Machine Learning-based Gait Analysis for Recognition of Amnestic Mild Cognitive Impairment and Alzheimer's Disease

doi:10.12114/j.issn.1007-9572.2022.0437

Chinese General Practice ›› 2022, Vol. 25 ›› Issue (31): 3857-3865.DOI: 10.12114/j.issn.1007-9572.2022.0437

Special Issue: 神经退行性病变最新文章合辑；阿尔茨海默病最新文章合辑；脑健康最新研究合辑；数智医疗最新文章合辑

• Article·Focus on Cognitive Impairment • Previous Articles Next Articles

Machine Learning-based Gait Analysis for Recognition of Amnestic Mild Cognitive Impairment and Alzheimer's Disease

1.Dalian Key Laboratory of Smart Medical and Health, Dalian University, Dalian 116622, China
2.Department of Neurology, the First People's Hospital of Foshan, Foshan 528010, China

Received:2022-06-22 Revised:2022-08-06 Published:2022-11-05 Online:2022-09-19
Contact: XIE Haiqun
About author:
TAO S, HAN X, KONG L W, et al. Machine learning-based gait analysis for recognition of amnestic mild cognitive impairment and Alzheimer's disease [J] . Chinese General Practice, 2022, 25 (31) : 3857-3865.

基于步态的机器学习模型识别遗忘型轻度认知障碍和阿尔茨海默病

1.116622　辽宁省大连市，大连大学大连市智慧医疗与健康重点实验室
2.528010　广东省佛山市，佛山市第一人民医院神经内科

通讯作者: 谢海群
作者简介:
陶帅，韩星，孔丽文，等.基于步态的机器学习模型识别遗忘型轻度认知障碍和阿尔茨海默病[J].中国全科医学，2022，25（31）：3857-3865.[www.chinagp.net] 作者贡献：陶帅负责资金提供、调查开展、概念提出；韩星负责形式分析、方法学、软件、原稿创作；孔丽文负责项目管理、监督、验证；汪祖民负责可视化、审查和写作；谢海群负责数据管理、资源提供。
基金资助:
国家重点研发计划(2018YFC2001700)

Abstract

Abstract:

Background

The prevalence of age-related cognitive impairment, including dementia, has significantly increased with population aging. It has been shown that cognitive function is associated with gait status. Previously, researchers used statistical analysis methods instead of machine learning methods to study the gait of amnestic mild cognitive impairment (aMCI) and Alzheimer's disease (AD) .

Objective

To develop a model to identify aMCI and AD based on gait status using machine learning methods, explore gait markers differentiating between aMCI and AD, and to assess their possible values as aided tools in diagnosing aMCI and AD.

Methods

We recruited 102 cases from the Rehabilitation Hospital Affiliated to National Research Center for Rehabilitation Technical Aids, the First People's Hospital of Foshan, and Affiliated Zhongshan Hospital of Dalian University from December 2018 to December 2020, and included 98 of them according to the screening criteria, including 55 patients with aMCI, 10 patients with AD, and 33 healthy controls (HC) . The gait parameters of the participants were collected during performing single-task (free walking) , dual-task (counting backwards in sevens) and another dual-task (counting backwards from 100) using a wearable device. Random forest (RF) algorithm and gradient boosting decision tree (GBDT) algorithm were separately used to establish a model to compare the effect of two algorithms in recognizing three groups, with 10 gait parameters as predictive variables and the physical status (healthy, aMCI, AD) as response variables. Then important features were chosen using a machine learning algorithm combined with recursive feature elimination (RFE) .

Results

No statistically significant differences were found among the three groups in terms of sex ratio, average age, height, body weight or shoe size (P>0.05) , while the differences in terms of average MMSE score and MoCA score were statistically significant (P<0.05) . In the free walking test, aMCI group and AD group had shorter average stride length and smaller average heel-to-ground angle (HtA) than HC group (P<0.05) . AD group had slower average gait speed and smaller average toe-off angle (ToA) than both HC group and aMCI group (P<0.05) . In performing the dual-task of counting backwards in sevens, compared with HC group, aMCI group and AD group had slower average gait speed and smaller average ToA and HtA (P<0.05) . AD group had longer average stance phase than HC group (P<0.05) . AD group had average smaller ToA than aMCI group (P<0.05) .In performing the dual-task of counting backwards from 100, AD group had slower average gait speed and smaller average HtA and ToA than both HC group and aMCI group (P<0.05) . Moreover, AD group had shorter average stride length than HC group (P<0.05) . The average HtA in aMCI group was smaller than that in HC group (P<0.05) . Using the GBDT-RFE method, we found important gait features in distinguishing between aMCI and AD to be the stride length, ToA and HtA, and the model using the RF algorithm performed better in identifying aMCI and AD, with an accuracy as high as 87.69%.

