Advanced Imaging


Advanced Imaging Magazine

Updated: July 8th, 2008 05:26 PM CDT

Image Analysis in Healthcare

Taking the Subjectivity Out of Specimen Analysis
Color deconvolution: A sophisticated and easy-to-use multipurpose algorithm that accurately separates the image to show the actual contribution of each stain and quantifies the area and intensities of the stains.
The IHC (immunohistochemistry) membrane image analysis algorithm is optimized for HER2 (Human Epidermal growth factor receptor 2) stained breast tissue slides, but also can be used for other tissue types and membrane stains.

By Allen H. Olson, Ph.D.

The healthcare pathology marketplace is poised for dramatic change in the coming years, prompted by a convergence of technology that is making digital pathology a practical alternative to traditional microscopy in numerous situations. Initially, the role of digital pathology is likely to augment traditional microscopy, rather than replace it. However, digital pathology's superior viewing and analysis capabilities, combined with the increased portability of images, will gradually increase its adoption. Glass slides and microscopes still will be found in the pathology and research departments of the future, but they are likely to be accompanied by digital slide scanners, sophisticated software, and extensive data storage capabilities.

While the term digital pathology is relatively new, the concept is an extension of what is known as virtual microscopy, which is the practice of converting entire glass microscope slides into high-resolution, whole-slide digital images. The difference in digital pathology is that it not only encompasses image acquisition, but also image analysis, remote viewing, and data management capabilities that include information security.

What has prompted the rise in digital pathology is a convergence of technology advances that have improved the speed of multiple processes while reducing the cost of solutions. For example, today's line-scanning technology can produce a digital image of an entire microscope slide with giga-pixel resolution and 24-bit true color in less than two minutes. High-capacity slide scanners can automatically load and scan several hundred slides at a time at submicron resolutions that correspond to conventional 20x and 40x microscope objective lenses. New scanners capable of using high numerical aperture oil-immersion objective lenses have become available to provide hematopathologists and microbiologists with the ability to digitize blood smear, Gram stain and other microscopic specimens requiring 100x oil-immersion resolutions.

The decreased cost of data storage also is a factor, since the file size of slide scans can be as large as 20 gigabytes. In addition, the increased use of picture archiving and communications systems (PACS) in radiology departments enables healthcare organizations to further leverage their existing technology investment to store digital pathology images in a DICOM-compliant format.

Image Analysis Algorithms:

New tools and technologies in digital pathology have given rise to more sophisticated image analysis capabilities. Image analysis is a cutting-edge capability that provides automated quantitative analytical techniques that can shorten research and analysis efforts by as much as 90 percent. With automated image analysis, complex computer algorithms are used to enhance the slide image, making interpretation straightforward and facilitating pathological interpretation.

1 2 3 next

Subscribe to our RSS Feeds