Although there are some studies on the automatic evaluation of impairment levels after stroke using machine learning (ML) models, few have delved into the predictive capabilities of raw motion data. In this study, we captured kinematic trajectories of the trunk and affected upper limb from 21 patients with chronic stroke when performing three reaching tasks. Employing ML models, we integrated the recorded trajectories to predict scores of the Fugl-Meyer Assessment of the Upper Extremity (FMA-UE) of stroke patients. A transformer-based model achieved better metrics than Residual Neural Network (ResNet) and support vector regression (SVR). The trajectory successfully predicted FMA-UE scores, with the forward task (R=0.905±0.028) outperforming the vertical task (R=0.875±0.019) and horizontal task (R=0.868±0.031). This pilot study demonstrated the capability of original trajectory data in tracking personal motor function after stroke and extended possibility of application in telerehabilitation.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/EMBC53108.2024.10781580 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Reversed loneliness development after age 65 for men and women: Modeling of the age trajectories of loneliness using national cohorts in Taiwan and Japan.
Geriatr Gerontol Int
March 2025
Department of History, College of Liberal Arts, National Cheng Kung University, Tainan, Taiwan.
Aim: This study depicts the age trajectories of loneliness and gender differences among older adults in Taiwan and Japan.
Methods: Two nationally representative data sets for older adults in Taiwan and Japan were obtained from the Taiwan Longitudinal Study on Aging (TLSA, 1996-2011) and the National Survey of the Japanese Elderly (NSJE, 1996-2012), respectively. The analytic sample included 3037 and 1974 older adults aged 65 and over at baseline in Taiwan and Japan, respectively.
Cumulative life course impairment: Evidence for hidradenitis suppurativa.
J Eur Acad Dermatol Venereol
March 2025
Clinical Laboratory for Epidemiology and Applied Research in Skin (CLEARS), Department of Dermatology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, USA.
Hidradenitis suppurativa (HS) adversely affects quality of life, education, work, relationships and mental health. The debilitating effects of HS can compound over a patient's lifetime and have lasting repercussions. The cumulative life course impairment (CLCI) model analyses the disease factors that could affect the life course trajectory of a patient, including effects on major life decisions and opportunities, such as relationships, career path, education and starting a family.
View Article and Find Full Text PDFElastographic magnetization prepared imaging with rapid encoding.
Magn Reson Med
March 2025
Department of Biomedical Engineering, University of Delaware, Newark, Delaware, USA.
Purpose: To introduce a novel sequence for achieving fast, whole-brain MR elastography data through the introduction of a magnetization preparation block for motion encoding along with rapid imaging readouts.
Theory And Methods: We implemented MRE motion encoding in a magnetization preparation pulse sequence block, where spins are excited, motion encoded, and then stored longitudinally. This magnetization is accessed through a train of rapid gradient echoes and encoded with a 3D stack-of-spirals trajectory.
Artificial intelligence-driven translational medicine: a machine learning framework for predicting disease outcomes and optimizing patient-centric care.
J Transl Med
March 2025
Unit for Data Science and Computing, North-West University, 11 Hofman Street, Potchefstroom, 2520, South Africa.
Article Synopsis
- Advancements in AI and ML are transforming the medical field, enhancing patient care and disease modeling, but challenges like data variability and class imbalance hinder optimal predictive performance.
- A new AI framework combining Gradient Boosting Machines and Deep Neural Networks was tested on two datasets, showing better results in accuracy metrics compared to traditional models, including achieving an AUROC of 0.96 on the UK Biobank dataset.
- The framework not only demonstrated superior accuracy but also trained quickly, making it well-suited for real-time clinical applications, with future enhancements aimed at improving scalability and interpretability for broader use.
Assessing the impact of insulin resistance trajectories on cardiovascular disease risk using longitudinal targeted maximum likelihood estimation.
Cardiovasc Diabetol
March 2025
School of Medical Technology and Information Engineering, Zhejiang Chinese Medical University, Hangzhou, China.
Background: Cardiovascular disease (CVD) is closely associated with Insulin Resistance (IR). However, there is limited research on the relationship between trajectories of IR and CVD incidence, considering both time-invariant and time-varying confounders. We employed advanced causal inference methods to evaluate the longitudinal impact of IR trajectories on CVD risk.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!