In 2017, the Event Horizon Telescope (EHT) network captured a picture of the black hole at the center of the galaxy Messier 87, located more than 53 million light-years from Earth. Now researchers have managed to improve the image using an artificial intelligence tool, reports The Astrophysical Journal Letters.

To improve the image, a new method called PRIMO was used – artificial intelligence that learns to restore high quality images even from fuzzy input data. Trained on generated simulations of more than 30,000 black holes, the AI uses data based on what we know about the physical laws of the universe, including black holes, to create more accurate pictures from the raw data obtained in 2017.

Black holes are mysterious regions of space where gravity is so strong that nothing, not even light, can escape. They form when dying stars are compressed by their own gravity. As a result, the collapse compresses the mass of the star into a tiny space. The boundary between a black hole and the mass surrounding it is called the event horizon – the point of no return, from where everything that crosses it will no longer come back.

The researchers note that the new image is consistent with the predictions of Albert Einstein. However, they expect that further research in machine learning and telescope hardware will lead to the possibility of obtaining even better images of black holes.