Scientists at the University of Warwick have developed a new statistical method for automatic mitotic cell detection to diagnose breast cancer.

The new statistical image analysis method segments tumor regions and detects dividing cells in tissue samples, facilitating breast cancer grading.

The University of Warwick Department of Computer Science associate professor Dr Nasir Rajpoot said objectivity needs to increase in the cancer grading process.

"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," Rajpoot said.

"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."

"While the current research has focused on breast cancer histology images, the scientists believe the method can be applied to other cancer types."

A three-step method to detect mitotic cells was developed by a team at the University of Warwick.

The method firstly segments the tumor margins for the precise detection of the mitotic cell.

The second step involves the statistical modelling of the intensity distribution of mitotic and non-mitotic cells in tumour areas, which aids the detection of potential mitotic cells.

Finally, the surrounding architecture of the potential cells will be observed to confirm them as mitotic, reducing the number of false positives.

While the current research has focused on breast cancer histology images, the scientists believe the method can be applied to other cancer types.

A pilot study successfully tested the method against two professional pathologists’ identification of mitotic cells, and larger scale trials are presently under way.