How Artificial Intelligence could transform veterinary education
Assessing the suturing skills of veterinary students is crucial to ensure they acquire the necessary competency for surgical procedures. In recent years, researchers have explored the use of innovative approaches, such as artificial intelligence (AI) and electromyographic (EMG) data analysis, to enhance the evaluation process. This article presents a study that aimed to evaluate suturing skills using different performance assessment methods and compare the results with EMG data.
A total of 16 second-year veterinary students participated in the study. They performed four suturing tasks on synthetic tissue models at various time intervals after completing a surgical skills course. The students' performances were recorded using digital videos, which were subsequently assessed by four expert surgeons. Three common performance assessment methods were utilized: 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). Additionally, EMG data was collected from the students' dominant forearm using an armband and analyzed using machine learning algorithms.
The reliability analysis revealed that the OSCE and OSATS checklists exhibited moderate reliability, whereas the GRS demonstrated poor reliability. When applying moderate tolerance, agreement was achieved for the checklists, but remained poor for the GRS. Surprisingly, the correlation between EMG data and assessment scores or task completion times was not strong. However, by extracting features from the EMG signals, it was possible to differentiate between different tasks and accurately classify them using machine learning algorithms.
The findings of this study suggest that the reliability and agreement of the OSCE-style checklist, OSATS checklist, and surgical GRS were inadequate to effectively evaluate suturing skills in veterinary students. The subjective nature of assessments conducted by human raters calls for further investigation into alternative evaluation approaches. The use of kinematics and EMG data analysis presents an intriguing avenue for future research, offering potential solutions to overcome the subjectivity associated with human assessments.
Enhancing the assessment of suturing skills in veterinary education is essential for producing competent and skilled practitioners. This study sheds light on the limitations of traditional assessment methods and emphasizes the need for innovative approaches. While the reliability and agreement of current performance assessment methods were deemed insufficient, the exploration of kinematics and EMG data analysis holds promise for the future development of objective and accurate evaluation tools. Further research is warranted to advance our understanding of these methodologies and their potential application in veterinary education.
Read more by clicking the link below:
https://avmajournals.avma.org/view/journals/ajvr/aop/ajvr.23.03.0058/ajvr.23.03.0058.xml