基于语音停顿度和平坦度的注意缺陷与多动障碍自动检测算法

doi:10.11772/j.issn.1001-9081.2021071213

《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (9): 2917-2925.DOI: 10.11772/j.issn.1001-9081.2021071213

• 多媒体计算与计算机仿真 • 上一篇

基于语音停顿度和平坦度的注意缺陷与多动障碍自动检测算法

李国中¹, 崔娅¹, 俄木依欣¹, 何凌¹, 李元媛²(), 熊熙³

^1.四川大学电气工程学院，成都 610065
^2.四川大学华西医院心理卫生中心，成都 610065
^3.成都信息工程大学网络空间安全学院，成都 610225

收稿日期:2021-07-13 修回日期:2021-11-14 接受日期:2021-11-17 发布日期:2022-09-19 出版日期:2022-09-10
通讯作者: 李元媛
作者简介:李国中（1994—），男，河南商丘人，硕士研究生，主要研究方向：自然语言分析、机器学习、语音信号处理、医学信号处理；
崔娅（2001—），女，重庆人，主要研究方向：语音特征处理、医学信号处理；
俄木依欣（2000—），女，四川乐山人，主要研究方向：医学信号处理、医学图像处理；
何凌（1981—），女，四川成都人，副教授，博士，主要研究方向：医学信号处理、医学图像处理；
李元媛（1984—），女，辽宁大连人，副教授，博士，主要研究方向：临床数据挖掘、医学信息抽取； guojipangxie@126.com
熊熙（1983—），男，四川成都人，副教授，博士，CCF会员，主要研究方向：信息抽取、自然语言分析、社会计算。
基金资助:
国家自然科学基金资助项目(81901389);四川省科技计划项目(2019YFS0236)

Automatic detection algorithm for attention deficit/hyperactivity disorder based on speech pause and flatness

Guozhong LI¹, Ya CUI¹, Yixin EMU¹, Ling HE¹, Yuanyuan LI²(), Xi XIONG³

^1.College of Electrical Engineering，Sichuan University，Chengdu Sichuan 610065，China
^2.Mental Health Center，West China Hospital of Sichuan University，Chengdu Sichuan 610065，China
^3.School of Cyberspace Security，Chengdu University of Information Technology，Chengdu Sichuan 610025，China

Received:2021-07-13 Revised:2021-11-14 Accepted:2021-11-17 Online:2022-09-19 Published:2022-09-10
Contact: Yuanyuan LI
About author:LI Guozhong， born in 1994， M. S. candidate. His research interests include natural language analysis， machine learning， speech signal processing， medical signal processing.
CUI Ya， born in 2001. Her research interests include speech feature processing， medical signal processing.
EMU Yixin， born in 2000. Her research interests include medical signal processing， medical image processing.
HE Ling， born in 1981， Ph. D.， associate professor. Her research interests include medical signal processing， medical image processing.
XIONG Xi， born in 1983， Ph. D.， associate professor. His research interests include information extraction， natural language analysis， social computing.
Supported by:
National Natural Science Foundation of China(81901389);Science and Technology Program of Sichuan Province(2019YFS0236)

摘要/Abstract

摘要：

针对注意缺陷与多动障碍（ADHD）临床诊断主要依靠医生主观评估，缺乏客观辅助依据的问题，提出了一种基于语音停顿度和平坦度的ADHD自动检测算法。首先，通过频带差能熵积（FDEEP）参数自动定位语音有话区间，并提取停顿度特征；然后，使用变换平均幅度平方差（TAASD）参数计算语音倍频率，并提取平坦度特征；最后，结合融合特征和支持向量机（SVM）分类器来实现ADHD的自动识别。实验共采集了17位正常对照组儿童和37位ADHD患儿的语音样本。实验结果表明，所提算法能自动检测正常儿童和ADHD患儿，识别正确率为91.38%。

关键词: 注意缺陷与多动障碍, 频带差能熵积, 停顿度, 变换平均幅度平方差, 平坦度

Abstract:

The clinicians diagnose Attention Deficit/Hyperactivity Disorder （ADHD） mainly based on on their subjective assessment， which lacks objective criteria to assist. To solve this problem， an automatic detection algorithm for ADHD based on speech pause and flatness was proposed. Firstly， the Frequency band Difference Energy Entropy Product （FDEEP） parameter was used to automatically locate the segment with voice from the speech and extract the speech pause features. Then， Transform Average Amplitude Squared Difference （TAASD） parameter was presented to calculate the voice multi-frequency and extract the flatness features. Finally， fusion features and the Support Vector Machine （SVM） classifier were combined to realize the automatic recognition of ADHD. The speech samples of the experiment were collected from 17 normal control children and 37 children with ADHD. Experimental results show that the proposed algorithm can effectively discriminate the normal children and children with ADHD， with an accuracy of 91.38%.