Conclusion

Stride length, ToA and HtA are important gait markers to identify aMCI and AD. These findings could help clinicians diagnose aMCI and AD in the future.

Key words: Cognitive dysfunction, Amnestic mild cognitive impairment, Alzheimer's disease, Gait analysis, Random forest, Gradient boosting decision tree

摘要：

背景

随着老龄化社会的到来，与年龄密切相关的认知障碍(包括痴呆)的患病率明显增加。先前的研究表明，具有不同认知能力的人群所表现的步态状态也不一样。过去研究者们在研究遗忘型轻度认知障碍（aMCI）和阿尔茨海默病（AD）的步态时，使用了统计分析方法，对机器学习方法的使用较少。

目的

构建基于步态的机器学习模型识别aMCI和AD，探索aMCI和AD之间的步态标志物，以便将其用作帮助诊断aMCI患者和AD患者的可能工具。

方法

于2018年12月至2020年12月，从国家康复辅具研究中心附属康复医院、佛山市第一人民医院、大连大学附属中山医院招募了102例受试者，按照筛选标准最终纳入98例受试者，其中55例为aMCI患者，10例为AD患者，33例为健康对照（HC）者。使用可穿戴设备采集参与者在单任务（自由行走）、双任务（倍数7）和双任务（倒数100）时的步态参数。使用随机森林算法（RF）和梯度提升决策树算法（GBDT）建立模型，10个步态参数作为预测变量，疾病状态（HC、aMCI、AD）作为响应变量，比较两种机器学习算法对3个疾病组的识别效果。然后使用机器学习算法结合递归特征消除法（RFE）进行重要特征选择。

结果

三组年龄、性别、身高、体质量、鞋码比较，差异无统计学意义（P>0.05）；MMSE评分、MoCA评分比较，差异有统计学意义（P<0.05）。自由行走测试时，aMCI组和AD组受试者步幅较HC组短，足跟着地角度较HC组小；AD组步速较HC组和aMCI组受试者慢，足趾离地角度较HC组小（P<0.05）。双任务倍数7测试时，aMCI组和AD组受试者步速较HC组慢，足趾离地角度和足跟着地角度较HC组小；AD组支撑时间较HC组长，足趾离地角度较aMCI组小（P<0.05）。双任务倒数100测试时，AD组步速较HC组和aMCI组受试者慢，足趾离地角度和足跟着地角度较HC组和aMCI组小，步幅较HC组短；aMCI组足跟着地角度较HC组小（P<0.05）。GBDT-RFE方法发现aMCI和AD之间的重要步态特征是步幅、足趾离地角度和足跟着地角度，并在RF模型中实现了识别aMCI和AD的最佳性能，最高准确率为87.69%。

结论

步幅、足趾离地角度和足跟着地角度是识别aMCI患者和AD患者的重要步态标志物，未来临床医生可依据重要步态标志物诊断和治疗aMCI患者和AD患者。

关键词: 认知功能障碍, 遗忘型轻度认知障碍, 阿尔茨海默病, 步态分析, 随机森林算法, 梯度提升决策树算法

TAO Shuai,HAN Xing,KONG Liwen, et al. Machine Learning-based Gait Analysis for Recognition of Amnestic Mild Cognitive Impairment and Alzheimer's Disease[J]. Chinese General Practice, 2022, 25(31): 3857-3865. DOI: 10.12114/j.issn.1007-9572.2022.0437.
陶帅,韩星,孔丽文等. 基于步态的机器学习模型识别遗忘型轻度认知障碍和阿尔茨海默病[J]. 中国全科医学, 2022, 25(31): 3857-3865. DOI: 10.12114/j.issn.1007-9572.2022.0437.

