ABSTRACT
The increasing use of electronic forms of communication presents new opportunities in the study of mental health, including the ability to investigate the manifestations of psychiatric diseases unobtrusively and in the setting of patients' daily lives. A pilot study to explore the possible connections between bipolar affective disorder and mobile phone usage was conducted. In this study, participants were provided a mobile phone to use as their primary phone. This phone was loaded with a custom keyboard that collected metadata consisting of keypress entry time and accelerometer movement. Individual character data with the exceptions of the backspace key and space bar were not collected due to privacy concerns. We propose an end-to-end deep architecture based on late fusion, named DeepMood, to model the multi-view metadata for the prediction of mood scores. Experimental results show that 90.31% prediction accuracy on the depression score can be achieved based on session-level mobile phone typing dynamics which is typically less than one minute. It demonstrates the feasibility of using mobile phone metadata to infer mood disturbance and severity.
- Charu C Aggarwal. 2002. On effective classification of strings with wavelets KDD. ACM, 163--172.Google Scholar
- David Ankers and Steven H Jones 2009. Objective assessment of circadian activity and sleep patterns in individuals at behavioural risk of hypomania. Journal of clinical psychology Vol. 65, 10 (2009), 1071--1086. Google ScholarCross Ref
- American Psychiatric Association and others 2013. Diagnostic and statistical manual of mental disorders (DSM-5®). American Psychiatric Pub.Google Scholar
- Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. 2014. Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473 (2014).Google Scholar
- Yoshua Bengio, Patrice Simard, and Paolo Frasconi. 1994. Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks Vol. 5, 2 (1994), 157--166. Google ScholarDigital Library
- Jedediah M Bopp, David J Miklowitz, Guy M Goodwin, Will Stevens, Jennifer M Rendell, and John R Geddes. 2010. The longitudinal course of bipolar disorder as revealed through weekly text messaging: a feasibility study. Bipolar disorders, Vol. 12, 3 (2010), 327--334.Google Scholar
- Bokai Cao, Lifang He, Xiangnan Kong, Philip S Yu, Zhifeng Hao, and Ann B Ragin. 2014. Tensor-based Multi-view Feature Selection with Applications to Brain Diseases ICDM.Google Scholar
- Bokai Cao, Xiangnan Kong, Jingyuan Zhang, Philip S Yu, and Ann B Ragin 2015. Mining Brain Networks using Multiple Side Views for Neurological Disorder Identification ICDM.Google Scholar
- Bokai Cao, Hucheng Zhou, Guoqiang Li, and Philip S Yu. 2016. Multi-view Machines WSDM. ACM, 427--436.Google Scholar
- Tianqi Chen and Carlos Guestrin 2016. XGBoost: A Scalable Tree Boosting System. In KDD. ACM. Google ScholarDigital Library
- Betty Yee Man Cheng, Jaime G Carbonell, and Judith Klein-Seetharaman 2005. Protein classification based on text document classification techniques. Proteins: Structure, Function, and Bioinformatics, Vol. 58, 4 (2005), 955--970.Google ScholarCross Ref
- Kyunghyun Cho, Bart Van Merriënboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. 2014. Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078 (2014).Google Scholar
- Franccois Chollet. 2015. Keras. https://github.com/fchollet/keras. (2015).Google Scholar
- Junyoung Chung, Caglar Gulcehre, KyungHyun Cho, and Yoshua Bengio 2014. Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555 (2014).Google ScholarDigital Library
- Mark A Demitrack, Doug Faries, John M Herrera, David J DeBrota, and William Z Potter 1998. The problem of measurement error in multisite clinical trials. Psychopharmacology bulletin Vol. 34, 1 (1998), 19.Google Scholar
- Hui Ding, Goce Trajcevski, Peter Scheuermann, Xiaoyue Wang, and Eamonn Keogh. 2008. Querying and mining of time series data: experimental comparison of representations and distance measures. VLDB, Vol. 1, 2 (2008), 1542--1552. Google ScholarDigital Library
- Martín Abadi et al. 2015. TensorFlow: Large-Scale Machine Learning on Heterogeneous Systems. (2015). http://tensorflow.org/shownoteSoftware available from tensorflow.org.Google Scholar
