Low-cost Medical In-situ Visualization

medicalARMedical in-situ visualization deals with the display of patient specific imaging data at the location where they actually are. The design of medical augmented reality applications involves significant challenges, including a fast response time and a high registration and calibration accuracy.

In this work we investigate how widely available and low-priced 3D point cloud sensors can be effective in meeting these challenges by simplifying the most computationally expensive and time consuming tasks. To achieve this aim, we have developed a proof-of-concept tool for medical in-situ visualization, which shows how the use of this new generation of sensors can ease the augmentation of live-video streams with patient specific imaging data.

The preliminary results, while confirming the feasibility of the approach, also show that the registration error is not at a satisfactory level and the whole process is still too time-consuming to execute it in real-time on commodity hardware. Whereas the registration process can be further refined by introducing non-linear optimization methods, highly parallel implementations of markerless spatial registration algorithms on the graphics processing units are still needed to speed up the full processing pipeline.


  • A. P. Placitelli and L. Gallo, "3D Point Cloud Sensors for Low-cost Medical In-situ Visualization," in Proceedings of the 2011 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM ’11, IEEE, 2011.
  • A. P. Placitelli and L. Gallo, "Low-Cost Augmented Reality Systems via 3D Point Cloud Sensors," in Proceedings of the 7th International Conference on Signal Image Technology & Internet Based Systems, SITIS ’11, IEEE Computer Society, 2011.