A qualitative and quantitative assessment for a bone marrow harvest simulator

Area of Science:

• Medical Simulation
• Virtual Reality Training
• Machine Learning in Healthcare

Background:

• Virtual reality (VR) simulators offer immersive training environments.
• Current online assessment methods in VR often struggle with specific cases involving mixed data types.
• Accurate and computationally efficient assessment is crucial for effective VR training.

Purpose of the Study:

• To develop an improved online assessment methodology for VR training simulators.
• To address limitations in handling both quantitative and qualitative data in existing assessment approaches.
• To enhance the accuracy and applicability of VR training assessment in specialized medical procedures.

Main Methods:

• Implementation of a Modified Naive Bayes classifier for simultaneous processing of quantitative and qualitative variables.
• Integration of the developed method into a virtual reality bone marrow harvest simulator.
• Evaluation of the assessment approach using simulated medical procedure data.

Main Results:

• The Modified Naive Bayes approach demonstrated satisfactory performance in the simulated medical scenario.
• The method successfully manipulated both quantitative and qualitative data for assessment.
• Results indicated the practical applicability of the proposed assessment technique.

Conclusions:

• The Modified Naive Bayes approach provides an effective solution for online training assessment in VR, particularly for complex medical procedures.
• This method enhances VR training by accurately assessing performance using diverse data types.
• Further application of this technique can improve medical training outcomes and patient safety.