Figures/Tables 8

References 48

[1]	GAWRONSKI K A B, KIM E S, LANGA K M，et al. Dispositional optimism and incidence of cognitive impairment in older adults[J]. Psychosomatic Medicine，2016，78（7）：819-828.
[2]	LANE C A, HARDY J, SCHOTT J M. Alzheimer's disease[J]. European Journal of Neurology，2018，25（1）：59-70. DOI：10.1111/ene.13439.
[3]	TANNENBAUM C, PAQUETTE A, HILMER S，et al. A systematic review of amnestic and non-amnestic mild cognitive impairment induced by anticholinergic，antihistamine，GABAergic and opioid drugs[J]. Drugs & Aging，2012，29（8）：639-658.
[4]	AHMED S, MITCHELL J, ARNOLD R，et al. Predicting rapid clinical progression in amnestic mild cognitive impairment[J]. Dementia and Geriatric Cognitive Disorders，2008，25（2）：170-177. DOI：10.1159/000113014.
[5]	BAHUREKSA L, NAJAFI B, SALEH A，et al. The impact of mild cognitive impairment on gait and balance：a systematic review and meta-analysis of studies using instrumented assessment[J]. Gerontology，2017，63（1）：67-83. DOI：10.1159/000445831.
[6]	SCHERDER E, EGGERMONT L, SWAAB D，et al. Gait in ageing and associated dementias：its relationship with cognition[J]. Neuroscience & Biobehavioral Reviews，2007，31（4）：485-497.
[7]	ROSSO A L, VERGHESE J, METTI A L，et al. Slowing gait and risk for cognitive impairment：the hippocampus as a shared neural substrate[J]. Neurology，2017，89（4）：336-342.
[8]	BEAUCHET O, BLUMEN H M, CALLISAYA M L，et al. Spatiotemporal gait characteristics associated with cognitive impairment：a multicenter cross-sectional study，the intercontinental "Gait，Cognition & Decline" initiative[J]. Current Alzheimer Research，2018，15（3）：273-282. DOI：10.1038/s41598-019-53656-7.
[9]	DE OLIVEIRA SILVA F, FERREIRA J V, PLÁCIDO J，et al. Gait analysis with videogrammetry can differentiate healthy elderly，mild cognitive impairment，and Alzheimer's disease：a cross-sectional study[J]. Experimental Gerontology，2020，131：110816. DOI：10.1016/j.exger.2019.110816.
[10]	BERG L. Clinical dementia rating[J]. British Journal of Psychiatry，1984，145（3）：339-339. DOI：10.1192/S0007125000118082.
[11]	DUBOIS B, FELDMAN H H, JACOVA C，et al. Research criteria for the diagnosis of Alzheimer's disease：revising the NINCDS-ADRDA criteria[J]. Lancet Neurology，2007，6（8）：734-746. DOI：10.1016/S1474-4422(07)70178-3.
[12]	BRAEKHUS A, LAAKE K, ENGEDAL K. The Mini-Mental State Examination：identifying the most efficient variables for detecting cognitive impairment in the elderly[J]. Journal of the American Geriatrics Society，1992，40（11）：1139-1145. DOI：10.1111/j.1532-5415.1992.tb01804.x.
[13]	PINTO T C C, MACHADO L, BULGACOV T M，et al. Is the Montreal Cognitive Assessment（MoCA） screening superior to the Mini-Mental State Examination（MMSE） in the detection of mild cognitive impairment（MCI） and Alzheimer's Disease（AD） in the elderly[J]. International Psychogeriatrics，2019，31（4）：491-504. DOI：10.1017/S1041610218001370.
[14]	VAKIL E, BLACHSTEIN H. Rey auditory-verbal learning test：structure analysis[J]. Journal of Clinical Psychology，1993，49（6）：883-890. DOI：10.1002/1097-4679(199311)49:63.0.CO;2-6.
[15]	JENSEN A R, ROHWER JR W D. The Stroop color-word test：a review[J]. Acta Psychologica，1966，25：36-93. DOI：10.1016/0001-6918(66)90004-7.
