et al. Editorial: predicting chronological age from structural neuroimaging: the Predictive Analytics Competition 2019[J]. Frontiers in Psychiatry, 2021, 12: No.710932. |
26 | NOONER K B, COLCOMBE S J, TOBE R H, et al. The NKI-Rockland sample: a model for accelerating the pace of discovery science in psychiatry [J]. Frontiers in Neuroscience, 2012, 6: No.152. |
27 | SHAFTO M A, TYLER L K, DIXON M, et al. The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) study protocol: a cross-sectional, lifespan, multidisciplinary examination of healthy cognitive ageing [J]. BMC Neurology, 2014, 14: No.204. |
28 | YAN C G, WANG X D, ZUO X N, et al. DPABI: data processing & analysis for (resting-state) brain imaging [J]. Neuroinformatics, 2016, 14(3): 339-351. |
29 | SHEN H, WANG L, LIU Y, et al. Discriminative analysis of resting-state functional connectivity patterns of schizophrenia using low dimensional embedding of fMRI [J]. NeuroImage, 2010, 49(4): 3110-3121. |
30 | LIU S, HUANG D, WANG Y. Receptive field block bet for accurate and fast object detection [C]// Proceedings of the 2018 European Conference on Computer Vision, LNCS 11215. Cham: Springer, 2018: 404-419. |
31 | WANG P, CHEN P, YUAN Y, et al. Understanding convolution for semantic segmentation[C]// Proceedings of the 18th IEEE Winter Conference on Applications of Computer Vision. Piscataway: IEEE, 2018: 1451-1460. |
32 | HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition [C]// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2016: 770-778. |
33 | FRANKE K, BUBLAK P, HOYER D, et al. In vivo biomarkers of structural and functional brain development and aging in humans[J]. Neuroscience and Biobehavioral Reviews, 2020, 117: 142-164. |
34 | SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions [C]// Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2015: 1-9. |
35 | HOWARD A, SANDLER M, CHEN B, et al. Searching for MobileNetV3[C]// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway: IEEE, 2019:1314-1324. |
36 | YE L. AugShuffleNet: communicate more, compute less[EB/OL]. [2023-12-01].. |
37 | HAN H, GE S, WANG H. Prediction of brain age based on the community structure of functional networks [J]. Biomedical Signal Processing and Control, 2023, 79(Pt 2): No.104151. |
38 | ZHAI J, LI K. Predicting brain age based on spatial and temporal features of human brain functional networks[J]. Frontiers in Human Neuroscience, 2019, 13: No.62. |
39 | JIANG R, SCHEINOST D, ZUO N, et al. A neuroimaging signature of cognitive aging from whole-brain functional connectivity [J]. Advanced Science, 2022, 9(24): No.2201621. |
40 | MILLAR P R, LUCKETT P H, GORDON B A, et al. Predicting brain age from functional connectivity in symptomatic and preclinical Alzheimer disease[J]. NeuroImage, 2022, 256: No.119228. |
41 | 吕书悦,孙文悦,吴凡,等. 海马亚区在阿尔茨海默病中的研究进展[J]. 重庆医科大学学报, 2021, 46(11):1391-1394. |
LYU S Y, SUN W Y, WU F, et al. Advances in the study of hippocampal subregion in Alzheimer's disease [J]. Journal of Chongqing Medical University, 2021, 46(11): 1391-1394. | |
42 | HISCOX L V, SCHWARB H, McGARRY M D J, et al. Aging brain mechanics: progress and promise of magnetic resonance elastography [J]. NeuroImage, 2021, 232: No.117889. |
43 | 徐家华,周莹,罗文波,等. 人类情绪发展认知神经科学:面向未来心理健康与教育[J]. 中国科学:生命科学, 2021, 51(6):663-678. |
