Software Increases Accuracy Of Breast Cancer Diagnosis

Depending on the type of tissue and the stage of growth at diagnosis, breast cancer has as much as a ninety percent survival rate. But even in the US with its current focus on preventive care and early diagnostics, cancer (off all types, admittedly) is the second leading cause of death for women. Researchers at Rice have developed a software tool, though, that could not only change that for women in the US, but can also increase the survival rate dramatically for women around the world.

Breast cancer detection at every stage has seen a dynamic shift in improved techniques, reduced costs, ease of patient treatment, and speed of results. Advancements such as needle biopsies over invasive surgical detection have also led to a change in the perception patients have, making them more likely to follow through with necessary procedures. But the fact remains that imaging techniques such as mammograms are still required in many places to detect an issue in the first place. Once a questionable result is found, tissue samples from biopsy are sent to a lab to separate out the specimen and examine it microscopically in order to detect cancerous cells. But new software is able to work with intact samples without the need for separation, using high-speed optical microscopy to produce nearly the same result.

This capability can mean the difference between early detection and no detection at all for women who live in regions where lab pathology simply isn’t feasible, cost effective, or fast. Unfortunately, the software is still in testing stages and isn’t ready for widespread clinical use just yet.

According to an interview with Rebecca Richards-Kortum, Rice’s Malcolm Gillis University Professor and professor of bioengineering and of electrical and computer engineering, “Although the software could have substantial clinical relevance, more research and refinement of the classification procedures are needed before the software can be used in a clinical setting.”