- Maria Faurholt-Jepsen, Maj Vinberg, Mads Frost, Sune Debel, Ellen Margrethe Christensen, Jakob E Bardram, and Lars Vedel Kessing. 2016. Behavioral activities collected through smartphones and the association with illness activity in bipolar disorder. International journal of methods in psychiatric research, Vol. 25, 4 (2016), 309--323. Google ScholarCross Ref
- Mads Frost, Afsaneh Doryab, Maria Faurholt-Jepsen, Lars Vedel Kessing, and Jakob E Bardram. 2013. Supporting disease insight through data analysis: refinements of the monarca self-assessment system. In UBICOMP. ACM, 133--142.Google Scholar
- Alex Graves, Abdel-rahman Mohamed, and Geoffrey Hinton. 2013. Speech recognition with deep recurrent neural networks ICASSP. IEEE, 6645--6649.Google Scholar
- Agnes Gruenerbl, Venet Osmani, Gernot Bahle, Jose C Carrasco, Stefan Oehler, Oscar Mayora, Christian Haring, and Paul Lukowicz. 2014. Using smart phone mobility traces for the diagnosis of depressive and manic episodes in bipolar patients. In AH. ACM, 38.Google Scholar
- Geoffrey E Hinton, Nitish Srivastava, Alex Krizhevsky, Ilya Sutskever, and Ruslan R Salakhutdinov. 2012. Improving neural networks by preventing co-adaptation of feature detectors. arXiv preprint arXiv:1207.0580 (2012).Google Scholar
- Sepp Hochreiter. 1998. The vanishing gradient problem during learning recurrent neural nets and problem solutions. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, Vol. 6, 02 (1998), 107--116. Google ScholarDigital Library
- Sepp Hochreiter and Jürgen Schmidhuber 1997. Long short-term memory. Neural computation, Vol. 9, 8 (1997), 1735--1780. Google ScholarDigital Library
- Xiaonan Ji, James Bailey, and Guozhu Dong 2007. Mining minimal distinguishing subsequence patterns with gap constraints. Knowledge and Information Systems Vol. 11, 3 (2007), 259--286. Google ScholarDigital Library
- Eamonn Keogh and Shruti Kasetty 2003. On the need for time series data mining benchmarks: a survey and empirical demonstration. Data Mining and Knowledge Discovery Vol. 7, 4 (2003), 349--371. Google ScholarDigital Library
- Eamonn J Keogh and Michael J Pazzani 2000. Scaling up dynamic time warping for datamining applications KDD. ACM, 285--289.Google Scholar
- Ronald C Kessler, Patricia Berglund, Olga Demler, Robert Jin, Kathleen R Merikangas, and Ellen E Walters 2005. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of general psychiatry Vol. 62, 6 (2005), 593--602.Google Scholar
- Kiran E Laxman, Kate S Lovibond, and Mariam K Hassan. 2008. Impact of bipolar disorder in employed populations. The American journal of managed care Vol. 14, 11 (2008), 757--764.Google Scholar
- Neal Lesh, Mohammed J Zaki, and Mitsunori Ogihara. 1999. Mining features for sequence classification. In KDD. ACM, 342--346.Google Scholar
- Christina Leslie and Rui Kuang 2004. Fast string kernels using inexact matching for protein sequences. Journal of Machine Learning Research Vol. 5, Nov (2004), 1435--1455.Google ScholarDigital Library
- Huma Lodhi, Craig Saunders, John Shawe-Taylor, Nello Cristianini, and Chris Watkins. 2002. Text classification using string kernels. Journal of Machine Learning Research Vol. 2, Feb (2002), 419--444.Google ScholarDigital Library
- Chun-Ta Lu, Lifang He, Weixiang Shao, Bokai Cao, and Philip S Yu 2017. Multilinear Factorization Machines for Multi-Task Multi-View Learning WSDM.Google Scholar
- Tomas Mikolov, Martin Karafiát, Lukas Burget, Jan Cernockỳ, and Sanjeev Khudanpur. 2010. Recurrent neural network based language model. In Interspeech, Vol. Vol. 2. 3.Google Scholar
- Pamela B Peele, Ying Xu, and David J Kupfer 2003. Insurance expenditures on bipolar disorder: clinical and parity implications. American Journal of Psychiatry Vol. 160, 7 (2003), 1286--1290.Google ScholarCross Ref
- Bruce M Psaty and Ross L Prentice 2010. Minimizing bias in randomized trials: the importance of blinding. Jama, Vol. 304, 7 (2010), 793--794. Google ScholarCross Ref
- Alessandro Puiatti, Steven Mudda, Silvia Giordano, and Oscar Mayora 2011. Smartphone-centred wearable sensors network for monitoring patients with bipolar disorder EMBC. IEEE, 3644--3647.Google Scholar
- Chotirat Ann Ratanamahatana and Eamonn Keogh 2004. Making Time-series Classification More Accurate Using Learned Constraints SDM. SIAM, 11.Google Scholar
- Steffen Rendle. 2012. Factorization machines with libfm. ACM Transactions on Intelligent Systems and Technology, Vol. 3, 3 (2012), 57. Google ScholarDigital Library
