Title :
Detecting Parkinsons´ symptoms in uncontrolled home environments: A multiple instance learning approach
Author :
Das, S. ; Amoedo, B. ; De la Torre, Fernando ; Hodgins, J.
Author_Institution :
Robot. Inst., Carnegie Mellon Univ., Pittsburgh, PA, USA
fDate :
Aug. 28 2012-Sept. 1 2012
Abstract :
In this paper, we propose to use a weakly supervised machine learning framework for automatic detection of Parkinson´s Disease motor symptoms in daily living environments. Our primary goal is to develop a monitoring system capable of being used outside of controlled laboratory settings. Such a system would enable us to track medication cycles at home and provide valuable clinical feedback. Most of the relevant prior works involve supervised learning frameworks (e.g., Support Vector Machines). However, in-home monitoring provides only coarse ground truth information about symptom occurrences, making it very hard to adapt and train supervised learning classifiers for symptom detection. We address this challenge by formulating symptom detection under incomplete ground truth information as a multiple instance learning (MIL) problem. MIL is a weakly supervised learning framework that does not require exact instances of symptom occurrences for training; rather, it learns from approximate time intervals within which a symptom might or might not have occurred on a given day. Once trained, the MIL detector was able to spot symptom-prone time windows on other days and approximately localize the symptom instances. We monitored two Parkinson´s disease (PD) patients, each for four days with a set of five triaxial accelerometers and utilized a MIL algorithm based on axis parallel rectangle (APR) fitting in the feature space. We were able to detect subject specific symptoms (e.g. dyskinesia) that conformed with a daily log maintained by the patients.
Keywords :
accelerometers; biomedical equipment; diseases; learning (artificial intelligence); medical computing; medical disorders; patient monitoring; support vector machines; APR fitting; MIL algorithm; MIL detector; PD patients; Parkinson disease motor symptom detection; Parkinson disease patients; automatic detection; axis parallel rectangle fitting; clinical feedback; daily living environments; feature space; in-home monitoring; medication cycles; monitoring system; multiple instance learning approach; supervised learning classifiers; supervised machine learning framework; support vector machines; symptom-prone time windows; time intervals; triaxial accelerometers; uncontrolled home environments; Accelerometers; Diseases; Feature extraction; Frequency domain analysis; Monitoring; Supervised learning; Training; Parkinson´s Disease (PD); continuous motor symptom monitoring; multiple instance learning; Algorithms; Artificial Intelligence; Dyskinesias; Female; Humans; Male; Middle Aged; Monitoring, Physiologic; Motor Activity; Parkinson Disease;
Conference_Titel :
Engineering in Medicine and Biology Society (EMBC), 2012 Annual International Conference of the IEEE
Conference_Location :
San Diego, CA
Print_ISBN :
978-1-4244-4119-8
Electronic_ISBN :
1557-170X
DOI :
10.1109/EMBC.2012.6346767