[16]	PASCOE M, ALAMRI Y, DALRYMPLE-ALFORD J，et al. The symbol-digit modalities test in mild cognitive impairment：evidence from Parkinson's disease patients[J]. European Neurology，2018，79（3/4）：206-210. DOI：10.1159/000485669.
[17]	WILLIAMS B W, MACK W, HENDERSON V W. Boston naming test in Alzheimer's disease[J]. Neuropsychologia，1989，27（8）：1073-1079. DOI：10.1016/0028-3932(89)90186-3.
[18]	BUCKS R S, ASHWORTH D L, WILCOCK G K，et al. Assessment of activities of daily living in dementia：development of the Bristol Activities of Daily Living Scale[J]. Age and Ageing，1996，25（2）：113-120. DOI：10.1093/ageing/25.2.113.
[19]	SHUAI T, ZEPING L, HAIQUN X. Application of wearable gait-assist technology in the fields of rehabilitation and elderly care[J]. Science & Technology Review，2019，37（22）：19-25.
[20]	BREIMAN L. Random forests[J]. Machine Learning，2001，45（1）：5-32. DOI：10.1023/A:1010933404324.
[21]	OSHIRO T M, PEREZ P S, BARANAUSKAS J A. How many trees in a random forest[EB/OL]. [2022-04-03].
[22]	PHAN D, NGUYEN N, PATHIRANA P N，et al. A random forest approach for quantifying gait ataxia with truncal and peripheral measurements using multiple wearable sensors[J]. IEEE Sensors Journal，2019，20（2）：723-734.
[23]	AÇICI K, ERDASÇ B, ASUROGLU T，et al. A random forest method to detect Parkinson's disease via gait analysis[EB/OL]. [2022-04-03].
[24]	PROBST P, WRIGHT M N, BOULESTEIX A L. Hyperparameters and tuning strategies for random forest[J]. Wiley Interdisciplinary Reviews，2019，9（3）：e1301. DOI：10.1002/widm.1301.
[25]	RONG G, ALU S, LI K，et al. Rainfall induced landslide susceptibility mapping based on bayesian optimized random forest and gradient boosting decision tree models：a case study of Shuicheng County，China[J]. Water，2020，12（11）：3066.
[26]	FRIEDMAN J H. Greedy function approximation：a gradient boosting machine[J]. Annals of Statistics，2001，29（5）：1189-1232. DOI：10.1214/aos/1013203451.
[27]	RAO H, SHI X, RODRIGUE A K，et al. Feature selection based on artificial bee colony and gradient boosting decision tree[J]. Applied Soft Computing，2019，74：634-642.
[28]	YAO M, ZHU Y, LI J，et al. Research on predicting line loss rate in low voltage distribution network based on gradient boosting decision tree[J]. Energies，2019，12（13）：2522. DOI：10.3390/en12132522.
[29]	PULLANAGARI R R, KERESZTURI G, YULE I. Integrating airborne hyperspectral，topographic，and soil data for estimating pasture quality using recursive feature elimination with random forest regression[J]. Remote Sensing，2018，10（7）：1117.
[30]	LINDH-RENGIFO M, JONASSON S B, ULLÉN S，et al. Components of gait in people with and without mild cognitive impairment[J]. Gait & Posture，2022，93：83-89.
[31]	KOO D K, JANG T S, KWON J W. Effects of dual-task training on gait parameters in elderly patients with mild dementia[J]. Healthcare（Basel），2021，9（11）：1444. DOI：10.3390/healthcare9111444.
[32]	SCHLACHETZKI J C M, BARTH J, MARXREITER F，et al. Wearable sensors objectively measure gait parameters in Parkinson's disease[J]. PLoS One，2017，12（10）：e0183989. DOI：10.1371/journal.pone.0183989.
[33]	吴壮，仲敏，蒋旭，等. 不同临床分期帕金森病患者的步态分析[J]. 临床神经病学杂志，2021，34（3）：169-172.
[34]	NI L, LV W, SUN D，et al. Pathological gait signatures of post-stroke dementia with toe-off and heel-to-ground angles discriminate from Alzheimer's disease[J]. Frontiers in Aging Neuroscience，2021，13：766884. DOI：10.3389/fnagi.2021.766884.
