MedicVR
Explore new possibilities to experience the human body.
MedicVR uses anatomical medical images from magnetic resonance imaging (MRI) or computer tomography (CT) scanners to visualize the resulting volume data in real-time, displaying it in the immersive stereoscopic 3D virtual reality environment.
By means of the accomplishing raycasting algorithm, MedicVR directly renders the volumetric density-isosurface of a given human body scan in the virtual environment with the Virtual Reality headset HTC-VIVE, exposing different anatomical structures and offering possibilities to immersively explore, interact and study the human-body for professional and educational use.
In VR the user is able to select and modify selected surfaces and perform geometrical transformations on the volume through editors and intuitive gesticulative head and controller motion. “Diving” into the human body and three dimensional visual feedback in real-time appeals to the “natural feel” - the immersive feeling in VR.
What we want to achieve is to fill the mental gap between the commonly abstract represented 2D scan image data and their look and feel of rather natural visual features and interaction methods of everyday activity to ease and clarify working with the scanned images.
Students are able to study the structure of the human anatomy directly extracted from a multitude of readily prepared data for educational purposes, effectively boosting the learning process in many ways.
Careful planning for surgical purposes surgeries or therapy with real patient data in a fully three dimensional environment enables the user to prepare in advance for complicated operations.
Because every algorithm is written in bare C++ and OpenGL/GLSL 4.x and no frameworks, other than Valve’s OpenVR for handling the interaction with the headset is used, the application runs at 90Hz for each lens of the Headset, delivering a smooth and comfortable experience in the virtual environment.
Using the parallel capabilities of modern GPU’s and the high resolution of the VR-headset, a detailed visualization is achieved. Currently supported Headsets are the HTC-VIVE. Tested graphic cards were Nvidia GTX Geforce 1080Ti and GTX Geforce 1060.
MedicVR offers a platitude of functionalities to interact with scanned medical datasets. Most functionalities can be used inside as well as outside of the Virtual Environment. The user has access to a classical graphical interface to tweak and work with the renderer outside of the VR and gestures, pointers, editors and three dimensional tools inside of the VR.
Of course, interaction in virtual reality does not only offer a lot of possibilities but also challenges the established comformity of common user-interfacing methods. As such it requires innovative and very different methods in contrast to classical two dimensional interfaces. Because of this we are constantly working on new ways to offer the best experience, immersion and most natural interaction for the user.
One of the most essential features if the transferfunction through which the user is able to visualize and distinguish different structures by assigning color and opacity to different density values. To see multiple different structures at once the clipping-plane offers the possibility to cut through the volume at any desirable angle. Combined with the coronal, axial and sagittal image planes the result is a assisted perspective visualization with the familiar standard views in medical imaging. Transforming the object in fully-immersive three dimensional geometrical space offers new insights, otherwise hidden in classical rendering techniques. For this the haptic controllers should offer the most natural experience to perform such actions.
We at MedicVR are a small team of students under the lead of Prof. Scholl. Belonging to the greater Mobile Autonomous Systems and Cognitive Robotics (MASCOR) institute of the University of Applied Sciences Aachen (FH Aachen), MedicVR grows to be a promising project beneath all the other achieving tasks of the Institute.
