A new study published on Wednesday notes that AI programs have become as good as human experts when it comes to medical diagnosis. The paper titled " A comparison of deep learning performance against healthcare professionals in detecting diseases from medical imaging: a systematic review and meta-analysis" was published on digital health journal the Lancet.

The researchers analysed over 31, 587 studies published and picked 14 studies, which reported good quality data. These 14 studies based on human diseases was used for the comparison between deep learning models and healthcare professionals. The team tested the deep learning systems with images separate dataset to the one used to train it and showed the same images to human experts. 

According to the paper, the researchers concluded that the deep learning system detected a disease correctly 87% of the time, as opposed to 86% for health care professionals. Furthermore, when it comes to giving the all-clear, the deep learning system was accurate 92.5% of times, while the healthcare professional was accurate, only 90.5% of times.

According to the authors, this review is the first to systematically compare the diagnostic accuracy of all deep learning models against healthcare professionals using medical imaging published to date. The study found that the diagnostic performance of deep learning models to be equivalent to that of healthcare professionals. However, they also noted that the latter was not provided with additional patient information; otherwise, they would have in the real world.