[35]	PIERUCCINI-FARIA F, SARQUIS-ADAMSON Y, MONTERO-ODASSO M. Mild cognitive impairment affects obstacle negotiation in older adults：results from "Gait and Brain Study"[J]. Gerontology，2019，65（2）：164-173. DOI：10.1159/000492931.
[36]	GHORAANI B, BOETTCHER L N, HSSAYENI M D，et al. Detection of mild cognitive impairment and Alzheimer's disease using dual-task gait assessments and machine learning[J]. Biomedical Signal Processing and Control，2021，64：102249. DOI：10.1016/j.bspc.2020.102249.
[37]	GIMMON Y, RASHAD H, KURZ I，et al. Gait coordination deteriorates in independent old-old adults[J]. Journal of Aging and Physical Activity，2018，26（3）：382-389.
[38]	KANG G E, YANG J, NAJAFI B. Does the presence of cognitive impairment exacerbate the risk of falls in people with peripheral neuropathy? An application of body-worn inertial sensors to measure gait variability[J]. Sensors，2020，20（5）：1328.
[39]	XIE H, WANG Y, TAO S，et al. Wearable sensor-based daily life walking assessment of gait for distinguishing individuals with amnestic mild cognitive impairment[J]. Frontiers in Aging Neuroscience，2019，11：285. DOI：10.3389/fnagi.2019.00285.
[40]	VERGHESE J, ROBBINS M, HOLTZER R，et al. Gait dysfunction in mild cognitive impairment syndromes[J]. Journal of the American Geriatrics Society，2008，56（7）：1244-1251.DOI：10.1111/j.1532-5415.2008.01758.x.
[41]	TSENG B Y, CULLUM C M, ZHANG R. Older adults with amnestic mild cognitive impairment exhibit exacerbated gait slowing under dual-task challenges[J]. Current Alzheimer Research，2014，11（5）：494-500. DOI：10.2174/1567205011666140505110828.
[42]	FOLEY J A, KASCHEL R, LOGIE R H，et al. Dual-task performance in Alzheimer's disease，mild cognitive impairment，and normal ageing[J]. Archives of Clinical Neuropsychology，2011，26（4）：340-348. DOI：10.1093/arclin/acr032.
[43]	ANSAI J H, ANDRADE L P, ROSSI P G，et al. Gait，dual task and history of falls in elderly with preserved cognition，mild cognitive impairment，and mild Alzheimer's disease[J]. Brazilian Journal of Physical Therapy，2017，21（2）：144-151. DOI：10.1016/j.bjpt.2017.03.010.
[44]	OH C. Single-task or dual-task? Gait assessment as a potential diagnostic tool for Alzheimer's dementia[J]. Journal of Alzheimer's Disease，2021，84（3）：1183-1192.
[45]	CASTRILLO A, OLMOS L M G, RODRÍGUEZ F，et al. Gait disorder in a cohort of patients with mild and moderate Alzheimer's disease[J]. American Journal of Alzheimer's Disease & Other Dementias®，2016，31（3）：257-262. DOI：10.1177/1533317515603113.
[46]	NADKARNI N K, MAWJI E, MCILROY W E，et al. Spatial and temporal gait parameters in Alzheimer's disease and aging[J]. Gait & Posture，2009，30（4）：452-454.
[47]	SEIFALLAHI M, SOLTANIZADEH H, HASSANI M A，et al. Detection of Alzheimer's disease in elder people using gait analysis and kinect camera[J]. Journal of Health and Biomedical Informatics，2019，6（3）：178-196.
[48]	MUIR S W, SPEECHLEY M, WELLS J，et al. Gait assessment in mild cognitive impairment and Alzheimer's disease：the effect of dual-task challenges across the cognitive spectrum[J]. Gait & Posture，2012，35（1）：96-100. DOI：10.1016/j.gaitpost.2011.08.014.