Key words: Attention Deficit/Hyperactivity Disorder (ADHD), Frequency band Difference Energy Entropy Product (FDEEP), pause, Transform Average Amplitude Squared Difference (TAASD), flatness

中图分类号:

TP391.1

李国中, 崔娅, 俄木依欣, 何凌, 李元媛, 熊熙. 基于语音停顿度和平坦度的注意缺陷与多动障碍自动检测算法[J]. 计算机应用, 2022, 42(9): 2917-2925.

Guozhong LI, Ya CUI, Yixin EMU, Ling HE, Yuanyuan LI, Xi XIONG. Automatic detection algorithm for attention deficit/hyperactivity disorder based on speech pause and flatness[J]. Journal of Computer Applications, 2022, 42(9): 2917-2925.

图/表 10

图1 基于语音停顿度和倍频率平坦度的注意缺陷与多动障碍自动识别算法流程

Fig. 1 Flowchart of automatic recognition algorithm for attention deficit/hyperactivity disorder based on speech pause and multi-frequency flatness

图2 基于语音FDEEP参数的停顿度声学特征提取算法流程

Fig. 2 Flowchart of acoustic pause feature extraction algorithm based on speech FDEEP parameter

图3 基于语音TAASD参数的平坦度声学特征提取算法流程

Fig. 3 Flowchart of acoustic flatness feature extraction algorithm based on speech TAASD parameters

图4 ADHD患儿与正常对照组的语音信号倍频率

Fig. 4 Speech signal multi-frequency of ADHD patient and normal control

表1 ADHD患者与正常对照组的语音样本子数据集

Tab. 1 Speech sample sub datasets of ADHD patients and normal controls

数据集序号	儿童类别及个数	语音样本数
1	NM17和NADHD16	99
2	NM17和RADHD21	114
3	NADHD16和RADHD21	111
4	NM17和ADHD37	162

表2 本文算法的ADHD自动检测结果 ( %)

Tab. 2 ADHD automatic detection results of the proposed algorithm

数据集序号	停顿度特征+平坦度特征			停顿度特征			平坦度特征
数据集序号	正确率	特异性	灵敏度	正确率	特异性	灵敏度	正确率	特异性	灵敏度
1	91.38	93.84	89.38	92.76	90.06	95.68	90.69	88.73	92.57
2	74.70	78.38	80.65	56.36	44.13	67.03	81.82	83.13	80.59
3	90.90	89.47	92.86	84.68	81.34	87.10	90.63	92.85	88.89
4	80.41	84.16	72.78	66.67	71.89	85.53	80.21	66.23	86.81

图5 语音停顿度特征箱线图

Fig. 5 Box plots of speech pause features

图6 倍频率平坦度特征箱线图

Fig. 6 Box plots of multi- frequency flatness features

表3 显著性分析

Tab. 3 Significance analysis

本文提取的语音特征	显著性水平（p）
最长停顿时长	1.392 9E-19
平均停顿时长	2.944 3E-16
停顿时间占比	2.657 5E-12
倍频率方差	1.028 2E-15
倍频率离散系数	6.284 7E-11
倍频率峰度	6.010 0E-23

表4 本文提取特征与传统病理语音特征自动检测ADHD的识别结果 (%)

Tab. 4 Recognition results of ADHD automatic detection of features extracted in this paper and traditional pathological voice features

语音特征类别	正确率	特异性	灵敏度
短时能量^［44-45］	81.11	82.82	74.78
MFCC^［41-43］	72.76	71.33	78.57
共振峰^［35-36］	62.50	69.23	54.55
基频^［47-49］	82.76	85.19	66.67
本文提出的停顿度和平坦度特征	91.38	93.84	89.38

