June 2, 2023

Taylor Daily Press

Complete News World

Better diagnosis of uterine cancer using artificial intelligence

Better diagnosis of uterine cancer using artificial intelligence

Uterine cancer is the most common tumor of the female reproductive system. A lot of research is being done at LUMC on this type of tumor. According to the researchers, studying the use of artificial intelligence contributes to further improving the diagnosis and treatment of uterine cancer.

2000 images of uterine cancer

To achieve this, the program must first be fed to the AI ​​with lots of images. LUMC researchers used microscopic images of uterine cancer from more than 2,000 women who participated in the PORTEC clinical studies, coordinated by LUMC by Professor Carien Creutzberg. Using these unique images, the AI ​​system was able to quickly identify and differentiate between different types of uterine cancer. Importantly, the developed model shows the researchers where in the tissues visual information for predictions is hidden. Thus, the developed AI model is not a black box like many other AI-driven systems.

Artificial intelligence helps in diagnosis and treatment

AI is increasingly being used, just as it is in LUMC, to support specialists in the diagnosis and treatment of cancer. For example, a new automated system, developed by researchers at UMC Utrecht, is able to quickly and accurately scan MRI images of breasts with dense mammary gland tissue using artificial intelligence. The system ensures that the radiologist sees only the images on it aberrations Diagnosed with possibly breast cancer. but also in Colon Cancer AI is already gratefully used, for example to indicate tumors arising during surgery by placing a box around it.

DNA changes

At LUMC, AI is now helping to identify different types of uterine cancer, because no changes in DNA can be seen with the naked eye. The researchers explain this on website Van Lumke: “DNA changes in the tumor determine tumor behavior and disease course, and this has become apparent in recent years, in part due to work at LUMC. There are four types of uterine cancer, each with a different disease course. It is important for the patient and the doctor to know Tumor type, but this currently requires additional expensive DNA tests.With artificial intelligence, this situation could change dramatically.

See also  Socio-demographic differences in COVID-19 mortality in the first year of the COVID-19 epidemic

Sarah Fremond, a doctoral student in pathology at LUMC, explains: “With this study we wanted to learn more about the relationship between tumor onset and underlying DNA changes. Through this work, we learned which areas in tumors contain the most important visual information for prognosis and, therefore, what which pathologists should focus on.”

Health and ICT Conference 2023

On January 30, 2023, ICT and Health usher in the new year for healthcare with healthcare’s premier and influential annual conference on healthcare transformation.
To be present too? Book your entry ticket quickly