Deep learning computer helps detect bladder cancer

A new deep learning computer model for testing bladder cancer patients in a biomedical clinic has been developed by researchers at the Georgia Institute of Technology and the Georgia Clinical and Translational Science Institute (GACT). The model, originally developed by Barry Misukow, professor of computer science at Georgia Tech, uses a bladder cancer hyper-specific antigen-specific T cell receptor (CAR) assay that also can be used to detect polycystic ovary syndrome (PCOS). This novel realization could lead to early detection of PCOS, a common female infertility disorder.

PCOS is commonly found in patients who carry a BRCA mutation, as well as other genes. The disease disrupts the function of female B cells, leading to infertility.

GACT and GACT researchers, along with their Georgia Tech and University of Georgia colleagues, have developed a highly sensitive assay to detect hormone-sensitive breast cancers. This assay has the potential to be used in both screening and pill-delivery in women with PCOS or HER2-positive cancers, a common female infertility disorder.

The published research in the peer-review Group of Clinical Cancer at GACT’s National Center for Advancing Translational Sciences includes a study evaluating the impact of using the Gabby-Rad enough T cell for detecting polycystic ovary on the quality of treatment decisions of patients.

Specifically, tests have shown that this assay will produce more accurate use of scarce biopsies of the female reproductive system, specifically the large ovaries, the endometrial gland and the pituitary gland. “This work is also helpful in development of alternative diagnostic tools and in the development of diagnostic tools for other medically relevant disease processes, such as cervical cancer, ” said Dr. Christopher Dyer, assistant professor in the Georgia Institute of Technology’s Commodite and Cancer Center and one of the authors of the study.