Can Tetris inspire AI for MRI breast cancer detection?

Borrowing concepts used in the classic computer game Tetris, Australian researchers have developed an artificial intelligence (AI) algorithm that they say can detect tumors on breast MRI exams nearly twice as fast as existing breast cancer detection methods without affecting accuracy.

Their fully automated medical image analysis program utilizes the traversal movement and style of Tetris to examine the breast area, according to the group from the University of Adelaide's Australian Institute for Machine Learning (AIML).

"Just as [the] vintage video game Tetris manipulated geometric shapes to fit a space, this program uses a green square to navigate and search over the breast image to locate lesions," said Gabriel Maicas Suso in a statement from the university. "The square changes to red in color if a lesion is detected."

Suso, a doctoral student, co-developed the deep reinforcement learning-based program with Gustavo Carneiro, PhD, an associate professor at the AIML. In testing, their method was 1.78 times faster than current approaches for detecting breast cancer. What's more, it was found to be just as accurate, according to the researchers. They also noted that they were able to train the algorithm using a relatively small amount of data.

"More research is needed before the program could be used clinically," Carneiro said in the statement. "Our ultimate aim is for this detection method to be used by radiologists to complement, support, and assist their important work in making a precise and quick prognosis."