Dylan Edwards
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Dylan Edwards has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Aaron D. Boes
Department of Psychiatry, University of Iowa , Iowa City, IA 52242
1 shared publication - Amy Kuceyeski
Department of Radiology, Weill Cornell Medicine, New York, NY, USA
1 shared publication - Ceren Tozlu
Department of Radiology, Weill Cornell Medicine, New York, USA
1 shared publication - Charles Yu Liu
Division of Biology and Biological Engineering, California Institute of Technology; Pasadena, CA, USA
1 shared publication - Douglas Labar
Department of Neurology, Weill Cornell Medical College, New York, NY, USA
1 shared publication - Heather Lane
Burke Neurological Institute, White Plains, NY, USA
1 shared publication
Latest Publications
- Open Access
Predicting response to motor therapy in chronic stroke patients using Machine Learning
Accurate predictions of motor improvement resulting from intensive therapy in chronic stroke patients is a difficult task for clinicians, but is key in prescribing appropriate therapeutic strategies. Statistical methods, including machine learning, are a highly promising avenue with which to improve prediction accuracy in clinical practice. The first main objective of this study was to use machine learning methods to predict a chronic stroke individual’s motor function improvement after 6 weeks of intervention using pre-intervention demographic, clinical, neurophysiological and imaging data. The second main objective was to identify which data elements were most important in predicting chronic stroke patients’ impairment after 6 weeks of intervention. Data from one hundred and two patients (Female: 31%, age 61±11 years) who suffered first ischemic stroke 3-12 months prior were included in this study. After enrollment, patients underwent 6 weeks of intensive motor and transcranial magnetic stimulation therapy. Age, gender, handedness, time since stroke, pre-intervention Fugl-Meyer Assessment, stroke lateralization, the difference in motor threshold between the unaffected and affected hemispheres, absence or presence of motor evoked potential in the affected hemisphere and various imaging metrics were used as predictors of post-intervention Fugl-Meyer Assessment. Five machine learning methods, including Elastic-Net, Support Vector Machines, Artificial Neural Networks, Classification and Regression Trees, and Random Forest, were used to predict post-intervention Fugl-Meyer Assessment based on either demographic, clinical and neurophysiological data alone or in combination with the imaging metrics. Cross-validated R-squared and root of mean squared error were used to assess the prediction accuracy and compare the performance of methods. Elastic-Net performed significantly better than the other methods for the model containing pre-intervention Fugl-Meyer Assessment, demographic, clinical and neurophysiological data as predictors of post-intervention Fugl-Meyer Assessment ( ). Pre-intervention Fugl-Meyer Assessment and difference in motor threshold between affected and unaffected hemispheres were commonly found as the strongest two predictors in the clinical model. The difference in motor threshold had greater importance than the absence or presence of motor evoked potential in the affected hemisphere. The various imaging metrics, including lesion overlap with the spinal cord, largely did not improve the model performance. The approach implemented here may enable clinicians to more accurately predict a chronic stroke patient’s individual response to intervention. The predictive models used in this study could assist clinicians in making treatment decisions and improve the accuracy of prognosis in chronic stroke patients.
Open AccessPredicting response to motor therapy in chronic stroke patients using Machine Learning
Accurate predictions of motor improvement resulting from intensive therapy in chronic stroke patients is a difficult task for clinicians, but is key in prescribing appropriate therapeutic strategies. Statistical methods, including machine learning, are a highly promising avenue with which to improve prediction accuracy in clinical practice. The first main objective of this study was to use machine learning methods to predict a chronic stroke individual’s motor function improvement after 6 weeks of intervention using pre-intervention demographic, clinical, neurophysiological and imaging data. The second main objective was to identify which data elements were most important in predicting chronic stroke patients’ impairment after 6 weeks of intervention. Data from one hundred and two patients (Female: 31%, age 61±11 years) who suffered first ischemic stroke 3-12 months prior were included in this study. After enrollment, patients underwent 6 weeks of intensive motor and transcranial magnetic stimulation therapy. Age, gender, handedness, time since stroke, pre-intervention Fugl-Meyer Assessment, stroke lateralization, the difference in motor threshold between the unaffected and affected hemispheres, absence or presence of motor evoked potential in the affected hemisphere and various imaging metrics were used as predictors of post-intervention Fugl-Meyer Assessment. Five machine learning methods, including Elastic-Net, Support Vector Machines, Artificial Neural Networks, Classification and Regression Trees, and Random Forest, were used to predict post-intervention Fugl-Meyer Assessment based on either demographic, clinical and neurophysiological data alone or in combination with the imaging metrics. Cross-validated R-squared and root of mean squared error were used to assess the prediction accuracy and compare the performance of methods. Elastic-Net performed significantly better than the other methods for the model containing pre-intervention Fugl-Meyer Assessment, demographic, clinical and neurophysiological data as predictors of post-intervention Fugl-Meyer Assessment ( ). Pre-intervention Fugl-Meyer Assessment and difference in motor threshold between affected and unaffected hemispheres were commonly found as the strongest two predictors in the clinical model. The difference in motor threshold had greater importance than the absence or presence of motor evoked potential in the affected hemisphere. The various imaging metrics, including lesion overlap with the spinal cord, largely did not improve the model performance. The approach implemented here may enable clinicians to more accurately predict a chronic stroke patient’s individual response to intervention. The predictive models used in this study could assist clinicians in making treatment decisions and improve the accuracy of prognosis in chronic stroke patients.