• AI
  • Molecular Imaging
  • CT
  • X-Ray
  • Ultrasound
  • MRI
  • Facility Management
  • Mammography

Lockheed teams with Rose Health Carefor computer-aided mammography

Article

MedDetect sets sights on telemammography With a new entry into the computer-aided mammography field comehopes to improve breast cancer detection rates with technologyinitially developed for military uses. Two Denver-based companies,Lockheed

MedDetect sets sights on telemammography

With a new entry into the computer-aided mammography field comehopes to improve breast cancer detection rates with technologyinitially developed for military uses. Two Denver-based companies,Lockheed Martin Astronautics and Rose Health Care System, joinedforces to form MedDetect to commercialize an optical processingtechnology for examining x-rays.

MedDetect's technology should be capable of detecting lesionsas small as 1 mm in diameter, according to Dr. Jeffrey Levy, directorof mammography at Rose. That should permit the detection of potentiallycancerous lesions a full year earlier, he said. However, severalyears of development work and Food and Drug Administration trialsare expected before commercial mammography reading services areavailable. Additional issues must be addressed regarding the useof computer-aided diagnosis for mammography applications.

These issues aside, MedDetect's technology is fast enough tobe able to read thousands of mammograms per hour. The technologycould be used in regional reading centers that are linked to perhapsdozens of clinics and hospitals through teleradiology. The efficiencyadvantages conferred by the system could allow these centers tocut in half the current $30 mammogram reading charge, accordingto Dr. E. Michael Henry, manager of advanced processing at LockheedMartin.

Initially, Rose and Lockheed Martin will provide a small amountof seed money for MedDetect, with additional sums expected tobe raised through private capital. This will fund product developmentand extensive clinical testing. Eventually, the technology couldbe applied to chest x-rays, pap smears and other cytological data.

To evaluate a mammogram, radiologists examine up to 42 differentfeatures, any of which can vary in size and orientation. Imageanalysis systems try to duplicate the radiologist's evaluationeither with electronic or optical techniques. But optical methodshave an advantage since they process data essentially in parallel,permitting much more efficient and speedy analysis.

The optical processor developed at Lockheed Martin uses Fouriertransform analysis. To detect features in the mammogram, a digitizedimage is first displayed on a device called a spatial light modulator(SLM), which is a high-resolution liquid crystal display. A laserilluminates the SLM to produce a coherent version of the mammogram,which passes through a set of lenses that transform it into aFourier image. In the Fourier domain, light diffraction causesthe input image to be represented as a new image pattern thatcorresponds to the size and shape (spatial frequencies) of eachfeature. Consequently, the smallest features are represented atthe edges of the lens, while larger features are near the center.

To detect 1-mm features in the mammogram, for example, a secondSLM is used to construct a light-passing ring. By properly settingthe radius and width of this ring, the processor will pass thespatial components of the image that are near 1 mm in breadth.A second set of lenses reconverts the light out of the Fourierdomain and images it on a video camera. The result is an opticalprocessor that can be tuned to pass any spatial frequency in animage. If certain features are present in the input mammogram,they will show up as bright spots in the processed image. Scanninga set of four mammograms can be done in a few seconds, comparedto approximately two minutes usually required by a radiologistto read mammograms.

Suspicious lesions are circled in red, while suspicious areasthat are less serious are indicated in yellow. Online expert radiologistscould review the mammogram should another opinion be requested.

Lockheed Martin originally developed similar technology todetect tanks and other targets in military images. The companylater applied this technique in an experiment to detect very finedefects in the x-rays of welds used to construct fuel tanks forTitan rockets. The inspiration for the formation of MedDetectcame when a radiologist at Rose Health Care System noticed strikingsimilarities between the weld x-rays and mammographic x-rays.

Other companies developing computer-aided diagnosis programsfor mammography include R2 Technology of Mountain View, CA, andStereometrix of San Jose, CA (SCAN 10/20/93 and 12/28/94).

Lockheed Martin's Henry believes that optical processing willvastly improve the speed and confidence with which a radiologistcan detect abnormalities, thus lowering costs. But there are otherbenefits as well. For example, the ability to identify featuresthat are below the detection abilities of the best radiologistsprovides an important new tool for the early detection of theselesions. Secondly, such systems provide an objective and quantitativeanalysis of mammograms.

"There is a lot of controversy right now concerning mammogramsbecause of the wide diversity in the reading of the x-rays,"Henry said. "Optical processors can add a level of confidenceto the reading and help ensure a level of uniformity."

Finally, the emergence of new and sophisticated x-ray readingservices is being helped by the budding of the telemedicine infrastructure.

"In four years, when we are ready to begin service, thetelemedicine infrastructure should be mature enough to make thisa viable option," Henry said.

Related Videos
Improving the Quality of Breast MRI Acquisition and Processing
Can Fiber Optic RealShape (FORS) Technology Provide a Viable Alternative to X-Rays for Aortic Procedures?
Does Initial CCTA Provide the Best Assessment of Stable Chest Pain?
Making the Case for Intravascular Ultrasound Use in Peripheral Vascular Interventions
Can Diffusion Microstructural Imaging Provide Insights into Long Covid Beyond Conventional MRI?
Assessing the Impact of Radiology Workforce Shortages in Rural Communities
Emerging MRI and PET Research Reveals Link Between Visceral Abdominal Fat and Early Signs of Alzheimer’s Disease
Reimbursement Challenges in Radiology: An Interview with Richard Heller, MD
Nina Kottler, MD, MS
The Executive Order on AI: Promising Development for Radiology or ‘HIPAA for AI’?
Related Content
© 2024 MJH Life Sciences

All rights reserved.