BOSTON, Oct. 27 /PRNewswire-FirstCall/ -- Medicsight (BULLETIN BOARD: MSHT) today announced it has received clearance from the U.S. Food and Drug Administration for Medicsight Colon CAR 1.2, an image analysis software tool designed to be used with CT colonography (virtual colonoscopy) to assist radiologists in searching for and measuring potential colorectal polyps. Virtual colonoscopy uses a CT scanner to generate both two-dimensional and three-dimensional views of the inside the colon, as opposed to traditional colonoscopy, in which a viewing instrument is inserted into the bowel. Colon cancers most often start when malignant cells form within polyps attached to the inner surface of the large bowel. Detection and removal of these polyps can prevent them from becoming cancerous.
Medicsight Colon CAR 1.2 works by using Medicsight's Computer Assisted Reader (CAR) technology to deploy a series of filters against image data derived from CT colonographies. These filters highlight spherical areas of the image as small as 5 mm or the size of a small pea which could be potential polyps. The radiologist is also able to manually highlight any irregularities for closer inspection. Once suspect polyps are found, the software can precisely identify the boundaries and features and show them in 3D with a volume measurement, diameter, shape and location. This allows the radiologist to accurately review and track any growth in the polyps.
"We know that polyps of a certain size will likely become cancerous and should be removed so accurately measuring polyps is key," said Dr. Mary Roddie, consultant radiologist at Hammersmith Hospitals NHS Trust in London and Medicsight's director of radiology. "To accurately calculate the volume or size of a polyp, you first must have accurate boundary identification. The Medicsight software reliably identifies the boundaries and calculates the volume, giving radiologists more dependable information on which to base their medical decisions. In addition, and more importantly, accurate boundary identification is what makes detection of these polyps possible."
Medicsight Colon CAR 1.2 is the first "joint-read" software available for CT colonography. Unlike "second-read" software, in which software is employed after radiologists complete their reviews, joint-read software enables radiologists to review the original, "unfiltered" image side-by-side and simultaneously with a software-enhanced image showing regions of interest. For a radiologist reviewing up to 1,400 images from a single CT colonography, the joint-read capability saves time and can produce optimal results in the evaluation of polyps.
Dr. Perry Pickhardt, the University of Wisconsin Medical School researcher who published a landmark study on virtual colonoscopy in the New England Journal of Medicine in December 2003, said software such as Medicsight's will help make virtual colonoscopy a practical and widely accepted means of screening for colorectal cancer. "With the enormous amount of data generated from a single CT scan, having an effective software tool to aid the radiologist in detection and accurate measurement will be critical," he said.
"Locating polyps hidden within folds in the colon or having to distinguish polyps from surrounding healthy tissue can be challenging for radiologists," said Jamshid Dehmeshki, Chief Technology Officer. "We have developed a software tool that is the most advanced and powerful in boundary identification and measurement of polyps, which in turn will assist radiologists in decision making and diagnosis."
Dr. Steve Halligan, a leading radiological researcher from St. Marks Hospital in the United Kingdom, is currently evaluating the benefits of the new software in a large-scale trial involving 10 radiologists with each interpreting over 200 CT colonographic datasets. Dr. Stuart Taylor, Dr. Halligan's colleague at St. Marks, points to the large differences in reader performance as one of the major challenges facing CT colonography. "There is good evidence that the diagnostic performance of CT colonography is in part dependent on the ability and experience of the reporting radiologist," said Dr. Taylor. "The current trial hypothesizes that the new joint read CAR software will significantly improve the ability of non-expert radiologists to detect colonic polyps, and thereby match the performance of their expert colleagues."
Colorectal cancer is the third most common cancer found in men and women in the United States, according to the American Cancer Society (ACS). The ACS estimates that about 106,370 new cases of colon cancer and 40,570 new cases of rectal cancer will be diagnosed in 2004. Colorectal cancer accounts for about 10 percent of annual cancer deaths. The ACS believes screening reduces the number of cancer deaths and can prevent the development of colon cancer by identifying and removing polyps, from which colon cancers often develop.
In August 2004, Medicsight announced that it received clearance from the U.S. Food and Drug Administration for Medicsight Lung CAR, an image analysis software tool that assists radiologists in evaluating lesions or nodules found during CT scans of the lung.
Medicsight is a software development business focused on the medical imaging market. The company is using its core technology to develop automatic detection and analytical tools for clinicians to improve their ability to diagnose and treat disease. Medicsight has initially focused on three key clinical areas -- lung cancer, colon cancer, and coronary heart disease -- which together account for almost half of all deaths in America. Medicsight's software will enable earlier and more accurate detection and treatment of suspicious lesions, which can save lives and reduce healthcare costs.
