-
Categories
-
Pharmaceutical Intermediates
-
Active Pharmaceutical Ingredients
-
Food Additives
- Industrial Coatings
- Agrochemicals
- Dyes and Pigments
- Surfactant
- Flavors and Fragrances
- Chemical Reagents
- Catalyst and Auxiliary
- Natural Products
- Inorganic Chemistry
-
Organic Chemistry
-
Biochemical Engineering
- Analytical Chemistry
-
Cosmetic Ingredient
- Water Treatment Chemical
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
Stroke is a neurological deficit caused by acute focal injury of the central nervous system caused by vascular causes, including cerebral infarction and cerebral hemorrhage, which is one of the leading causes of
disability and death worldwide.
Upper limb motor dysfunction and wrist dysfunction caused by stroke seriously affect the normal work and life
of stroke patients.
At present, the distal upper limb - wrist motor function has received less attention in the research field of rehabilitation training and evaluation, and the effective intervention and rehabilitation research of wrist motor function of stroke patients is of great significance
for the improvement of their quality of life and the maximum recovery of the overall function of the upper limb after stroke.
Recently, Xiong Daxi's team from Suzhou Medical Institute proposed a quantitative assessment system for wrist motor function of stroke patients based on force feedback equipment and machine learning algorithms, and its overall workflow is shown in
Figure 1.
Figure 1 Workflow of wrist function quantitative evaluation system based on force feedback
In view of the problems that the motor function assessment of stroke patients cannot be objective and quantitative, based on tactile force feedback, virtual reality visual feedback and artificial intelligence machine learning algorithms, the team proposed a new quantitative evaluation method
for wrist motor function of stroke patients 。 Firstly, stroke patients were manipulated to collect motion and kinetic data by manipulating force feedback equipment, and then the collected data were filtered and noise reduction and other data preprocessing and kinematic feature extraction, and four supervised machine learning algorithms were selected as regression algorithms to establish quantitative evaluation models, and the prediction scores of the evaluation models were compared with the scale scores of clinicians, so as to obtain the quantitative evaluation model
with the highest accuracy.
To verify the effectiveness of the system, the team recruited 25 stroke patients to participate in a clinical trial
assessing wrist motor function.
The real scene of the subject participating in the trial is shown in Figure 2, Figure 2 (d shows a comparison between
the real-time trajectory plotted by the subject's movement during the trial and the curve trajectory given by the system.
At the same time, during the trial, an experienced clinician was invited to score and record
the patient's wrist motor function using a clinical assessment scale.
Fig.
2 Comparison of test scene and trajectory of wrist motor function evaluation
The results show that among the four quantitative evaluation models based on machine learning algorithms, the BPNN (Backpropagation Neural Network) neural network model has the best evaluation effect and the highest accuracy rate is 94.
26%.
The fitted curve of its prediction score basically coincides with the scatter curve of the scale score, while the slightly inferior evaluation model based on support vector machine regression has a large
gap between the prediction results of the low score (0-10) range and the actual scale score.
Figure 3 shows a comparison between
the prediction scores of four machine learning algorithm-based evaluation models and the doctor's scale scores.
Fig.
3 Comparison of the prediction scores of the four evaluation models with the doctor's scale scores
Through further Pearson correlation test, it can be found that there is a significant correlation between the prediction scores of the four models and the scale scores of rehabilitation physicians, which also indicates that the quantitative evaluation system proposed in this paper has high clinical application prospects and can effectively evaluate the wrist motor function
of stroke patients.
The results were published as "Quantitative Evaluation System of Wrist Motor Function for Stroke Patients Based on Force Feedback.
" Sensors (IF: 3.
576), doctoral student Kangjia Ding is the first author
.
Links to papers: https://doi.
org/10.
3390/s22093368