组别	例数	年龄〔M（QR），岁〕	性别（男/女）	身高（±s，cm）	体质量（±s，kg）	鞋码〔M（QR），码〕	MMSE量表评分〔M（QR），分〕	MoCA量表评分〔M（QR），分〕
HC组	33	67.0（5.5）	14/19	160.24±6.82	58.65±9.46	38.00（3.00）	27（2）	24.0（3.5）
aMCI组	55	68.0（5.0）	27/28	161.40±6.96	60.92±8.92	39.00（3.00）	25（3）	20.0（4.0）
AD组	10	66.0（11.0）	3/7	159.35±8.21	60.00±9.67	38.00（4.25）	21（3）	14.0（5.2）
检验统计量值		1.390^a	1.370^b	0.515^c	0.629^c	2.743^a	33.247^a	48.238^a
P值		0.499	0.504	0.599	0.535	0.254	<0.001	<0.001

组别	例数	年龄〔M（QR），岁〕	性别（男/女）	身高（±s，cm）	体质量（±s，kg）	鞋码〔M（QR），码〕	MMSE量表评分〔M（QR），分〕	MoCA量表评分〔M（QR），分〕
HC组	33	67.0（5.5）	14/19	160.24±6.82	58.65±9.46	38.00（3.00）	27（2）	24.0（3.5）
aMCI组	55	68.0（5.0）	27/28	161.40±6.96	60.92±8.92	39.00（3.00）	25（3）	20.0（4.0）
AD组	10	66.0（11.0）	3/7	159.35±8.21	60.00±9.67	38.00（4.25）	21（3）	14.0（5.2）
检验统计量值		1.390^a	1.370^b	0.515^c	0.629^c	2.743^a	33.247^a	48.238^a
P值		0.499	0.504	0.599	0.535	0.254	<0.001	<0.001

组别	例数	步幅（±s，m）	步速（±s，m/s）	步频（±s，步/min）	支撑相〔M（QR），%〕	摆动相〔M（QR），%〕	跨步时间（±s，s）	支撑时间（±s，s）	摆动时间（±s，s）	足趾离地角度（±s，°）	足跟着地角度〔M（QR），°〕
HC组	33	1.16±0.13	0.92±0.14	96.19±10.40	65.70（3.80）	34.30（3.80）	1.26±0.16	0.83±0.13	0.43±0.04	42.62±3.96	31.70（6.70）
aMCI组	55	1.08±0.16^b	0.87±0.14	96.10±10.18	66.30（2.60）	33.70（2.60）	1.27±0.15	0.84±0.12	0.44±0.04	41.05±4.48	28.90（5.90）^b
AD组	10	1.00±0.17^b	0.75±0.17^bc	89.69±9.35	67.55（5.95）	32.45（5.95）	1.37±0.17	0.93±0.16	0.44±0.03	38.34±5.20^b	28.05（8.47）^b
F（H）值		5.265	4.927	1.800	2.633^a	2.633^a	2.053	2.719	0.112	3.872	12.323^a
P值		0.007	0.009	0.171	0.236	0.236	0.134	0.071	0.894	0.024	0.002

组别	例数	步幅（±s，m）	步速（±s，m/s）	步频（±s，步/min）	支撑相〔M（QR），%〕	摆动相〔M（QR），%〕	跨步时间（±s，s）	支撑时间（±s，s）	摆动时间（±s，s）	足趾离地角度（±s，°）	足跟着地角度〔M（QR），°〕
HC组	33	1.16±0.13	0.92±0.14	96.19±10.40	65.70（3.80）	34.30（3.80）	1.26±0.16	0.83±0.13	0.43±0.04	42.62±3.96	31.70（6.70）
aMCI组	55	1.08±0.16^b	0.87±0.14	96.10±10.18	66.30（2.60）	33.70（2.60）	1.27±0.15	0.84±0.12	0.44±0.04	41.05±4.48	28.90（5.90）^b
AD组	10	1.00±0.17^b	0.75±0.17^bc	89.69±9.35	67.55（5.95）	32.45（5.95）	1.37±0.17	0.93±0.16	0.44±0.03	38.34±5.20^b	28.05（8.47）^b
F（H）值		5.265	4.927	1.800	2.633^a	2.633^a	2.053	2.719	0.112	3.872	12.323^a
P值		0.007	0.009	0.171	0.236	0.236	0.134	0.071	0.894	0.024	0.002

组别	例数	步幅（±s，m）	步速〔M（QR），m/s 〕	步频（±s，步/min）	支撑相（±s，%）	摆动相（±s，%）	跨步时间（±s，s）	支撑时间（±s，s）	摆动时间（±s，s）	足趾离地角度（±s，°）	足跟着地角度〔M（QR），°〕
HC组	33	1.07±0.14	0.77（0.14）	87.58±8.39	67.11±1.66	32.89±1.66	1.39±0.15	0.94±0.12	0.46±0.04	40.13±3.93	29.40（4.90）
aMCI组	55	1.01±0.16	0.72（0.21）^b	85.23±15.34	68.11±2.62	31.89±2.62	1.48±0.31	1.01±0.25	0.47±0.07	36.92±4.53^b	26.10（7.10）^b
AD组	10	0.96±0.23	0.54（0.25）^b	77.75±16.43	68.94±4.00	31.06±4.00	1.61±0.32	1.11±0.22^b	0.50±0.13	33.39±6.10^bc	24.40（10.65）^b
F（H）值		2.167	10.069^a	2.026	2.654	2.654	2.788	3.126	1.426	10.158	9.024^a
P值		0.120	0.007	0.138	0.076	0.076	0.067	0.048	0.245	<0.001	0.011