Headquartered in London, Medicsight employs more than 60 people and has offices in Nashville, Tenn. and Singapore. Since its inception in 1999, Medicsight has been developing its core algorithms through a team of more than 30 scientists and software developers. For further information, please visit http://www.medicsight.com/.
Medicsight
Medicsight Colon CAR 1.2 works by using Medicsight's Computer Assisted Reader (CAR) technology to deploy a series of filters against image data derived from CT colonographies. These filters highlight spherical areas of the image as small as 5 mm or the size of a small pea which could be potential polyps. The radiologist is also able to manually highlight any irregularities for closer inspection. Once suspect polyps are found, the software can precisely identify the boundaries and features and show them in 3D with a volume measurement, diameter, shape and location. This allows the radiologist to accurately review and track any growth in the polyps.
"We know that polyps of a certain size will likely become cancerous and should be removed so accurately measuring polyps is key," said Dr. Mary Roddie, consultant radiologist at Hammersmith Hospitals NHS Trust in London and Medicsight's director of radiology. "To accurately calculate the volume or size of a polyp, you first must have accurate boundary identification. The Medicsight software reliably identifies the boundaries and calculates the volume, giving radiologists more dependable information on which to base their medical decisions. In addition, and more importantly, accurate boundary identification is what makes detection of these polyps possible."
Medicsight Colon CAR 1.2 is the first "joint-read" software available for CT colonography. Unlike "second-read" software, in which software is employed after radiologists complete their reviews, joint-read software enables radiologists to review the original, "unfiltered" image side-by-side and simultaneously with a software-enhanced image showing regions of interest. For a radiologist reviewing up to 1,400 images from a single CT colonography, the joint-read capability saves time and can produce optimal results in the evaluation of polyps.
Dr. Perry Pickhardt, the University of Wisconsin Medical School researcher who published a landmark study on virtual colonoscopy in the New England Journal of Medicine in December 2003, said software such as Medicsight's will help make virtual colonoscopy a practical and widely accepted means of screening for colorectal cancer. "With the enormous amount of data generated from a single CT scan, having an effective software tool to aid the radiologist in detection and accurate measurement will be critical," he said.
"Locating polyps hidden within folds in the colon or having to distinguish polyps from surrounding healthy tissue can be challenging for radiologists," said Jamshid Dehmeshki, Chief Technology Officer. "We have developed a software tool that is the most advanced and powerful in boundary identification and measurement of polyps, which in turn will assist radiologists in decision making and diagnosis."
Dr. Steve Halligan, a leading radiological researcher from St. Marks Hospital in the United Kingdom, is currently evaluating the benefits of the new software in a large-scale trial involving 10 radiologists with each interpreting over 200 CT colonographic datasets. Dr. Stuart Taylor, Dr. Halligan's colleague at St. Marks, points to the large differences in reader performance as one of the major challenges facing CT colonography. "There is good evidence that the diagnostic performance of CT colonography is in part dependent on the ability and experience of the reporting radiologist," said Dr. Taylor. "The current trial hypothesizes that the new joint read CAR software will significantly improve the ability of non-expert radiologists to detect colonic polyps, and thereby match the performance of their expert colleagues."
Colorectal cancer is the third most common cancer found in men and women in the United States, according to the American Cancer Society (ACS). The ACS estimates that about 106,370 new cases of colon cancer and 40,570 new cases of rectal cancer will be diagnosed in 2004. Colorectal cancer accounts for about 10 percent of annual cancer deaths. The ACS believes screening reduces the number of cancer deaths and can prevent the development of colon cancer by identifying and removing polyps, from which colon cancers often develop.
In August 2004, Medicsight announced that it received clearance from the U.S. Food and Drug Administration for Medicsight Lung CAR, an image analysis software tool that assists radiologists in evaluating lesions or nodules found during CT scans of the lung.
Medicsight is a software development business focused on the medical imaging market. The company is using its core technology to develop automatic detection and analytical tools for clinicians to improve their ability to diagnose and treat disease. Medicsight has initially focused on three key clinical areas -- lung cancer, colon cancer, and coronary heart disease -- which together account for almost half of all deaths in America. Medicsight's software will enable earlier and more accurate detection and treatment of suspicious lesions, which can save lives and reduce healthcare costs.
Headquartered in London, Medicsight employs more than 60 people and has offices in Nashville, Tenn. and Singapore. Since its inception in 1999, Medicsight has been developing its core algorithms through a team of more than 30 scientists and software developers. For further information, please visit http://www.medicsight.com/.
Medicsight
© 2004 PR Newswire
