"The future of surgery is not about blood and guts; the future of surgery is about bits and bytes.”
/Dr. Richard Satava/

Tuesday, July 29, 2008

GE Healthcare

Recently, I visited the General Electric (GE) Healthcare division’s site in Waukesha (WI), there main profile being imaging. Even though GE has not been involved with interventional technologies, there is a recent shift towards intra-operative navigation and image guided surgery. (Interventional radiology has long been on their pallet.) Out of the six main US sites where they deal with medical imaging and other technologies, Waukesha is the and biggest. Most of their focus is on the big imaging machines of GE, including several product lines of CTs, MRs, PETs and X-rays.
The first floor of the main (CT) building is taken up by the assembly halls, the test facilities and the showcase for the visitors. The second floor is more for sales, marketing and other product related activities, while an entire floor is dedicated to engineering R&D.
One of my friends works here now, so first she gave me a tour, than I got an insight guided visit from her team leader. I was introduced to the latest imaging machines, just about to enter the market. The VCT HD Lightspeed dual scanner can perform scans with two different radiation levels, saving the patient from significant dose exposure. This is the fastest and highest resolution device available (0.23 mm slice thickness); with 64 detector rows it can scan 40 mm tissue at a time, making it capable of doing e.g. brain perfusion imaging or dynamic heart imaging. The new PET/CT will hit the market next year, dedicated to tumor imaging combined with physiological analysis. It features high-sensitivity crystals and Vue Point HD high-definition image processing techniques.
The manufacturing hall has three parallel assembly lines, and although the machines are constructed manually, they have the capacity of 1100 unit / year. Further assembly sites are in Japan and China. There are more than 70 service and development bays attached to the hall, each with a shielded room to contain an imaging device and additional controller. Besides testing and servicing, developers can also get to work on their machines here. The facilities include clean rooms and a separate detector research area as well.
Many of the R&D projects are out of site, and GE often acquire smaller companies with promising technologies. In Waukesha, they focus on the software side, both researching advanced algorithms and implementing additional features to their main support software, the AW. The AW consists of many modules specifically optimized and tuned to certain procedures (e.g. to assist with coronary artery imaging), and the costumers can buy each module separately. Its segmenting capabilities are very impressive and believed to be superior to the competitors’.

Sunday, July 20, 2008

Surgery for engineers

I was given the unique chance to take the Surgery for Engineers course this summer at Hopkins. As the official brochure says, it gives “orientation to undergraduate and graduate engineering students to basics of surgery to enable them to more effectively engage in engineering problem solving for current and future technologies. This course will not make engineers into surgeons, but will orient engineers to the operating room environment, sterile technique, equipment, laparoscopic, robotic, and 'open' instruments, basic surgical exposure and skills, and region specific fundamentals for specific targeted organ systems, as the liver, lungs, GI tract, pancreas, adrenal glands, kidneys, etc. “
This summer training consists of 6 4-hour-long classes in the medical school (in the Center for Minimally Invasive Technologies within) with joint practical sessions. The first day was more introductory, presenting the facilities and history of the Hopkins Hospital. It was founded in 1889, and by today it’s one of the biggest hospitals in the state. (They add a new block every year, and handled 400.000 inpatient days in 2007.) They have been selected for the No.1. hospital by the U.S.News & World Report journal in the past 17 years consecutively. They have received 14 Nobel prices in total. The second occasion was about suturing. We learnt about cutlery, electric knives and their use, and then how to tie a square knot, a surgeon knot, one and two handed. Later we had the opportunity to practice on pig legs. First we had to preparate some vessels or other structures, and then do the tying.
The consecutive class was about the abdomen. Basic anatomy, placement of incision, operating approaches and the preparation of gastro-intestinal (GI) anasthamoses. We were introduced to the linear and circular staplers that helps to faster stitch the tissues. In the OR, we had a pig in anesthesia to practice on. We had to open, fix and examine the abdomen, look for the different tissues, identify organs. Then we preformed a few side-to-side anatsthamoses on its small intestine, and later a gastric bypass. Another group was ultra-sounding the arteria carotis in the mean time. Finally, we closed the abdomen to practice the skills learnt the previous day.

