1
|
O’Brien MK, Lanotte F, Khazanchi R, Shin SY, Lieber RL, Ghaffari R, Rogers JA, Jayaraman A. Early Prediction of Poststroke Rehabilitation Outcomes Using Wearable Sensors. Phys Ther 2024; 104:pzad183. [PMID: 38169444 PMCID: PMC10851859 DOI: 10.1093/ptj/pzad183] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/15/2023] [Revised: 11/13/2023] [Accepted: 12/03/2023] [Indexed: 01/05/2024]
Abstract
OBJECTIVE Inpatient rehabilitation represents a critical setting for stroke treatment, providing intensive, targeted therapy and task-specific practice to minimize a patient's functional deficits and facilitate their reintegration into the community. However, impairment and recovery vary greatly after stroke, making it difficult to predict a patient's future outcomes or response to treatment. In this study, the authors examined the value of early-stage wearable sensor data to predict 3 functional outcomes (ambulation, independence, and risk of falling) at rehabilitation discharge. METHODS Fifty-five individuals undergoing inpatient stroke rehabilitation participated in this study. Supervised machine learning classifiers were retrospectively trained to predict discharge outcomes using data collected at hospital admission, including patient information, functional assessment scores, and inertial sensor data from the lower limbs during gait and/or balance tasks. Model performance was compared across different data combinations and was benchmarked against a traditional model trained without sensor data. RESULTS For patients who were ambulatory at admission, sensor data improved the predictions of ambulation and risk of falling (with weighted F1 scores increasing by 19.6% and 23.4%, respectively) and maintained similar performance for predictions of independence, compared to a benchmark model without sensor data. The best-performing sensor-based models predicted discharge ambulation (community vs household), independence (high vs low), and risk of falling (normal vs high) with accuracies of 84.4%, 68.8%, and 65.9%, respectively. Most misclassifications occurred with admission or discharge scores near the classification boundary. For patients who were nonambulatory at admission, sensor data recorded during simple balance tasks did not offer predictive value over the benchmark models. CONCLUSION These findings support the continued investigation of wearable sensors as an accessible, easy-to-use tool to predict the functional recovery after stroke. IMPACT Accurate, early prediction of poststroke rehabilitation outcomes from wearable sensors would improve our ability to deliver personalized, effective care and discharge planning in the inpatient setting and beyond.
Collapse
Affiliation(s)
- Megan K O’Brien
- Max Nader Lab for Rehabilitation Technologies and Outcomes Research, Shirley Ryan AbilityLab, Chicago, Illinois, USA
- Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, Illinois, USA
| | - Francesco Lanotte
- Max Nader Lab for Rehabilitation Technologies and Outcomes Research, Shirley Ryan AbilityLab, Chicago, Illinois, USA
- Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, Illinois, USA
| | - Rushmin Khazanchi
- Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA
| | - Sung Yul Shin
- Max Nader Lab for Rehabilitation Technologies and Outcomes Research, Shirley Ryan AbilityLab, Chicago, Illinois, USA
- Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, Illinois, USA
| | - Richard L Lieber
- Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, Illinois, USA
- Department of Biomedical Engineering, Northwestern University, Evanston, Illinois, USA
- Shirley Ryan AbilityLab, Chicago, Illinois, USA
| | - Roozbeh Ghaffari
- Department of Biomedical Engineering, Northwestern University, Evanston, Illinois, USA
- Querrey Simpson Institute for Bioelectronics, Northwestern University, Evanston, Illinois, USA
| | - John A Rogers
- Department of Biomedical Engineering, Northwestern University, Evanston, Illinois, USA
- Querrey Simpson Institute for Bioelectronics, Northwestern University, Evanston, Illinois, USA
- Departments of Materials Science and Engineering, Chemistry, Mechanical Engineering, Electrical Engineering and Computer Science, Northwestern University, Evanston, Illinois, USA
- Department of Neurological Surgery, Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA
| | - Arun Jayaraman
- Max Nader Lab for Rehabilitation Technologies and Outcomes Research, Shirley Ryan AbilityLab, Chicago, Illinois, USA
- Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, Illinois, USA
| |
Collapse
|
2
|
Song J, Hardin EC. Monitoring walking asymmetries and endpoint control in persons living with chronic stroke: Implications for remote diagnosis and telerehabilitation. Digit Health 2024; 10:20552076231220450. [PMID: 38188863 PMCID: PMC10768577 DOI: 10.1177/20552076231220450] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/17/2023] [Accepted: 11/23/2023] [Indexed: 01/09/2024] Open
Abstract
Objective The objective of this study was to assess the feasibility of monitoring and diagnosing compromised walking motion in the frontal plane, particularly in persons living with the chronic effects of stroke (PwCS). The study aimed to determine whether active control of walking in the frontal plane could be monitored and provide diagnostic insights into compensations made by PwCS during community living. Methods The study recruited PwCS with noticeable walking asymmetries and employed a monitoring method to assess frontal plane motion. Monitoring was conducted both within a single assessment and between assessments. The study aimed to uncover baseline data and diagnostic information about active control in chronic stroke survivors. Data were collected using sensors during 6 minutes of walking and compared between the paretic and non-paretic legs. Results The study demonstrated the feasibility of monitoring frontal plane motion and diagnosing disturbed endpoint control (p < 0.0125) in chronic stroke survivors when comparing the paretic leg to the non-paretic leg. A greater variability was observed in the paretic leg (p < 0.0125), and sensors were able to diagnose a stronger coupling of the body with its endpoint on the paretic side (p < 0.0125). Similar results were obtained when monitoring was conducted over a six-minute walking period, and no significant diagnostic differences were found between the two monitoring assessments. Monitoring did not reveal performance fatigue or debilitation over time. Conclusions This study's findings indicate that monitoring frontal plane motion is a feasible approach for diagnosing compromised walking motion. The results suggest that individuals with walking asymmetries, exhibit differences in endpoint control and variability between their paretic and non-paretic legs. These insights could contribute to more effective rehabilitation strategies and highlight the potential for monitoring compensations during various activities of daily living.
