Development of a decision support model for screening attention-deficit hyperactivity disorder with actigraph-based measurements of classroom activity

Summary

Objective: Questionnaire-based ADHD screening tests may not always be objective or accurate, owing to both subjectivity and prejudice. Despite attempts to develop objective measures to characterize ADHD, no widely applicable index currently exists. The principal aim of this study was to develop a decision support model for ADHD screening by monitoring children’s school activities using a 3-axial actigraph.

Methods: Actigraphs were placed on the non-dominant wrists of 153 children for 3 hours, while they were at school. Children who scored high on the questionnaires were clinically examined by child psychiatrists, who then confirmed ADHD. Mean, variance, and ratios of low-level (0.5-1.0G) and high-level (1.6-3.2G) activity were extracted as activity features from 142 children (10 ADHD, 132 non-ADHD). Two decision-tree models were constructed using the C5.0 algorithm: [A] from whole hours (class + playtime) and [B] during classes. Accuracy, sensitivity, and specificity were evaluated. PPV, NPV, likelihood ratio, and AUC were also calculated for evaluation.

Results: [Model A] One child without ADHD was misclassified, resulting in an accuracy score of 99.30%. Sensitivity and NPV were 1.0000. Specificity and PPV were 0.992 and 0.803-0.909, respectively. [Model B] Two children without ADHD were misclassified, resulting in an accuracy score of 98.59%. Specificity and PPV were scored at 0.985 and 0.671-0.832, respectively.

Conclusion: The selected features were consistent with the findings of previous studies. Objective screening of latent patients with ADHD can be accomplished with a simple watch-like sensor, which is worn for just a few hours while the child attends school. The model proposed herein can be applied to a great many children without heavy cost in time and manpower cost, and would generate valuable results from a public health perspective.

Keywords

Attention deficit disorder with hyperactivity - actigraphy - activities of daily living - mass screening