Monday, July 14, 2008

DTI tractography

Even though it is not closely connected to CIS, I was fascinated by a relatively new brain visualization technique, the Diffusion Tensor Imaging (DTI). Google can show many of the incredible images created by analyzing and clustering the neurons. There is a free software, the MedINRIA that can help creating these images, and open source platform 3D Slicer has also got a recent tractography plugin. (Their picture on the right shows the lateral connections of the brain.)
Wikipedia describes briefly the technology behind it: "Tractography is performed utilizing Diffusion Weighted Imaging, an MR technique which is sensitive to the diffusion of water in the body, and can be used to reveal its 3D shape. Free diffusion occurs equally in all directions. If the water diffuses in a medium having barriers, the diffusion will be uneven. In such a case, the relative-mobility of the molecules from the origin has a shape different from a sphere. This shape is often modeled as an ellipsoid, and the technique is then called Diffusion Tensor Imaging. Barriers can be many things --cell membranes, axons, myelin, etc; but in white matter the principal barrier is the myelin sheath of axons. Bundles of axons provide a barrier to perpendicular diffusion and a path for parallel diffusion along the orientation of the fibers. This is termed "anisotropic" diffusion. Anisotropic diffusion is expected to be increased in areas of high mature axonal order. " So the water diffusion gives an idea of approximate fiber orientation. By data analysis techniques we can track the principal diffusion direction and estimate fiber trajectories in white matter. By automatically grouping fibers into structures, we can visualize physiological structures, highly interconnected and functionally integrated regions in the brain.

Thursday, July 10, 2008

MR compatible robotics

Recently I had a chance to attend a labmate, Greg Fisher's defense for the doctoral title in mechanical engineering. He has done an enormous amount of work here in the past couple of years, taking part in many projects, doing work in mechanical design, US signal processing, robot control, 3D display implementation, system verification and many other issues. His major work was to develop an MR compatible robotic system for prostate biopsy and brachitherapy. He designed built and tested two generations of the robots, and the later one is in clinical trials now. A recent publication on the robot is available here (IEEE Xplore). More information on MR compatible surgical robotics can be found in the latest special issue of the IEEE Engineering in Medicine and Biology magazine (Issue: 3, May/June, 2008) .
For further details, here is the synopsis of Greg's work:

by Gregory Fischer

"Magnetic Resonance Imaging (MRI) can provide high-quality 3D visualization of the target anatomy and surrounding tissue, thus granting potential to be a superior medical imaging modality for guiding and monitoring interventions. However, the benefits can not be readily harnessed for interventional procedures due to difficulties that surround the use of Highfield (1.5T or greater) MRI. The inability to use conventional mechatronics and the confined physical space make it extremely challenging to access the patient.
This work describes the development of two apparently very different systems that represent different approaches to the same surgical problem - coupling information and action to perform percutaneous (through the skin) needle placement with MR imaging. The first system addressed takes MR images and projects them along with a surgical plan directly on the interventional site, thus providing in-situ imaging and is know and the MR Image Overlay. With anatomical images and a corresponding plan visible in the appropriate pose, the clinician can use this information to perform the surgical action. The Image Overlay system, with its ability to help guide in-plane needle insertions without the requirement for real-time visualization, is ideally suited for applications where needle insertion in the MRI suite is beneficial. Development of the system for providing optically stable, in-situ images under MRI is presented here.
My primary research effort has focused on a robotic assistant system that overcomes the difficulties inherent to MR-guided procedures, and promises safe and reliable intra-prostatic needle placement inside closed high-field MRI scanners. The robot is a servo pneumatically operated automatic needle guide, and effectively guides needles under real-time MR imaging. MRI compatibility of the robot has been evaluated under 3T field strength using standard prostate imaging sequences and average SNR loss is limited to 5%. Needle alignment accuracy of the robot under servo pneumatic control is better than 0.94mm RMS per axis. The complete system workflow has been evaluated in phantom studies with accurate visualization and targeting. The thesis describes development of the robotic system including requirements, workspace analysis, mechanism design and optimization, and evaluation of MR compatibility. Further, a generally applicable MR compatible robot controller is developed, the pneumatic control system is implemented and evaluated, and the system is deployed in pre-clinical trials."