Early Illness Recognition Using In-home Monitoring Sensors and Multiple Instance Learning

Summary

Background: Many older adults in the US prefer to live independently for as long as they are able, despite the onset of conditions such as frailty and dementia. Solutions are needed to enable independent living, while enhancing safety and peace of mind for their families. Elderly patients are particularly at-risk for late assessment of cognitive changes.

Objectives: We predict early signs of illness in older adults by using the data generated by a continuous, unobtrusive nursing home monitoring system.

Methods: We describe the possibility of employing a multiple instance learning (MIL) framework for early illness detection. The MIL framework is suitable for training classifiers when the available data presents temporal or location uncertainties.

Results: We provide experiments on three datasets that prove the utility of the MIL framework. We first tuned our algorithms on a set of 200 normal/abnormal behavior patterns produced by a dedicated simulator. We then conducted two retrospective studies on residents from the Tiger Place aging in place facility, aged over 70, which have been monitored with motion and bed sensors for over two years. The presence or absence of the illness was manually assessed based on the nursing visit reports.

Conclusions: The use of simulated sensor data proved to be very useful for algorithm development and testing. The results obtained using MIL for six Tiger Place residents, an average area under the receiver operator characteristic curve (AROC) of 0.7, are promising. However, more sophisticated MIL classifiers are needed to improve the performance.

Keywords

Automated pattern recognition - multiple instance learning - eldercare

• References

• 1 Andrews S, Tsochantaridis I, Hofmann T. Support vector machines for multiple-instance learning. Advances NIPS 2003: 561-568.
• 2 Chen JP, Reich L, Chung H. Anxiety disorders. West J Med 2002; 176: 249-253.
• 3 Cuddihy P, Weisenberg J, Graichen C, Ganesh M. Algorithm to automatically detect abnormally long periods of inactivity in a home. Proc. of the 1st ACM SIGMOBILE Intl Workshop New York: 2007: 89-94.
• 4 Dietterich TG, Lathrop RH, Lozano-Perez T. Solving the Multiple-Instance Problem with Axis-Parallel Rectangles. Artificial Intelligence Journal 1997; 89: 31-71.
• 5 Gale CK, Millichamp J. Generalised anxiety disorder. Clinical Evidence 2011; 10: 1002
• 6 Godsey C, Skubic M. Using Elements of Game Engine Architecture to Simulate Sensor Networks for ElderCare. Proc 31st Ann Int Conf of the IEEE EMBS; Minneapolis, MN; Sept.(2-6). 2009: 6143-6146.
• 7 Hayes TL, Pavel M, Kaye JA. An unobtrusive in-home monitoring system for detection of key motor changes preceding cognitive decline. Proc of the 26th Annual Intl Conf of the IEEE EMBS. San Francisco, CA: 2004: 2480-2483.
• 8 Maron O, Lozano-Pérez T. A framework for multiple-instance learning. Proc of the 1997 Conf on ANIPS 1998; (10) 570-576.
• 9 Maron O, Ratan AL. Multiple-Instance Learning for Natural Scene Classification. Proc 15th ICML 1998: 341-349.
• 10 Morris M, Intille S, Beaudin JS. Embedded assessment: overcoming barriers to early detection with pervasive computing. In: Proc of PERVASIVE 2005 Gellersen GW, Want R, Schmidt A. eds Springer-Verlag; 2005: 333-346.
• 11 Popescu M, Florea E, Skubic M, and Rantz M. Prediction of Elevated Pulse Pressure in Elderly Using In-Home Monitoring Sensors: A Pilot Study. 4th IET Int Conf on Int Env; Seattle; July 21-22. 2008.
• 12 Popescu M, Skubic M, Rantz M. Predicting abnormal clinical events using non-wearable sensors in elderly. AMIA Fall Symp; San Francisco, CA: 2009. Nov 14-18 1010
• 13 Popescu M, Mahnot A. Early illness recognition in older adults using in-home monitoring sensors and multiple instance learning. IDAMAP-10 Washington DC: Nov 12, 2010 21-25.
• 14 Rantz MJ, Marek KD, Aud MA, Johnson RA, Otto D, Porter R. Tiger Place: A New Future for Older Adults. J of Nursing Care Quality 2005; 20 (01) 1-4.
• 15 Rowan J, Mynatt ED. Digital Family Portrait field trial: support for Aging in Place”. Proc SIGCHI Conf on Human Factors in Comp. Sys. New York: ACM Press; 2005: 521-530.
• 16 Skubic M, Alexander G, Popescu M, Rantz M, Keller J. A Smart Home Application to Eldercare: Current Status and Lessons Learned. Tech and Health Care 2009; 17 (03) 183-201.
• 17 Tax DMJ. One-class classification. PhD Thesis; TU Delft, NL: 2001.
• 18 Zhang Q, Goldman SA, Yu W, Fritts JE. Content-Based Image Retrieval Using Multiple-Instance Learning. Proc of the 19th ICML; July 2002: 682-689.
• 19 Demiris G, Rantz MJ, Aud MA, Marek KD, Tyrer HW, Skubic M, and Hussam AA. Seniors’ Attitudes Towards Home-based Assistive Technologies. 29th Annual MNRS Research Conference; Cincinnati, Ohio, April 1-4. 2005.
• 20 Demiris G, Skubic M, Rantz MJ, Harris K, Hensel B, Aud MA, Lee J, Burks K, Oliver DR, He Z, Tyrer HW, Keller J. Older Adults’ Attitudes Towards Smart Home Features. International Conference on Aging, Disability and Independence (ICADI); St Petersburg, Florida; February 2006.
• 21 Demiris G, Skubic M, Rantz M, Hensel B. Smart Home Sensors for Aging in Place: Older Adults' Attitudes and Willingness to Adopt. The Gerontologist 2006; 46 (Special Issue 1) 430
• 22 Mahnot A, Popescu M. “FUMIL-Fuzzy Multiple Instance Learning for Early Illness Recognition in Older Adults. To appear in Proc IEEE WCCI; Brisbane, Australia: 2012.
• 23 Rantz MJ, Skubic M, Koopman RJ, Alexander G, Phillips L, Musterman KI, Back JR, Aud MA, Galambos C, Guevara RD, Miller SJ. Automated technology to speed recognition of signs of illness in older adults. In press -2012, Journal Gerontological Nursing 2012; 38 (04) 18-23.
• 24 Popescu M, Chronis G, Ohol R, Skubic M, Rantz M. An Eldercare Electronic Health Record System for Predictive Health Assessment. IEEE 13th International Conference on e-Health Networking, Applications and Services; Columbia, MO; June 13-15 2011: 194-196.
• 25 Skubic M, Guevara RD, Rantz M. Testing classifiers for embedded health assessment. To appear in Proc of Intl Conf on Smart Homes and Health Telematics; June 12-15. 2012. Italy
• 26 Mack DCM, Alwanand B. Turner. A Passive and Portable System for Monitoring Heart Rate and Detecting Sleep Apnea and Arousals: Preliminary validation. Proc of the Transdisciplinary Conference on Distributed Diagnosis and Home Healthcare 2006 D2H2; Arlington, VA 2006: 51-54.