Deforming a 3D surface mesh while preserving its local detail is useful for editing anatomical atlases or for mesh based segmentation. This contribution introduces new classes for performing hard and soft constraints deformation in a flexible design which allows user to switch easily in between Laplacian discretization operators, area weighing and solvers. The usage of these new
classes is demonstrated on a sphere.


Laplacian Mesh Deformation

The computation of geodesic distances on a triangle mesh has many applications in geometry processing. The fast marching method provides an approximation of the true geodesic distance field. We provide VTK classes to compute geodesics on triangulated surface meshes. This includes classes for computing the geodesic distance field from a set of seeds and to compute the geodesic curve between source and destination point(s) by back-tracking along the gradient of the distance field. The fast marching toolkit (Peyre et. al.) is internally used. A variety of options are exposed to guide front propagation including the ability to specify propagation weights, constrain to a region, specify exclusion regions, and distance based termination criteria. Interpolators that plug into a contour widget, are provided to enable interactive tracing of paths on meshes.

Geodesic Computations on Surfaces

We present our work to develop a telerobotics research platform that provides complete access to all levels of control via open-source custom electronics and software. The electronics employs an FPGA to enable a centralized computation and distributed I/O architecture in which all control computations are implemented in a familiar development environment (Linux PC) and low-latency I/O is performed over an IEEE-1394a (Firewire) bus at speeds up to 400 Mbits/sec. The mechanical components of the system are provided by the Research Kit for the da Vinci System, which consists of the Master Tool Manipulators (MTMs), Patient Side Manipulators (PSMs), and stereo console of the first-generation da Vinci surgical robot. This system is currently installed, or planned for installation, at 11 research institutions, with additional installations likely in the future, thereby creating a research community around a common open-source hardware and software platform.

An Open-Source Hardware and Software Platform for Telesurgical Robotics Research

This paper is about a design policy for operation-room (OR) friendly research system, software and a software toolkit SCCToolKit for this purpose, and a few examples. Examples are image processor for endoscope, OR event detector, vision based registration of ultrasound and endoscopic images, etc. We name the design policy as Small Computations for Clinicals (SCC). A system is SCC savvy if it is 1) single purpose, 2) turn-key system without need of using mouse and keyboard, 3) small form factor PC based, and 4) cost less than $1,000 USD. SCCToolKit is OpenCV extension that could minimize the number of buffer copies from receiving image frames from operating system to render it to display buffer. It is open source under BSD license. We measured the latency of image capture to display of the endoscope image processor was between 0.13 to 0.2 second for 1920x1080 size display with chroma-key image mixing. Considering of the acceptable latency of 0.2 second, the performance was as good as commercially available, custom hardware processor that could cost over 10,000 USD, while our PC based system cost 773 USD.

Small Computings for Clinicals and SCCToolKit, OR friendly Trial Package and Software Development Kit

Hospital Information Systems (HIS) are the primary source of information in clinical routine. In most clinics and medical centers, an electronic health record (EHR) stores relevant patient information which is passed between sub-systems using the ISO standard Health Level 7 (HL7). In Radiology, the Picture Archiving and Communication System (PACS) is used to store and transfer medical images and other image related patient data and reports. 
The paper promotes the development of dedicated patient models based on HIS information. This allows specialized and tailored models for each purpose without the need for a commonly accepted data format.

Steps towards the Integration of Model Guided Therapy Systems into the Healthcare Environment

This document describes an implementation for GPU and CPU of Young and Van Vliet's recursive Gaussian smoothing as an external module for the Insight Toolkit ITK, version 4.* www.itk.org. In the absence of an OpenCL-capable platform, the code will run the CPU implementation as an alternative to the existing Deriche recursive Gaussian smoothing filter in ITK.

GPU and CPU implementation of Young - Van Vliet's Recursive Gaussian Smoothing Filter

Image-guided interventional systems become an essential part of the modern minimally invasive surgical procedures, especially for the cardiac surgery. Transapical aortic valve implantation (TAVI) is a recently developed surgical technique to treat severe aortic valve stenosis in elderly and high-risk patients. The placement of stented aortic valve prosthesis is crucial and typically performed under live 2-D fluoroscopy guidance. To assist the placement of the prosthesis during the surgical procedure, a few image-guided TAVI systems have been developed. This work therefore presents current difficulties and challenges of these guidance systems, with consideration of future prospects.

Perspectives on Image-Guided Transapical Beating Heart Aortic Valve Intervention

In this paper, we introduce a new concept of surgical navigation which processes information interactively between the real and virtual spaces, namely, updating preoperative images using the positional information of surgical tools. Although the organs are deformed by operative procedures during surgery, surgical navigation systems usually do not change the reference images that are taken prior to surgery. It is useful to generate deformed reference images during surgery while it progresses. We develop a skull base surgery navigation system that updates the preoperative images during surgery. To estimate the resected regions, our proposed system utilizes the positional information of the surgical tools that can be tracked by a surgical navigation system. Our proposed system reflects the bone removal on preoperative images by changing the voxel values of the preoperative images using the positional information of the tracked tools. The updated reference images are generated by visualizing the updated preoperative images using a volume rendering method. We evaluated the proposed system on a skull phantom created from CT images by a 3D printer. The experimental results showed that the proposed system updated the reference images in real time based on the surgical tasks including bone removal process. The accuracy of our proposed method was about 1 mm. It is very useful for surgeons to drill into such complex bone structure as the skull base.

Skull Base Surgery Navigation System Based on Updating Preoperative Images Using Positional Information of Surgical Tools

This document describes the Image-based Vascular Analysis Toolkit (IvanTk), a set of cross-platform C++ open source software libraries for the detection, extraction and modeling of vessels in 2D/3D medical images based on the Insight Toolkit. We provide an overview of the libraries and their purpose and describe their functional modules and future directions of the toolkit.

The Image-based Vascular Analysis Toolkit (IvanTk)

In this paper we applied one of our regularly used processing pipelines for fully automated brain tissue segmentation. Brain tissue was segmented in cerebrospinal fluid (CSF), gray matter (GM) and white matter (WM). Our algorithms for skull stripping, tissue segmentation and white matter lesion (WML) detection were slightly adapted and applied to twelve data sets within the MRBrainS13 brain tissue segmentation challenge. Skull stripping is performed using non-rigid registration of 5 atlas masks. Our tissue segmentation is based on an automatically trained kNN-classifier. Training samples were obtained by non-rigid registration of 5 manually labeled scans followed by a pruning step in feature space to remove any residual erroneously sampled tissue voxels. The kNN-classification incorporates voxel intensities from a T1-weighted scan and a FLAIR scan. The white matter lesion detection is based on an automatically determined threshold on the FLAIR scan. The application of the algorithms on the data from the MRBrainS13 Challenge showed that our pipeline produces acceptable segmentations. Average resulting Dice scores were 77.86 (CSF), 81.22 (GM), 87.27 (WM), 93.78 (total parenchyma), and 96.26 (all intracranial structures). Total processing time was about 2 hours per subject.