XU J H, ZHOU Y, LUO W B, et al. Human developmental cognitive and affective neuroscience: future-oriented mental health and education[J]. SCIENTIA SINICA Vitae, 2021, 51(6):663-678. | |
44 | WASCHKE L, KLOOSTERMAN N A, OBLESER J, et al. Behavior needs neural variability [J]. Neuron, 2021, 109(5): 751-766. |
45 | NAKAGAWA S, TAKEUCHI H, TAKI Y, et al. Lenticular nucleus correlates of general self-efficacy in young adults [J]. Brain Structure and Function, 2017, 222(7): 3309-3318. |
46 | KONDO K, KIKUTA S, UEHA R, et al. Age-related olfactory dysfunction: epidemiology, pathophysiology, and clinical management [J]. Frontiers in Aging Neuroscience, 2020, 12: No.208. |
47 | MAREK S, DOSENBACH U F. The frontoparietal network: function, electrophysiology, and importance of individual precision mapping [J]. Dialogues in Clinical Neuroscience, 2018, 20(2): 133-140. |
48 | CASSADY K, RUITENBERG M F L, REUTER-LORENZ P A, et al. Neural dedifferentiation across the lifespan in the motor and somatosensory systems [J]. Cerebral Cortex, 2020, 30(6): 3704-3716. |
49 | XUE C, SUN H, YUE Y, et al. Structural and functional disruption of salience network in distinguishing subjective cognitive decline and amnestic mild cognitive impairment[J]. ACS Chemical Neuroscience, 2021, 12(8): 1384-1394. |
50 | PAN N, WANG S, QIN K, et al. Common and distinct neural patterns of attention-deficit/hyperactivity disorder and borderline personality disorder: a multimodal functional and structural meta-analysis [J]. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 2023, 8(6): 640-650. |
[1] | 秦璟, 秦志光, 李发礼, 彭悦恒. 基于概率稀疏自注意力神经网络的重性抑郁疾患诊断[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2970-2974. |
[2] | 王熙源, 张战成, 徐少康, 张宝成, 罗晓清, 胡伏原. 面向手术导航3D/2D配准的无监督跨域迁移网络[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2911-2918. |
[3] | 潘烨新, 杨哲. 基于多级特征双向融合的小目标检测优化模型[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2871-2877. |
[4] | 李顺勇, 李师毅, 胥瑞, 赵兴旺. 基于自注意力融合的不完整多视图聚类算法[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2696-2703. |
[5] | 黄云川, 江永全, 黄骏涛, 杨燕. 基于元图同构网络的分子毒性预测[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2964-2969. |
[6] | 刘禹含, 吉根林, 张红苹. 基于骨架图与混合注意力的视频行人异常检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2551-2557. |
[7] | 顾焰杰, 张英俊, 刘晓倩, 周围, 孙威. 基于时空多图融合的交通流量预测[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2618-2625. |
[8] | 石乾宏, 杨燕, 江永全, 欧阳小草, 范武波, 陈强, 姜涛, 李媛. 面向空气质量预测的多粒度突变拟合网络[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2643-2650. |
[9] | 赵亦群, 张志禹, 董雪. 基于密集残差物理信息神经网络的各向异性旅行时计算方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2310-2318. |
[10] | 徐松, 张文博, 王一帆. 基于时空信息的轻量视频显著性目标检测网络[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2192-2199. |
[11] | 孙逊, 冯睿锋, 陈彦如. 基于深度与实例分割融合的单目3D目标检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2208-2215. |
[12] | 吴筝, 程志友, 汪真天, 汪传建, 王胜, 许辉. 基于深度学习的患者麻醉复苏过程中的头部运动幅度分类方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2258-2263. |
[13] | 李欢欢, 黄添强, 丁雪梅, 罗海峰, 黄丽清. 基于多尺度时空图卷积网络的交通出行需求预测[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2065-2072. |
[14] | 张郅, 李欣, 叶乃夫, 胡凯茜. 基于暗知识保护的模型窃取防御技术DKP[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2080-2086. |
[15] | 赵雅娟, 孟繁军, 徐行健. 在线教育学习者知识追踪综述[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1683-1698. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||