Sign in Subscribe
Cancer

Stroke Rehabilitation: Machine Learning Approach to Personalize Recovery

Strategic Insights

— 2 min read

New research on stroke rehabilitation has highlighted the potential of machine learning algorithms to personalize and improve the recovery process for stroke survivors. According to a study published in the Journal of Engineering, stroke is a significant global public health challenge, ranking as the second leading cause of death after heart disease. The study focused on 45 post-stroke patients who experienced either hemorrhagic or ischemic strokes, categorizing them based on the location of brain damage. Gait analysis was conducted using the GaitRite system, measuring 39 spatio-temporal parameters. The research employed machine learning algorithms, including Random Forest, Support Vector Machine, and k-Nearest Neighbors, to classify the stroke types based on gait variables.

Key Takeaways:

  • The study found that machine learning models, particularly Random Forest, can identify stroke types based on gait variables when traditional tests cannot.
  • Random Forest outperformed other algorithms, demonstrating superior performance in accuracy, precision, recall, and F1 score, with all exceeding 85%.
  • The research suggests that machine learning can be used to personalize rehabilitation programs for stroke survivors, taking into account the specific characteristics of each individual's brain damage and gait patterns.
  • The study highlights the role of sensory input in post-stroke rehabilitation, particularly the importance of basal ganglia lesions in motor control, sensory processing, and postural control.
  • The clinical implication of the study is that rehabilitation programs should take into account the specific gait parameters and cerebral lesion location of each patient to optimize recovery.
  • The research was conducted by a team of experts from the Institute for Cancer Research and Treatment (IRCCS) and published in the Journal of Engineering.

Statistics:

  • 45 post-stroke patients were included in the study, categorized based on the location of brain damage (cortical-subcortical, corona radiata, and basal ganglia).
  • Gait analysis was conducted using the GaitRite system, measuring 39 spatio-temporal parameters.
  • Machine learning algorithms were employed, including Random Forest, Support Vector Machine, and k-Nearest Neighbors.
  • The study reported an accuracy, precision, recall, and F1 score of over 85% for Random Forest.
  • The study found a significant correlation between gait parameters and cerebral lesion location, particularly linking basal ganglia lesions to prolonged double support time.

Sources:

  • NewsRx. Institute for Cancer Research and Treatment (IRCCS) Reports Findings in Stroke (Older people and stroke: a machine learning approach to personalize the rehabilitation of gait). Journal of Engineering. June 23, 2025; p 1030.
  • Older people and stroke: a machine learning approach to personalize the rehabilitation of gait. Frontiers in Aging, 2025;6:1562355.

Read more