Artificial Intelligence is reaching every branch of the sector including the medical sector. Recently, scientists developed an AI that can detect the outcome of the most dreadful disease, cancer, in a body by just studying tissue samples. This AI can also give you personalised treatment strategies.
The study was published in the journal Nature Communications. According to the researchers, the technology studies the spatial arrangement of cells in tissue samples like a jigsaw puzzle.
The study leader, Guanghua Xiao, “The study showcases the remarkable ability of AI to grasp these intricate spatial relationships among cells within tissues, extracting subtle information previously beyond human comprehension while predicting patient outcomes,” who is a professor at the University of Texas Southwestern Medical Center in the US
In standard medical practice, tissue samples are regularly gathered from patients and placed on slides, a procedure aimed at enabling pathologists to interpret them for diagnostic purposes. However, this traditional process is time-intensive, and there can be variations in interpretations among pathologists, as per the researchers.
Over the past few years, several AI models have been developed to replicate certain aspects of a pathologist’s role. For instance, these models can identify cell types or use cell proximity as a way to approximate interactions between cells. Nevertheless, these models fall short of accurately reproducing more intricate aspects of how pathologists analyze tissue images. This includes tasks like recognizing patterns in cell spatial organization and filtering out extraneous “noise” in images, which can complicate the interpretation process.
