Since the advent of virtual reality (VR), it has been implemented in medical education for surgical training and anatomy education so that the Internet of Medical Things (IoMT) could be further boosted. There have been rare studies on the research trends of the evaluation of endoscopic training through different XR modalities. Position-based dynamics (PBD) has been utilized to optimize the surgical thread simulation, This paper aims to quantitatively evaluate the training performance of each XR modality in general and in terms of the medical fields studied and outcomes measured. Sensors and devices are utilized to form the Internet of Medical Things for healthcare, where the data is uploaded to the cloud and then analyzed as follows before being fed back to the doctor so that he or she can understand his or her level of operation. Through subjective and objective evaluation, the potential promoting effects of vision and touch in module training were discussed.

Abstract Link residual closeness is a newly proposed measure for network vulnerability. In this model, vertices are perfectly reliable and the links fail independently of each other. It measures the vulnerability even when the removal of links does not disconnect the graph. In this paper, we characterize those graphs that maximize the link residual closeness over the connected graphs with fixed order and one additional parameter such as connectivity, edge connectivity, bipartiteness, independence number, matching number, chromatic number, number of cut vertices and number of cut edges.

Virtual reality technology has been widely used in surgical simulators, providing new opportunities for assessing and training surgical skills. Machine learning algorithms are commonly used to analyze and evaluate the performance of participants. However, their interpretability limits the personalization of the training for individual participants. Seventy-nine participants were recruited and divided into three groups based on their skill level in intracranial tumor resection. Data on the use of surgical tools were collected using a surgical simulator. Feature selection was performed using the Minimum Redundancy Maximum Relevance and SVM-RFE algorithms to obtain the final metrics for training the machine learning model. Five machine learning algorithms were trained to predict the skill level, and the support vector machine performed the best, with an accuracy of 92.41% and Area Under Curve value of0.98253. The machine learning model was interpreted using Shapley values to identify the important factors contributing to the skill level of each participant. This study demonstrates the effectiveness of machine learning in differentiating the evaluation and training of virtual reality neurosurgical per- formances. The use of Shapley values enables targeted training by identifying deficiencies in individual skills. This study provides insights into the use of machine learning for personalized training in virtual reality neurosurgery. The interpretability of the machine learning models enables the development of individualized training programs. In addition, this study highlighted the potential of explanatory models in training external skills.

The advancement of virtual reality technology has ushered in new developments in the medical field. The use of virtual surgery training simulators alleviates the paucity of training resources and high training expenses associated with traditional surgical capabilities. Regardless of the type of schooling, doctors must continue to educate themselves. The postoperative evaluation mechanism is incomplete. Traditional objective evaluation indicators are unable to meet surgeons' stringent expectations. The electroencephalograph (EEG) rhythm index is proposed in this article as a new tool for evaluating and distinguishing between novice and expert doctors. The experiment uses a cutting training module from neurosurgery training and compares it with established assessment metrics to determine the correct rate of classification of new evaluation metrics, classifying testers by both metrics and finding a 20% increase in correctness. Additionally, this article compares the energy topographic maps of different EEG rhythms of novices and experts. For classification, two-machine learning algorithms, SVM and random forest, are utilized at the same time. The findings reveal that the accuracy of distinguishing indicators based on EEG cycles is 10% higher than that of typical objective evaluation indicators, regardless of the categorization method. ROC curve analysis was also used to compare the two classification models. The AUC value for the EEG rhythm evaluation index model was 0.971, whereas the AUC value for the classic objective evaluation index model was 0.761, which explains the EEG rhythm evaluation index. The model demonstrates a categorization standard that is reliable.