| NSF Org: |
IIS Division of Information & Intelligent Systems |
| Recipient: |
|
| Initial Amendment Date: | August 18, 2016 |
| Latest Amendment Date: | August 18, 2016 |
| Award Number: | 1602337 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Sylvia Spengler
sspengle@nsf.gov (703)292-7347 IIS Division of Information & Intelligent Systems CSE Directorate for Computer and Information Science and Engineering |
| Start Date: | September 1, 2016 |
| End Date: | August 31, 2020 (Estimated) |
| Total Intended Award Amount: | $678,850.00 |
| Total Awarded Amount to Date: | $678,850.00 |
| Funds Obligated to Date: |
|
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
5000 FORBES AVE PITTSBURGH PA US 15213-3815 (412)268-8746 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
5000 Forbes Ave Pittsburgh PA US 15213-1000 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): | Smart and Connected Health |
| Primary Program Source: |
|
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.070 |
ABSTRACT
Parkinson's Disease (PD) poses a serious threat to the elderly population, affecting as many as one million Americans. There is no cure, and medications can only provide symptomatic relief. In addition, costs associated with PD, including treatment, social security payments, and lost income from inability to work, is estimated to be nearly $25 billion per year in the United States alone. The current state-of-the-art in PD management suffers from several shortcomings: (1) frequent clinic visits are a major contributor to the high cost of PD treatment and are inconvenient for the patient, especially in a population for which traveling is difficult; (2) inaccurate patient self-reports and 15-20 minute clinic visits are not enough information for doctors to accurately assess their patients, leading to difficulties in monitoring patient symptoms and medication response; and (3) motor function assessments are subjective, making it difficult to monitor disease progression. Furthermore, because they must be performed by a trained clinician, it is infeasible to do frequent motor function assessments. This project aims to promote a paradigm shift in PD management through in-home monitoring using wearable accelerometers and machine learning. Novel algorithms and experimental protocols are developed to allow for robust detection and assessment of PD motor symptoms during daily living environments.
Specifically, this project develops algorithms for weakly-supervised learning, time series analysis, and personalization of classifiers. In previous studies, data was collected in controlled environments for a short amount of time (1-4 hours) and manually labeled for fully-supervised learning. In contrast, this project collects long-term (several weeks), in-home data where the participants' actions are natural and unscripted. Participants use a cell phone app to label their own data, marking segments of time as containing or not containing the occurrence of a PD motor symptom. Since the exact time of the symptom is unknown, this constitutes weakly-labeled data. This project extends multiple-instance learning algorithms for learning from weakly-labeled data in time series. Additional major technical challenges include detection of subtle motor symptoms and local minima during optimization. To further increase robustness and generalization, this project explores the use of personalization algorithms to learn person-specific models of motor symptoms from unsupervised data. The proposed techniques for weakly-supervised learning and personalization are general, and they can be applied to other human sensing problems.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
An estimated 10 million people worldwide live with Parkinson’s Disease (PD), a chronic, neuro-degenerative disorder that leads to both non-motor and motor symptoms. These symptoms include, but are not limited to, depression, anxiety, sleep disorders, slowness, muscle rigidity, postural instability,and tremors. While there is no cure, medication can provide symptomatic relief. However, dosages need to be adjusted as a patient’s disease progresses and symptoms worsen.
This research presents a continuous PD motor symptom monitoring for tremor that would potentially enable clinicians to better adjust medication andthereby improve their patients’ quality of life. Our eventual goal is to build a system for continuous monitoring of PD motor symptoms “in the wild” –i.e., in natural environments without requiring any specific interaction from the patient. Such a system would use machine learning algorithms to detect symptoms in data collected from wearable accelerometers (see Fig. 1).
Fig. 1.Depiction of automated tremor detection with wearable sensorsduring everyday activities
Machine learning algorithms for symptom detection typ-ically require accurate labels (i.e., the start and end of each symptom). Labels are time consuming to annotate and the exact onset of a symptom can be subjective. These issues are compounded in wild settings. Therefore, researchers oftenuse data collected in laboratory settings for training. Fewer researchers have explored the use of wild data for trainingbecause labels for these data are typically supplied by PD patients via paper diaries. Therefore, only approximate tremor timestamps of symptom occurrences are available (typically±1 hour). That is, these data are weakly-labeled.
In this research, we assess how well laboratory data represents wild data by comparing PD symptom detection performance of three models on laboratory versus wilddata. Similar performance across datasets would imply that findings on laboratory data should transfer to the wild. Results from each of the models, however, show that laboratory data is not representative of wild data. Of the three models, we find that the one trained on weakly labeled wild data has better performance on wild data than the onestrained on accurately labeled laboratory data. While this research focuses on upper-limb PD tremor, we expect results totranslate to other PD symptoms. The findings in this research may also generalize to other problems in human activity understanding, such as monitoring of other motor impairments,activity tracking, or sports performance analysis
Last Modified: 05/06/2021
Modified by: Jessica Hodgins
Please report errors in award information by writing to: awardsearch@nsf.gov.