参考文献 51

1	KUBO Y， ANAZAWA T， KAWABATA Y， et al. Comparative analysis of the WISC between two ADHD subgroups［J］. Psychiatry Investigation， 2018， 15（2）：172-177. 10.30773/pi.2017.07.12
2	WANG T T， LIU K H， LI Z Z， et al. Prevalence of attention deficit/hyperactivity disorder among children and adolescents in China： a systematic review and meta-analysis［J］. BMC Psychiatry， 2017， 17： No.32. 10.1186/s12888-016-1187-9
3	TSAI C J， LIN H Y， TSENG I W Y， et al. Brain voxel-based morphometry correlates of emotion dysregulation in attention-deficit hyperactivity disorder［J］. Brain Imaging and Behavior， 2021， 15（3）：1388-1402. 10.1007/s11682-020-00338-y
4	GAIR S L， BROWN H R， KANG S， et al. Early development of comorbidity between symptoms of ADHD and anxiety［J］. Research on Child and Adolescent Psychopathology， 2021， 49（3）：311-323. 10.1007/s10802-020-00724-6
5	VALERO S， BOSCH R， COROMINAS M， et al. Remittance or persistence of Attention Deficit-Hyperactivity Disorder （ADHD） and its impact on recidivism in risky driving behaviors［J］. Traffic Injury Prevention， 2018， 19（7）：701-707. 10.1080/15389588.2018.1505043
6	DEW R E， KOLLINS S H， KOENIG H G. ADHD， religiosity， and psychiatric comorbidity in adolescence and adulthood［J］. Journal of Attention Disorders， 2022， 26（2）：307-318. 10.1177/1087054720972803
7	李园园，张红霞，马丽华，等. 美国《精神障碍诊断与统计手册》第5版跨界症状量表在消化系统恶性肿瘤患者精神症状评估中的应用［J］. 临床精神医学杂志， 2020， 30（4）：260-262.
	LI Y Y， ZHANG H X， MA L H， et al. Application of Diagnostic and Statistical Manual of Mental Disorders（DSM）-5 cross-cutting symptom measures in assessing the mental symptoms of patients with gastrointestinal malignant tumor［J］. Journal of Clinical Psychiatry， 2020， 30（4）：260-262.
8	朱琳，李斐，陈立. 4种常见评定量表在儿童注意缺陷多动障碍诊断与随访管理中的应用［J］. 重庆医科大学学报， 2020， 45（1）：32-35.
	ZHU L， LI F， CHEN L. Application of four commonly used rating scales in diagnosis and follow-up management of children with attention deficit hyperactivity disorder［J］. Journal of Chongqing Medical University， 2020， 45（1）：32-35.
9	KHALEGHI A， BIRGANI P M， FOOLADI M F， et al. Applicable features of electroencephalogram for ADHD diagnosis［J］. Research on Biomedical Engineering， 2020， 36（1）：1-11. 10.1007/s42600-019-00036-9
10	TANG Y B， LI X F， CHEN Y， et al. High-accuracy classification of attention deficit hyperactivity disorder with l_2，1-norm linear discriminant analysis and binary hypothesis testing［J］. IEEE Access， 2020， 8：562828-56237. 10.1109/access.2020.2982401
11	MUNOZ-ORGANERO M， POWELL L， HELLER B， et al. Using recurrent neural networks to compare movement patterns in ADHD and normally developing children based on acceleration signals from the wrist and ankle［J］. Sensors， 2019， 19（13）： No.2935. 10.3390/s19132935
12	NAKATANI M， OKADA S， SHIMIZU S， et al. Body movement analysis during sleep for children with ADHD using video image processing［C］// Proceedings of the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway： IEEE， 2013： 6389-6392. 10.1109/embc.2013.6611016
13	LEVANTINI V， MURATORI P， INGUAGGIATO E， et al. Eyes are the window to the mind： eye-tracking technology as a novel approach to study clinical characteristics of ADHD［J］. Psychiatry Research， 2020， 290： No.113135. 10.1016/j.psychres.2020.113135
14	JAYAWARDENA G， JAYARATHNA S. Automated filtering of eye gaze metrics from dynamic areas of interest［C］// Proceedings of the IEEE 21st International Conference on Information Reuse and Integration for Data Science. Piscataway： IEEE， 2020：67-74. 10.1109/iri49571.2020.00018
15	MI L， WANG L， LI X Z， et al. Reduction of phonetic mismatch negativity may depict illness course and predict functional outcomes in schizophrenia［J］. Journal of Psychiatric Research， 2021，137：290-297. 10.1016/j.jpsychires.2021.02.065
16	张静，崔东红. 语音信号在精神疾病领域的应用［J］. 临床精神医学杂志， 2016， 26（3）：203-205.
	ZHANG J， CUI D H. The application of speech signal in the field of mental illness［J］. Journal of Clinical Psychiatry， 2016， 26（3）：203-205.
