A virtual reality simulator for training the surgical reduction of patient-specific supracondylar humerus fractures.

PURPOSE: Virtual reality has been used as a training platform in medicine, allowing the repetition of a situation/scenario as many times as needed and making it patient-specific prior to an operation. Of special interest is the minimally invasive plate osteosynthesis (MIPO). It represents a novel technique for orthopedic trauma surgery, but requires intensive training to acquire the required skills. In this paper, we propose a virtual reality platform for training the surgical reduction of supracondylar fractures of the humerus using MIPO. The system presents a detailed surgical theater where the surgeon has to place the bone fragments properly. METHODS: Seven experienced users were selected to perform a surgical reduction using our proposal. Two paired humeri were scanned from a dataset obtained from the Complejo Hospitalario de Jaén. A virtual fracture was performed in one side of the pair, using the other as contralateral part. Users have to simulate a reduction for each case and fill out a survey about usability, using a five-option Likert scale. RESULTS: The subjects have obtained excellent scores in both simulations. The users have notably reduced the time employed in the second experiment, being 60% less in average. Subjects have valued the usability (5.0), the intuitiveness (4.6), comfort (4.5), and realism (4.9) in a 1-5 Likert scale. The mean score of the usability survey was 4.66. CONCLUSION: The system has shown a high learning rate, and it is expected that the trainees will reach an expert level after additional runs. By focusing on the movement of bone fragments, specialists acquire motor skills to avoid the malrotation of MIPO-treated fractures. A future study can fulfill the requirements needed to include this training system into the protocol of real surgeries. Therefore, we expect the system to increase the confidence of the trainees as well as to improve their decision making.

Inter-observer variability assessment of a left ventricle segmentation tool applied to 4D MDCT images of the heart.

Multiple detector row computed tomography (MDCT) cardiac angiography provides a large amount of data concerning multiple cardiac phases which are not often considered. Segmentation is a first step towards exploring how this additional data can be used to perform left ventricle functional analysis or myocardial perfusion assessment. We present preliminary results regarding the assessment of inter-observer variability for a semi-automatic (multi-phase) segmentation tool for the left-ventricle.