• References

• 1 Teicher MH, Ito Y, Glod CA, Barber NI. Objective measurement of hyperactivity and attentional problems in ADHD. J Am Acad Child Adolesc Psychiatry 1996; 35 (03) 334-342.
• 2 Kofler MJ, Rapport MD, Alderson RM. Quantifying ADHD classroom inattentiveness, its moderators, and variability: a meta-analytic review. J Child Psychol Psychiatry 2008; 49 (01) 59-69.
• 3 Goodyear P, Hynd GW. Attention-deficit disorder with (ADD/H) and without (ADD/WO) hyperactivity: behavioral and neuropsychological differentiation. J Clin Psychol 1992; 21 (03) 273-305.
• 4 Dane AV, Schachar RJ, Tannock R. Does actigraphy differentiate ADHD subtypes in a clinical research setting?. J Am Acad Child Adolesc Psychiatry 2000; 39 (06) 752-760.
• 5 Teicher MH. Actigraphy and motion analysis: new tools for psychiatry. Harv Rev Psychiatry 1995; 3 (01) 18-35.
• 6 Swanson JM, Gupta S, Williams L, Agler D, Lerner M, Wigal S. Efficacy of a new pattern of delivery of methylphenidate for the treatment of ADHD: Effects on activity level in the classroom and on the playground. J Am Acad Child Adolesc Psychiatry 2002; 41 (11) 1306-1314.
• 7 Pärkkä J, Ermes M, Korpipää P, Mäntyjärvi J, Peltola J, Korhonen I. Activity classification using realistic data from wearable sensors. IEEE Trans Inf Technol Biomed 2006; 10 (01) 119-128.
• 8 Choudhury T, Philipose M, Wyatt D, Lester J. Towards activity databases: using sensors and statistical models to summarize people’s lives. IEEE Data Eng Bull 2006; 29 (01) 49-58.
• 9 Hong JH, Kim NJ, Cha EJ, Lee TS. Classification technique of human motion context based on wireless sensor network. Conf Proc IEEE Eng Med Biol Soc 2005; 5: 5201-5202.
• 10 Liao L, Fox D, Kautz H. Location-based activity recognition using relational markov networks. In: Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI). 2005, 30 July - 5 August 2005 Edinburgh, Scotland: 2005: 773.
• 11 Burchfield TR, Venkatesan S. Accelerometer-based human abnormal movement detection in wireless sensor networks. In: 1st International Workshop on Systems and Networking support for Healthcare and Assisted living Environments (HealthNet’s 07). 11 June 2007; San Juan, Puerto Rico, USA: 2007: 67-69.
• 12 Zhang T, Wang J, Liu P, Hou J. Fall detection by embedding an accelerometer in cellphone and using KFD algorithm. Intl. J Comp Sci Net Sec 2006; 6 (10) 277-284.
• 13 Chen J, Kwong K, Chang D, Luk J, Bajcsy R. Wearable sensors for reliable fall detection. Conf Proc IEEE Eng Med Biol Soc 2005; 4: 3551-3554.
• 14 Laerhoven KV, Gellersen HW, Malliaris YG. Long-term activity monitoring with a wearable sensor node. In: International Workshop on Wearable and Implantable Body Sensor Networks (BSN‘06). 3-5 April 2006 Cambridge, Massachusetts, USA: 2006: 170-174.
• 15 Tuisku K, Virkkunen M, Holi M, Lauerma H, Naukkarinen H, Rimon R. et al. Antisocial violent offenders with attention deficit hyperactivity disorder demonstrate akathisia-like hyperactivity in three-channel actometry. J Neuropsychiatry Clin Neurosci 2003; 15 (02) 194-199.
• 16 Tuisku K, Lauerma H, Holi M, Markkula J, Rimon R. Measuring neuroleptic-induced akathisia by three-channel actometry. Schizophr Res 1999; 40 (02) 105-110.
• 17 Van Someren EJW, Pticek MD, Speelman JD, Schuurman PR, Esselink R, Swaab DF. New actigraph for long-term tremor recording. Mov disord 2006; 21 (08) 1136-1143.
• 18 Halperin JM, Matier K, Bedi G, Sharma V, Newcorn JH. Specificity of inattention, impulsivity, and hyperactivity to the diagnosis of attention-deficit hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 1992; 31 (02) 190-196.
• 19 Porrino LJ, Rapoport JL, Behar D, Sceery W, Ismond DR, Bunney Jr. WE. A naturalistic assessment of the motor activity of hyperactive boys. I. Comparison with normal controls. Arch Gen Psychiatry 1983; 40 (06) 681-687.
• 20 Halperin JM, Newcorn JH, Matier K, Sharma V, KMS Mckay KE, Schwartz S. Discriminant validity of attention-deficit hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 1993; 32 (05) 1038-1043.
• 21 Tsujii N, Okada A, Kaku R, Kuriki N, Hanada K, Matsuo J. et al. Association between activity level and situational factor in children with attention deficit/hyperactivity disorder in elementary school. Psychiatry Clin Neurosci 2007; 61 (02) 181-185.
• 22 Wood AC, Asherson P, Rijsdijk F, Kuntsi J. Is overactivity a core feature in ADHD? Familial and receiver operating characteristic curve analysis of mechanically assessed activity level. J Am Acad Child Adolesc Psychiatry 2009; 48 (10) 1023-1030.
• 23 Dupaul GJ, Power TJ, Anastopoulos AD, Reid R. ADHD Rating Scale-IV: Checklists, norms, and clinical interpretation. New York, NY: The Guilford Press; 1998
• 24 Kim YS, Cheon KA, Kim BN, Chang SA, Yoo HJ, Kim JW. et al. The reliability and validity of Kiddie-Schedule for affective disorders and schizophrenia-present and lifetime version-Korean version (K-SADSPL-K). Yonsei Med J 2004; 45 (01) 81-89.
• 25 Teicher MH, McGreenery CE, Ohashi K. Actigraph assessment of rest-activity disturbances in psychiatric disorders. In: Psychosomatic Medicine - Proceedings of the 18th World Congress on Psychosomatic Medicine. 21-26 August 2005 Kobe, Japan: 2006: 32-37.
• 26 Yun C, Shin D, Jo H, Yang J, Kim S. An experimental study on feature subset selection methods. In: 7th IEEE International Conference on Computer and Information Technology (CIT). 16-19 October 2007 Fukushima, Japan: 2007: 77-82.
• 27 Han J, Kamber M. Data Mining: Concepts and Techniques. 2nd ed. Academic Press (CA): Morgan Kaufmarm Publishers, Inc.; 2001: 284-291.
• 28 Tan P-N, Steinbach M, Kumur V. Introduction to data mining. 1st ed. Addison-Wesley (MA): Pearson Education, Inc.; 2006
• 29 Matthew NA, Sajjan S. Comparative analysis of serial decision-tree classification algorithms. International Journal of Computer Science and Security 2009; 3 (03) 230-240.
• 30 Rulequest research 2009 [Internet]. Australia: RuleQuest Research Pty Ltd. c1997-2008 [updated 2009 Nov; cited 2010 Jul 23]. Is see5/C5.0 setter than C4.5? Available from: http://www.rulequest.com/see5-comparison.html
• 31 Quinlan JR. Programs for Machine Learning. Morgan Kaufmann Publishers, Inc.; 1993
• 32 Japkowicz N, Stephen S. The class imbalance problem: a systematic study. Intelligent Data Analysis 2002; 6 (05) 429-449.
• 33 Kim JW, Park KH, Cheon KA, Kim BN, Cho SC, Hong KEM. The child behavior checklist together with the ADHD rating scale can diagnose ADHD in Korean community-based samples. Can J Psychiatry 2005; 50: 802-805.
• 34 Dabkowska MM, Pracka D, Pracki T. Does actigraphy differentiate ADHD subtypes?. Eur Psychiatry 2007; 22 (Suppl. 01) S319.