17	COELHO R M， DRUMMOND C， MOTA N B， et al. Network analysis of narrative discourse and attention-deficit hyperactivity symptoms in adults［J］. PLoS ONE， 2021， 16（4）： No.e0245113. 10.1371/journal.pone.0245113
18	KONRAD K， EICKHOFF S B. Is the ADHD brain wired differently？ a review on structural and functional connectivity in attention deficit hyperactivity disorder［J］. Human Brain Mapping， 2010， 31（6）：904-916. 10.1002/hbm.21058
19	ALTINKAYNAK M， GÜVEN A， DOLU N， et al. Prefrontal brain activation in subtypes of attention deficit hyperactivity disorder： a functional near-infrared spectroscopy study［J］. Electrica， 2018， 18（2）：256-262. 10.26650/electrica.2018.99730
20	HANKEY J R， LIU Y S， CHOKKA P R. Adult ADHD： prevalence， comorbidities， and patient-reported dysfunction in a tertiary mental health clinic［J］. Psychiatry and Clinical Psychopharmacology， 2020， 30（1）：3-9.
21	PEHLIVANIDIS A， PAPANIKOLAOU K， KOROBILI K， et al. Trait-based dimensions discriminating adults with Attention Deficit Hyperactivity Disorder （ADHD）， Autism Spectrum Disorder （ASD） and， co-occurring ADHD/ASD［J］. Brain Sciences， 2021， 11（1）： No.18. 10.3390/brainsci11010018
22	FIGUEIREDO T， FORTES D， ERTHAL P， et al. Impulsivity as an endophenotype in ADHD： negative findings［J］. Journal of Attention Disorders， 2021， 25（4）：502-507. 10.1177/1087054718816161
23	ROSELLÓ B， BERENGUER C， BAIXAULI I， et al. Empirical examination of executive functioning， ADHD associated behaviors， and functional impairments in adults with persistent ADHD， remittent ADHD， and without ADHD［J］. BMC Psychiatry， 2020， 20： No.134. 10.1186/s12888-020-02542-y
24	ETTER N M， CADELY F A， PETERS M G， et al. Speech motor control and orofacial point pressure sensation in adults with ADHD［J］. Neuroscience Letters， 2020， 744： No.135592. 10.1016/j.neulet.2020.135592
25	PETROLINI V， JORBA M， VICENTE A. The role of inner speech in executive functioning tasks： schizophrenia with auditory verbal hallucinations and autistic spectrum conditions as case studies［J］. Frontiers in Psychology， 2020， 11： No.572035. 10.3389/fpsyg.2020.572035
26	BECKER S P， BURNS G L， SMITH Z R， et al. Sluggish cognitive tempo in adolescents with and without ADHD： differentiation from adolescent-reported ADHD inattention and unique associations with internalizing domains［J］. Journal of Abnormal Child Psychology， 2020， 48（3）：391-406. 10.1007/s10802-019-00603-9
27	ALMEIDA J S， REBOUÇAS FLIHO P P， CARNEIRO T， et al. Detecting Parkinson’s disease with sustained phonation and speech signals using machine learning techniques［J］. Pattern Recognition Letters， 2019， 125： 55-62. 10.1016/j.patrec.2019.04.005
28	POLAT K， NOUR M. Parkinson disease classification using one against all based data sampling with the acoustic features from the speech signals［J］. Medical Hypotheses， 2020， 3（140）： No.109678.
29	NARENDRA N P， ALKU P. Dysarthric speech classification from coded telephone speech using glottal features［J］. Speech Communication， 2019， 110： 47-55. 10.1016/j.specom.2019.04.003
30	VINCZE V， SZATLÓCZKI G， TÓTH L， et al. Telltale silence： temporal speech parameters discriminate between prodromal dementia and mild Alzheimer’s disease［J］. Clinical Linguistics and Phonetics， 2021， 35（8）：727-742. 10.1080/02699206.2020.1827043
31	NASROLAHZADEH M， MOHAMMADPOORY Z， HADDADNIA J. Higher-order spectral analysis of spontaneous speech signals in Alzheimer’s disease［J］. Cognitive Neurodynamics， 2018， 12（6）：583-596. 10.1007/s11571-018-9499-8
32	MARTINC M， HAIDER F， POLLAK S， et al. Temporal integration of text transcripts and acoustic features for Alzheimer’s diagnosis based on spontaneous speech［J］. Frontiers in Aging Neuroscience， 2021， 13： No.642647. 10.3389/fnagi.2021.642647
33	GOSZTOLYA G， VINCZE V， TÓTH L， et al. Identifying mild cognitive impairment and mild Alzheimer’s disease based on spontaneous speech using ASR and linguistic features［J］. Computer Speech and Language， 2019， 53： 181-197. 10.1016/j.csl.2018.07.007