Machine Learning-based Gait Analysis for Recognition of Amnestic Mild Cognitive Impairment and Alzheimer's Disease

基于步态的机器学习模型识别遗忘型轻度认知障碍和阿尔茨海默病

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 8

References 48

Related Articles 15

Recommended Articles

Metrics

Comments

组别	例数	步幅（±s，m）	步速（±s，m/s）	步频（±s，步/min）	支撑相〔M（QR），%〕	摆动相〔M（QR），%〕	跨步时间（±s，s）	支撑时间（±s，s）	摆动时间（±s，s）	足趾离地角度（±s，°）	足跟着地角度〔M（QR），°〕
HC组	33	1.09±0.16	0.77±0.11	84.97±12.39	67.90（3.95）	32.10（3.95）	1.46±0.25	1.00±0.20	0.47±0.06	39.94±4.26	29.30（4.15）
aMCI组	55	1.03±0.16	0.72±0.14	83.31±14.40	68.00（3.40）	32.00（3.40）	1.50±0.28	1.02±0.22	0.47±0.07	38.07±4.27	27.10（5.40）^b
AD组	10	0.96±0.19^b	0.62±0.14^bc	79.54±12.04	70.80（5.15）	29.20（5.15）	1.59±0.26	1.12±0.22	0.48±0.05	34.27±5.57^bc	23.05（7.15）^bc
F（H）值		3.230	4.759	0.629	4.080^a	4.080^a	0.861	1.165	0.160	6.531	13.584^a
P值		0.044	0.011	0.535	0.130	0.130	0.426	0.316	0.852	0.002	0.001

项目		RF		GBDT
项目		准确度	AUC	准确度	AUC
自由行走
	HC组/aMCI组	0.623 5	0.608 3	0.636 6	0.576 5
	HC组/AD组	0.811 1	0.888 9	0.816 7	0.888 9
	aMCI组/AD组	0.861 5	0.833 3	0.815 4	0.785 7
倍数7测试
	HC组/aMCI组	0.647 7	0.628 6	0.637 3	0.606 8
	HC组/AD组	0.838 9	0.916 7	0.836 1	0.875 0
	aMCI组/AD组	0.830 8	0.812 5	0.769 2	0.809 5
倒数100测试
	HC组/aMCI组	0.683 7	0.695 8	0.717 0	0.742 9
	HC组/AD组	0.838 9	0.900 0	0.741 7	0.777 8
	aMCI组/AD组	0.800 0	0.850 0	0.753 8	0.800 0

项目		RF		GBDT
项目		准确率	AUC	准确率	AUC
自由行走
	HC组/AD组	0.858 3	0.900 0	0.811 1	0.791 7
	aMCI组/AD组	0.876 9	0.933 3	0.846 1	0.812 5
倍数7测试
	HC组/AD组	0.905 6	0.944 4	0.905 6	0.944 4
	aMCI组/AD组	0.876 9	0.817 3	0.830 8	0.866 7
倒数100测试
	HC组/AD组	0.858 3	0.875 0	0.819 4	0.833 3
	aMCI组/AD组	0.846 2	0.866 7	0.769 2	0.781 3

