Australian Team's AI Method Independently Confirms the First Black Hole Image

A small team at Australia's national science agency has validated one of the most famous images in science. Using a new imaging technique, researchers at CSIRO recovered the glowing ring around the black hole at the center of galaxy M87 from the original Event Horizon Telescope data, and the result matches the 2019 picture in size and shape.

The 2019 image, produced by the work of hundreds of scientists, was the first direct view of a black hole's shadow. Some researchers have since asked how much the final picture was shaped by the processing needed to turn sparse telescope measurements into an image. The new work, published in Nature, addresses that question directly.

"It is like standing in your backyard and trying to take a photo of a doughnut on the moon," said lead author Samuel Lai, a CSIRO astronomer. The team of four built machine learning and artificial intelligence tools to pull out the black hole's structure while filtering out interference from Earth's atmosphere. "When we applied it to the original Event Horizon Telescope dataset, we found a black hole almost identical in size and shape to the historic EHT image," Lai said. "Importantly, our model had no idea what a black hole looked like when we applied it to the M87 data."

The Event Horizon Telescope links radio dishes around the globe so they act as one telescope the size of the planet, a technique called very long baseline interferometry. It is tricky, because it can introduce calibration errors and biases. CSIRO astronomer Nithyanandan Thyagarajan, who led the project, said the team drew on his experience with similar problems in synchrotron experiments. Their approach combines triplets of telescope signals instead of the usual pairs, which cancels out those errors.

The method is called GenDIReCT. Laurent Loinard, director of the EHT Collaboration, welcomed it. "It is extremely gratifying to see that GenDIReCT, an elegant, entirely independent method, recovers the ring structure of M87* across several years, in excellent agreement with the EHT Collaboration's published results," he said. "Studies like this strengthen our confidence in what we have learned about black holes, and enrich the toolkit available to the whole community."

The finding strengthens confidence that the ring around M87's central black hole is a real astrophysical feature and not an artifact of how the image was reconstructed. Lai called it the first truly independent validation of the EHT image, an important step in the scientific process.

The technique may reach well beyond astronomy. Thyagarajan said it could improve medical imaging and help map earthquakes, calling it "an out-of-this-world solution to real-world problems."