Medical Image Analysis Overview

Voting Technology

Parascript initially applied a voting mechanism in the postal automation industry. A combination of human-like holistic analysis, multiple neural networks and sophisticated statistical voting algorithms enabled a significant improvement in recognition rates and a decrease in error rates in mail processing. These advancements in mail automation were first achieved in 1998, when the Remote Computer Reader applied by the United States Postal Service was recognizing about 35% of machine printed and 2% of handwritten letter mail pieces. Thanks to the use of voting methodology, modern systems recognize 93% of machine printed and about 88% of handwritten mail - or more than 90% cumulatively. Similarly, the application of multiple engines in a voting scheme for the banking and financial services industry raised the read rates in payment automation from 40% in 1997 to the current 80% level at 1% error rate.

Medical image analysis and image processing

Medical image analysis encompasses a broad range of techniques and processes aimed at creating and analyzing images of the human body to reveal, diagnose or examine disease. Mammography is one such procedure that uses low-dose X-rays to examine the human breast. Mammography is primarily used to detect and diagnose breast cancer, as well as to evaluate palpable masses and no palpable breast lesions. It has been proven to reduce mortality from breast cancer and is an essential part of regular breast care. Mammography is the tool of choice, used in conjunction with physical examination, for early breast cancer screening.

Medical image analysis and image processing are often applied as an interpretive aid during the mammography image review. Radiologists frequently use Computer-Aided Detection (CAD) software to help them read medical images. The purpose of CAD is to identify and highlight hard-to-find features and anomalies on medical images that may be indicative of cancer and bring them to the attention of the radiologist for further review, while also minimizing false-positive readings (i.e. identification of non-cancerous regions of interest as suspects) that burden the physician’s review process and can lead to higher recall rates.

Medical imaging centers, radiology departments and hospitals are embracing the benefits of computerized second read in mammogram images to detect breast cancers earlier. Radiologists report that on many occasions CAD systems helped them to confirm that a suspicious area required further investigation. The American College of Radiology, Medicare, American Cancer Society and patient advocacy groups have recognized the benefits of CAD technology. As a result, mammograms and CAD screenings are now encouraged as a preventive measure and are reimbursed by Medicare and other insurance companies.

As new Computer Aided Detection (CAD) technologies emerge, mammography CAD continues to prove its advantages in an evolving breast imaging world. The reality of mammography is that fewer radiologists are entering the mammography field. CAD provides a critical aid as the crisis hits, because it can reduce double-reading while increasing the detection rate. With CAD algorithms on a continuous improvement path, more and more CAD systems are deployed in double-read environments and faster adoption is only hampered by reports that current CAD systems still deliver unacceptably high false-positive rates.

Advancements in medical image analysis and recognition technology will result in the introduction of next generation CAD systems that will be readily adapted as a second opinion with false-positive rates approaching those of a second radiologist.

One way to reach this point is to use voting technology in medical image analysis. While new to CAD for mammography, this approach has been used for over a decade in image processing for postal and payment automation industries.

Voting in Medical Image Analysis

Based on its successful implementation of image analysis and recognition software in postal and payments automation industries, Parascript believes that the integration of several complementary algorithms within one CAD system or the utilization of two CAD systems and sophisticated voting methods in medical image analysis, can enable next generation CAD systems to achieve high sensitivity and low false-positive rates for greater levels of certainty.

Voting mechanisms can offer many technological advances in medical image analysis. Each image recognition process may identify areas of interest on the mammogram image independently, without sharing information with other image recognition processes. After image recognition processes individually identify areas of interest or objects on the mammogram image, the different areas can be compared to determine a confidence value related to the accuracy of the identifications. Comparing the results of multiple image recognition processes allows for the mitigation of the inherent faults of the image recognition process, thus leading to reduced false-positive and false-negative rates.

Medical image analysis voting is fully utilized within Parascript's Computer Aided Detection algorithm intended for OEM customers that are interested in improving the performance of their CAD and mammography systems.

With Parascript, you get a powerful and flexible medical image analysis and processing algorithm that can locate hard-to-find regions of interest with high sensitivity and low false-positive rates. The algorithm may allow you to:

Detect breast anomalies earlier
Locate hard-to-find regions of interest with high sensitivity
Achieve low false-positive rates
Integrate easily with existing analog and digital systems

Parascript® AccuDetect® - Medical image analysis and processing algorithm: detects breast lesions earlier and with a higher level of assurance



		HOME \| CONTACT US \| CAREERS \| CUSTOMER LOGIN \| SEARCH \| SITE MAP