Vitrea takes advantage of speedy new O2 workstationTwo developments this month indicate that 3-D visualization technologymay finally be ready for routine clinical use. On the hardwareside, Silicon Graphics released its new O2 workstation,
Two developments this month indicate that 3-D visualization technologymay finally be ready for routine clinical use. On the hardwareside, Silicon Graphics released its new O2 workstation, whichpromises high-end 3-D performance at a price point under $10,000,allaying concerns about the price of high-end 3-D technology (seestory, page 3).
SGI's progress on the hardware front has been matched on thesoftware side by Vital Images, a Fairfield, IA, developer of 3-Dimage processing software. The company this week announced therelease of Vitrea, a new 3-D visualization workstation designedto be easier to use than the company's VoxelView 2.5 product (SCAN2/14/96).
VoxelView broke new ground in 3-D due to the speed with whichit conducted volume rendering, rather than surface rendering,when processing images, according to Andrew Weiss, president ofVital Images. Volume rendering allows clinicians to see throughthe surface of an organ to view objects inside, a capability notpossible with surface rendering.
VoxelView requires a high degree of user interaction, however,which may have contributed to the product's limited success outsideluminary institutions that are using 3-D for research applications.To target the clinical market, Vital Images decided to developsoftware that contains customized rendering algorithms adaptedfor the type of study being processed. As a result, cliniciansdon't have to input visualization parameters to create 3-D images,as they do with VoxelView.
"We want to provide the same quality (as VoxelView) withoutrequiring clinicians to program visualization parameters,"Weiss said. "It's so easy to use, you never even touch thekeyboard."
Vitrea is programmed to automatically render an image basedon information contained in the DICOM header of that image, aswell as the image's histogram, which refers to data found withinthe image that can indicate characteristics such as image density.For example, an image identified in the DICOM header as a CT angiogramwould be rendered based on clinical protocols developed by researchsites that have been working with VoxelView.
Users have a number of choices available to display images,including 2-D slices, 3-D views, and 3-D views with orthogonaland multiplanar reformat (MPR) modes. Real-time fly-throughs arealso possible, at a rendering speed that the company claims isorders of magnitude faster than other 3-D visualization workstations.Vitrea renders images at 10 to 20 million voxels/second for full-resolutionvolume rendering and 8 frames/second during interactive volumemanipulation. This enables users to move through volumes interactively,according to Weiss.
In addition to image visualization, Vitrea includes featuresthat make image distribution easier, such as a standardized reportfunction that allows clinicians to create reports for referringphysicians. These reports can be sent to DICOM-compliant printersor can be posted on a World Wide Web server.
Vital Images applied for 510(k) clearance for Vitrea in September(SCAN 9/25/96). The company plans to sell Vitrea as part of anintegrated workstation that includes SGI's O2 platform, and istargeting a price point around $50,000.
That's far less than what Vitrea would have cost if O2 neverexisted, according to Weiss. Without O2, Vital Images would havebeen required to use a more expensive platform.
"O2 has a degree of price performance that is really unparalleled,"Weiss said. "If the O2 didn't exist, we could offer the samefunctionality on the SGI Indigo2 Impact hardware, but we wouldpay at minimum a factor of three as a price penalty for the hardware."
Study Assesses Lung CT-Based AI Models for Predicting Interstitial Lung Abnormality
September 6th 2024A machine-learning-based model demonstrated an 87 percent area under the curve and a 90 percent specificity rate for predicting interstitial lung abnormality on CT scans, according to new research.