Collapse
Affiliation(s)
- Jiafeng Song
- Department of Biomedical Engineering, Georgia Institute of Technology, Atlanta, GA, USA
- Department of Biomedical Informatics, Emory University, Atlanta, GA, USA
| | - Elizabeth C Hardin
- Human Performance Virtual Reality Lab, Cleveland FES Center, Cleveland VA Medical Center, Cleveland, OH, USA
| |
Collapse
|
3
|
Lanotte F, O’Brien MK, Jayaraman A. AI in Rehabilitation Medicine: Opportunities and Challenges. Ann Rehabil Med 2023; 47:444-458. [PMID: 38093518 PMCID: PMC10767220 DOI: 10.5535/arm.23131] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/18/2023] [Accepted: 11/23/2023] [Indexed: 01/03/2024] Open
Abstract
Artificial intelligence (AI) tools are increasingly able to learn from larger and more complex data, thus allowing clinicians and scientists to gain new insights from the information they collect about their patients every day. In rehabilitation medicine, AI can be used to find patterns in huge amounts of healthcare data. These patterns can then be leveraged at the individual level, to design personalized care strategies and interventions to optimize each patient's outcomes. However, building effective AI tools requires many careful considerations about how we collect and handle data, how we train the models, and how we interpret results. In this perspective, we discuss some of the current opportunities and challenges for AI in rehabilitation. We first review recent trends in AI for the screening, diagnosis, treatment, and continuous monitoring of disease or injury, with a special focus on the different types of healthcare data used for these applications. We then examine potential barriers to designing and integrating AI into the clinical workflow, and we propose an end-to-end framework to address these barriers and guide the development of effective AI for rehabilitation. Finally, we present ideas for future work to pave the way for AI implementation in real-world rehabilitation practices.
Collapse
Affiliation(s)
- Francesco Lanotte
- Max Nader Lab for Rehabilitation Technologies and Outcomes Research, Shirley Ryan AbilityLab, Chicago, IL, United States
- Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, IL, United States
| | - Megan K. O’Brien
- Max Nader Lab for Rehabilitation Technologies and Outcomes Research, Shirley Ryan AbilityLab, Chicago, IL, United States
- Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, IL, United States
| | - Arun Jayaraman
- Max Nader Lab for Rehabilitation Technologies and Outcomes Research, Shirley Ryan AbilityLab, Chicago, IL, United States
- Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, IL, United States
| |
Collapse
|
4
|
Yamamoto M, Shimatani K, Yoshikawa D, Washida T, Takemura H. Perturbation-Based Balance Exercise Using a Wearable Device to Improve Reactive Postural Control. IEEE J Transl Eng Health Med 2023; 11:515-522. [PMID: 38059063 PMCID: PMC10697292 DOI: 10.1109/jtehm.2023.3310503] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 02/08/2023] [Revised: 04/22/2023] [Accepted: 08/22/2023] [Indexed: 12/08/2023]
Abstract
Reactive postural control is an important component of the balance function for fall prevention. Perturbation-based balance exercises improve reactive postural control; however, these exercises require large, complex instruments and expert medical guidance. This study investigates the effects of unexpected perturbation-based balance exercises using a wearable balance exercise device (WBED) on reactive postural control. Eighteen healthy adult males participated in this study. Participants were assigned to the WBED and Sham groups. In the intervention session, participants in the WBED group randomly underwent unexpected perturbation in the mediolateral direction, while the Sham group performed the same exercises without perturbation. Before and after the intervention session, all participants underwent evaluation of reactive balance function using air cylinders. Peak displacement (D), time at peak displacement (T), peak velocity (V), and root mean square (RMS) of center of pressure (COP) data were measured. For mediolateral and anteroposterior COP (COPML and COP[Formula: see text]), the main effects of group and time factors (pre/post) were investigated through the analysis of variance for split-plot factorial design. In the WBED group, the D-COPML and V-COPML of the post-test significantly decreased compared to those of the pre-test (p = 0.017 and p = 0.003, respectively). Furthermore, the D-COPAP and RMSAP of the post-test significantly decreased compared to those of the pre-test (p = 0.036 and p = 0.015, respectively). This study proved that the perturbation-based balance exercise using WBED immediately improved reactive postural control. Therefore, wearable exercise devices, such as WBED, may contribute to the prevention of falls and fall-related injuries.
Collapse
Affiliation(s)
- Masataka Yamamoto
- Faculty of Science and TechnologyTokyo University of ScienceNodaChiba278-8510Japan
- Graduate School of Advanced Science and EngineeringHiroshima UniversityHigashihiroshima739-8527Japan
- Department of RehabilitationFukuyama Memorial HospitalFukuyama721-0964Japan
| | - Koji Shimatani
- Faculty of Health and WelfarePrefectural University of HiroshimaMiharaHiroshima723-0053Japan
| | - Daiki Yoshikawa
- Faculty of Health and WelfarePrefectural University of HiroshimaMiharaHiroshima723-0053Japan
| | - Taku Washida
- Faculty of Science and TechnologyTokyo University of ScienceNodaChiba278-8510Japan
| | - Hiroshi Takemura
- Faculty of Science and TechnologyTokyo University of ScienceNodaChiba278-8510Japan
| |
Collapse
|