34	PETTI U， BAKER S， KORHONEN A. A systematic literature review of automatic Alzheimer’s disease detection from speech and language［J］. Journal of the American Medical Informatics Association， 2020， 27（11）：1784-1797. 10.1093/jamia/ocaa174
35	BARRAGÁN-PULIDO M L， HERNÁNDEZ J B A， FERRER-BALLESTER M A， et al. Alzheimer’s disease and automatic speech analysis： a review［J］. Expert Systems with Applications， 2020， 150： No.113213. 10.1016/j.eswa.2020.113213
36	YAMAMOTO M， TAKAMIYA A， SAWADA K， et al. Using speech recognition technology to investigate the association between timing-related speech features and depression severity［J］. PLoS ONE， 2020， 15（9）： No.e0238726. 10.1371/journal.pone.0238726
37	TAGUCHI T， TACHIKAWA H， NEMOTO K， et al. Major depressive disorder discrimination using vocal acoustic features［J］. Journal of Affective Disorders， 2018， 225：214-220. 10.1016/j.jad.2017.08.038
38	DONG Y Z， YANG X Y. A hierarchical depression detection model based on vocal and emotional cues［J］. Neurocomputing， 2021， 441：279-290. 10.1016/j.neucom.2021.02.019
39	JAGGER-RICKELS A C， KIBBY M Y， CONSTANCE J M. Global gray matter morphometry differences between children with reading disability， ADHD， and comorbid reading disability/ADHD［J］. Brain and Language， 2018， 185：54-66. 10.1016/j.bandl.2018.08.004
40	ARICÒ M， ARIGLIANI E， GIANNOTTI F， et al. ADHD and ADHD-related neural networks in benign epilepsy with centrotemporal spikes： a systematic review［J］. Epilepsy and Behavior， 2020， 112： No.107448. 10.1016/j.yebeh.2020.107448
41	SRIMADHUR N S， LALITHA S. An end-to-end model for detection and assessment of depression levels using speech［J］. Procedia Computer Science， 2020， 171： 12-21. 10.1016/j.procs.2020.04.003
42	HE L， CAO C. Automated depression analysis using convolutional neural networks from speech［J］. Journal of Biomedical Informatics， 2018， 83： 103-111. 10.1016/j.jbi.2018.05.007
43	CHLASTA K， WOŁK K， KREJTZ I. Automated speech-based screening of depression using deep convolutional neural networks［J］. Procedia Computer Science， 2019， 164： 618-628. 10.1016/j.procs.2019.12.228
44	PAROLA A， SIMONSEN A， BLIKSTED V， et al. Voice patterns in schizophrenia： a systematic review and Bayesian meta-analysis［J］. Schizophrenia Research， 2020， 216：24-40. 10.1016/j.schres.2019.11.031
45	RAPCAN V， D’ARCY S， YEAP S， et al. Acoustic and temporal analysis of speech： a potential biomarker for schizophrenia［J］. Medical Engineering and Physics， 2010， 32（9）：1074-1079. 10.1016/j.medengphy.2010.07.013
46	MARTÍNEZ-SÁNCHEZ F， MUELA-MARTÍNEZ J A， CORTÉS-SOTO P， et al. Can the acoustic analysis of expressive prosody discriminate schizophrenia？［J］. The Spanish Journal of Psychology， 2015， 18： No.e86. 10.1017/sjp.2015.85
47	ZOUGHI T， HOMAYOUNPOUR M M， DEYPIR M. Adaptive windows multiple deep residual networks for speech recognition［J］. Expert Systems with Applications， 2020， 139： No.112840. 10.1016/j.eswa.2019.112840
48	WU Y， MAO H， YI Z. Audio classification using attention-augmented convolutional neural network［J］. Knowledge-Based Systems， 2018， 161：90-100. 10.1016/j.knosys.2018.07.033
49	LIU B， WANG Z， GUO S S， et al. An energy-efficient voice activity detector using deep neural networks and approximate computing［J］. Microelectronics Journal， 2019， 87：12-21. 10.1016/j.mejo.2019.03.009
50	BREZNITZ Z. The speech and vocalization patterns of boys with ADHD compared with boys with dyslexia and boys without learning disabilities［J］. The Journal of Genetic Psychology， 2003， 164（4）：425-452. 10.1080/00221320309597888
51	BARONA-LLEO L， FERNANDEZ S. Hyperfunctional voice disorder in children with Attention Deficit Hyperactivity Disorder （ADHD）. a phenotypic characteristic？［J］. Journal of Voice， 2016， 30（1）：114-119. 10.1016/j.jvoice.2015.03.002

基于语音停顿度和平坦度的注意缺陷与多动障碍自动检测算法

Automatic detection algorithm for attention deficit/hyperactivity disorder based on speech pause and flatness

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 51

相关文章 1

编辑推荐

Metrics