Deep learning model uncovers hidden stellar bubbles in Milky Way

Researchers in Japan have leveraged advanced artificial intelligence to explore the distant regions of our galaxy, revealing previously undetected cosmic structures. Led by Osaka Metropolitan University, the team created a deep learning model capable of sifting through immense datasets from space telescopes, uncovering numerous bubble-like formations absent from existing astronomical records.

These structures, commonly referred to as Spitzer bubbles, are prevalent in the Milky Way and other galaxies. They typically emerge during the formation and active life stages of massive stars, offering valuable insights into the mechanisms behind star formation and galactic evolution.

Shimpei Nishimoto, a graduate student at the Graduate School of Science, and Professor Toshikazu Onishi collaborated with researchers nationwide to build and refine the AI-driven system. The model analyzed imagery captured by both the Spitzer Space Telescope and the James Webb Space Telescope, using image recognition to efficiently identify Spitzer bubbles. Additionally, the AI detected shell-shaped features believed to result from supernova events.

"Our results show it is possible to conduct detailed investigations not only of star formation, but also of the effects of explosive events within galaxies," stated graduate student Nishimoto.

Professor Onishi added, "In the future, we hope that advancements in AI technology will accelerate the elucidation of the mechanisms of galaxy evolution and star formation."

Research Report:Infrared Bubble Recognition in the Milky Way and Beyond Using Deep Learning