Reliability in performance assessment creates a potential application of artificial intelligence in veterinary education: evaluation of suturing skills at a single institution.

OBJECTIVES: To evaluate suturing skills of veterinary students using 3 common performance assessments (PAs) and to compare findings to data obtained by an electromyographic armband. SAMPLE: 16 second-year veterinary students. PROCEDURES: Students performed 4 suturing tasks on synthetic tissue models 1, 3, and 5 weeks after a surgical skills course. Digital videos were scored by 4 expert surgeons using 3 PAs (an Objective Structured Clinical Examination [OSCE]- style surgical binary checklist, an Objective Structured Assessment of Technical Skill [OSATS] checklist, and a surgical Global Rating Scale [GRS]). Surface electromyography (sEMG) data collected from the dominant forearm were input to machine learning algorithms. Performance assessment scores were compared between experts and correlated to task completion times and sEMG data. Inter-rater reliability was calculated using the intraclass correlation coefficient (ICC). Inter-rater agreement was calculated using percent agreement with varying levels of tolerance. RESULTS: Reliability was moderate for the OSCE and OSATS checklists and poor for the GRS. Agreement was achieved for the checklists when moderate tolerance was applied but remained poor for the GRS. sEMG signals did not correlate well with checklist scores or task times, but features extracted from signals permitted task differentiation by routine statistical comparison and correct task classification using machine learning algorithms. CLINICAL RELEVANCE: Reliability and agreement of an OSCE-style checklist, OSATS checklist, and surgical GRS assessment were insufficient to characterize suturing skills of veterinary students. To avoid subjectivity associated with PA by raters, further study of kinematics and EMG data is warranted in the surgical skills evaluation of veterinary students.

Classification of Activities of Daily Living Based on Grasp Dynamics Obtained from a Leap Motion Controller.

Stroke is one of the leading causes of mortality and disability worldwide. Several evaluation methods have been used to assess the effects of stroke on the performance of activities of daily living (ADL). However, these methods are qualitative. A first step toward developing a quantitative evaluation method is to classify different ADL tasks based on the hand grasp. In this paper, a dataset is presented that includes data collected by a leap motion controller on the hand grasps of healthy adults performing eight common ADL tasks. Then, a set of features with time and frequency domains is combined with two well-known classifiers, i.e., the support vector machine and convolutional neural network, to classify the tasks, and a classification accuracy of over 99% is achieved.

Hand gesture recognition using surface electromyography.

Surface electromyography has become one of the popular methods for recognizing hand gestures. In this paper, the performance of four classification methods on sEMG signals have been investigated. These methods are developed by combinations of two feature extraction methods, including Mean Absolute Value and Short-Time Fourier Transform, and two classifiers, including Support Vector Machine and Convolutional Neural Network. These classification methods achieved an accuracy over 97 % on the NinaPro dataset 1. In addition, a new dataset, which includes the Activities of Daily Living, was proposed and an accuracy over 98 % was obtained by applying the presented classification methods.This methodology can provide the basis for a robust quantitative technique to evaluate hand grasps of stroke patients in performing activities of daily living that in turn can lead to a more efficient rehabilitation regimen.