[1]	LIU Meixia, YIN Jinnian, WU Mei, YANG Xing, ZHOU Quanxiang, YANG Jingyuan. Impact of Body Mass Index on the Association of Triglyceride Glucose Index with Cognitive Function: a Cross-sectional Study in Rural Older Adults in Guizhou Province [J]. Chinese General Practice, 2025, 28(22): 2806-2812.
[2]	TAN Wenbin, LI Jia, LIU Mingyu, LU Yongxin, CHENG Yaxin. Research Progress on the Influence of Nervous System Diseases and Related Therapeutic Drugs on Osteoporosis [J]. Chinese General Practice, 2025, 28(17): 2092-2100.
[3]	LIU Feiran, CHEN Minghuang, ZHAO Shuaihong, BAI Wenpei. Application of Random Forest Algorithm in Pregnancy Prediction after Fallopian Tube Recanalization [J]. Chinese General Practice, 2025, 28(11): 1361-1366.
[4]	WEN Yining, HUANG Huichang, ZHAO Mingming. Advances in the Study of Sleep-related Subjective Cognitive Decline [J]. Chinese General Practice, 2025, 28(11): 1403-1410.
[5]	HAN Shukui, REN Yitao, MA Xin, SONG Panpan, MA Jinxiang, ZHANG Ziyu, CHEN Hongru. Analysis of Disease Burden Trends and Forecast of Alzheimer's Disease and Other Dementias among the Elderly in China from 1992 to 2021 [J]. Chinese General Practice, 2025, 28(08): 996-1003.
[6]	PEI Jinying, SONG Jinzhou, MA Daofeng, LUO Xiaoting, LIU Bin, DONG Xiaohong, CONG Shuyuan. Research Progress in the Study of Traditional Chinese Herbal Medicine Extracts Regulating Ferroptosis thus Improving the Symptom of Alzheimer's Disease [J]. Chinese General Practice, 2025, 28(05): 639-648.
[7]	TIAN Yinghan, LIU Lewei, YANG Cheng, LING Chen, YANG Xiaoxue, FAN Haojie, ZHAO Xin, LI Jun, XIA Lei, LIU Huanzhong. Associations and Sex Differences between Depression and Cognitive Function in the Urban Elderly [J]. Chinese General Practice, 2025, 28(01): 47-52.
[8]	WANG Chenyu, ZHANG Chen, SHI Lulu, ZHU Mengna, WANG Lina. Advances in Cognitive Function Health Management Strategies for Individuals with Mild Cognitive Impairment in Community [J]. Chinese General Practice, 2024, 27(26): 3273-3280.
[9]	MA Jia, ZHANG Minjue, ZHANG Shaowei, YU Haiyan, CHEN Shen, Gulibaier MAMUTI, HONG Juan, LU Yuan. Community Management of Elderly Patients with Type 2 Diabetes Complicated with Mild Cognitive Impairment Related Factors [J]. Chinese General Practice, 2024, 27(16): 1984-1989.
[10]	ZHANG Min, LU Yuan, GAO Song, YU Dehua. Study of the Perceptions of Mild Cognitive Impairement among Community Populations Questionnaire Developed Based on Delphi Method [J]. Chinese General Practice, 2024, 27(16): 1990-1997.
[11]	ZHAI Jiayi, LU Yuan, TAO Qiongying, HUANG Wuquan, LIU Yalin, YU Dehua. A Study on the Factors Influencing Performance of Chronic Disease Management among Population with Mild Cognitive Impairment in the Context of Family Doctor Service Based on Grounded Theory [J]. Chinese General Practice, 2024, 27(13): 1608-1615.
[12]	QIN Xiaoyu, ZHANG Binsen, ZHANG Xiaojia, LU Xiaoting, LIU Hongxin, WANG Chunai. Effect of Electroacupuncture on Inflammation Response and Ferroptosis in Rats Modeling Postoperative Cognitive Dysfunction [J]. Chinese General Practice, 2024, 27(06): 723-732.
[13]	HUANG Xiaotong, WANG Qiong, LIU Chen, HOU Xiaochun, XU Xinrui, WU Bingyi, YANG Xiao. A Path Analysis of the Effect of Childhood Socioeconomic Status on Cognitive Function in Middle-aged and Older Adults [J]. Chinese General Practice, 2024, 27(03): 322-327.
[14]	LI Yaping, LI Xia. Research Progress of Correlation and Underlying Mechanisms between Metabolic Syndrome and Alzheimer's Disease [J]. Chinese General Practice, 2024, 27(03): 268-272.
[15]	WANG Ying, DONG Zhixiao, YANG Kehu. Non-pharmacological Integrated Interventions in Older Adults with Mild Cognitive Impairment: a Single Case Pilot Study [J]. Chinese General Practice, 2024, 27(03): 315-321.