A statistical image analysis method has been developed to assist in grading breast cancer by automatically segmenting tumor regions and detecting dividing cells in tissue samples. The system promises to bring objectivity and automation to the cancer grading process, which determines the aggressiveness of the treatment offered to the patient.

The number of mitotic cells, which are cells that are dividing to create new cells, is a key indication used by histopathologists to diagnose and grade cancer. The currently predominant system in much of the world is based on expert analysis of tissue samples to determine the severity of the cancer. This subjective system depends on visual analysis, and so it can have substantial variability in diagnostic assessments and thus low agreement between pathologists. A pilot study by the researchers found 19% agreement between three pathologists in identifying mitotic cells.

This research team sought to increase objectivity by developing a three-stage method that takes an image of tissue samples and applies statistical modeling to detect mitotic cells in that image.

Continue Reading

“It has long been recognized that there is a need to increase objectivity in the cancer grading process,” explained Nasir Rajpoot, PhD, of University of Warwick, United Kingdom. “This grading process determines the treatment offered to people who have been diagnosed with cancer, so it’s vital to get it right in order to prevent patients undergoing unnecessarily aggressive treatments. We believe our method takes a significant step towards this by offering an objective, automatic technique to assist the pathologists in grading of breast cancer.”

The method consists of three steps. First, tumor margins are segmented, which is critical to accurately detecting mitotic cells. Second, the intensity distribution of mitotic and nonmitotic cells in tumor areas is modeled statistically, which identifies potential mitotic cells in tumor areas. Third, the surrounding architecture of these potential mitotic cell candidates is examined in order to confirm them as mitotic cells, which reduces the number of false alarms.

This study was presented at the International Conference on Pattern Recognition, held November 11-16 in Tsukuba, Japan.