Agreement and Reliability Analysis of Machine Learning Scaling and Wireless Monitoring in the Assessment of Acute Proximal Weakness by Experts and Non-Experts: A Feasibility Study
Round 1
Reviewer 1 Report
This study has conducted agreement and reliability analysis using ensemble learning model-based scaling for assessment of proximal weakness obtained from sensors and have demonstrated improved agreement. The study is novel and the potential improvement by incorporating ML-based scaling for stroke management.
Reviewer 2 Report
The authors apply machine learning to assess proximal weakness by experts and non-specialists. The authors train an ensemble training model using signals from sensors attached to the limbs of patients in the neurological intensive care unit. To analyze the correspondence, they investigated the percentage correspondence of proximal MRC estimates and Bland-Altman kinematic characteristics graphs between expert and non-expert scaling.
The authors demonstrated that scaling using machine learning has significantly improved mutual reliability for assessing proximal weakness in clinical assessments. This can improve the informed application of reliable and to optimize medical care.
Moreover, the results shown clear and accurate. My decision is "Accept the paper in present form".