- O Schleusing, Ph Renevey, M Bertschi, J-M Koller, R Paradiso, and others. 2011. Monitoring physiological and behavioral signals to detect mood changes of bipolar patients ISMICT. IEEE, 130--134.Google Scholar
- Rong She, Fei Chen, Ke Wang, Martin Ester, Jennifer L Gardy, and Fiona SL Brinkman 2003. Frequent-subsequence-based prediction of outer membrane proteins KDD. ACM, 436--445.Google Scholar
- Sören Sonnenburg, Gunnar Rätsch, and Bernhard Schölkopf 2005. Large scale genomic sequence SVM classifiers. In ICML. ACM, 848--855. Google ScholarDigital Library
- Prashant K Srivastava, Dhwani K Desai, Soumyadeep Nandi, and Andrew M Lynn 2007. HMM-ModE--Improved classification using profile hidden Markov models by optimising the discrimination threshold and modifying emission probabilities with negative training sequences. BMC bioinformatics, Vol. 8, 1 (2007), 1. Google ScholarCross Ref
- Ilya Sutskever, James Martens, and Geoffrey E Hinton. 2011. Generating text with recurrent neural networks. In ICML. 1017--1024.Google Scholar
- Tijmen Tieleman and Geoffrey Hinton 2012. Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude. COURSERA: Neural Networks for Machine Learning, Vol. 4, 2 (2012).Google Scholar
- Gaetano Valenza, Mimma Nardelli, Antonio Lanata, Claudio Gentili, Gilles Bertschy, Rita Paradiso, and Enzo Pasquale Scilingo. 2014. Wearable monitoring for mood recognition in bipolar disorder based on history-dependent long-term heart rate variability analysis. IEEE Journal of Biomedical and Health Informatics, Vol. 18, 5 (2014), 1625--1635. Google ScholarCross Ref
- Li Wei and Eamonn Keogh 2006. Semi-supervised time series classification. In KDD. ACM, 748--753. Google ScholarDigital Library
- Janet BW Williams. 1988. A structured interview guide for the Hamilton Depression Rating Scale. Archives of general psychiatry Vol. 45, 8 (1988), 742--747. Google ScholarCross Ref
- Xiaopeng Xi, Eamonn Keogh, Christian Shelton, Li Wei, and Chotirat Ann Ratanamahatana. 2006. Fast time series classification using numerosity reduction ICML. ACM, 1033--1040.Google Scholar
- Zhengzheng Xing, Jian Pei, and Eamonn Keogh 2010. A brief survey on sequence classification. ACM SIGKDD Explorations Newsletter Vol. 12, 1 (2010), 40--48. Google ScholarDigital Library
- Oksana Yakhnenko, Adrian Silvescu, and Vasant Honavar. 2005. Discriminatively trained markov model for sequence classification ICDM. IEEE, 8--pp.Google Scholar
- Lexiang Ye and Eamonn Keogh 2009. Time series shapelets: a new primitive for data mining KDD. ACM, 947--956. Google ScholarDigital Library
- RC Young, JT Biggs, VE Ziegler, and DA Meyer. 1978. A rating scale for mania: reliability, validity and sensitivity. The British Journal of Psychiatry Vol. 133, 5 (1978), 429--435. Google ScholarCross Ref
- Jingyuan Zhang, Bokai Cao, Sihong Xie, Chun-Ta Lu, Philip S Yu, and Ann B Ragin. 2016natexlaba. Identifying Connectivity Patterns for Brain Diseases via Multi-side-view Guided Deep Architectures. In SDM.Google Scholar
- Weinan Zhang, Tianming Du, and Jun Wang 2016. Deep Learning over Multi-field Categorical Data. ECIR. Springer, 45--57. Google ScholarCross Ref
Index Terms
- DeepMood: Modeling Mobile Phone Typing Dynamics for Mood Detection
Recommendations
DeepMood: Forecasting Depressed Mood Based on Self-Reported Histories via Recurrent Neural Networks
WWW '17: Proceedings of the 26th International Conference on World Wide WebDepression is a prevailing issue and is an increasing problem in many people's lives. Without observable diagnostic criteria, the signs of depression may go unnoticed, resulting in high demand for detecting depression in advance automatically. This ...
Towards personalised ambient monitoring of mental health via mobile technologies
Managing bipolar disorder is an important health issue that can strongly affect the patient's quality of life during occurrences of depressive or manic episodes and is therefore a growing burden to healthcare systems. A widely practised method of ...
Towards long term monitoring of electrodermal activity in daily life
Manic depression, also known as bipolar disorder, is a common and severe form of mental disorder. The European research project MONARCA aims at developing and validating mobile technologies for multi-parametric, long term monitoring of physiological and ...
Comments