Using automated deep learning software, physicians without any background in computer programming were able to develop effective artificial intelligence technology for medical image diagnostic classification, a recent study found.
As described in The Lancet Digital Health, scientists trained a deep learning system to classify common diseases based on medical imaging data.
Physicians were tasked with using the system to create algorithms that could perform clinical classification and which fit into their individual workflows.
Nearly all of the resulting algorithms were found to be just as accurate in diagnostic classification as existing professionally developed algorithms. According to the study’s authors, these findings are promising for multiple reasons:
Not only do they demonstrate physicians’ abilities and willingness to play a role in the development of digital health tools, but they also prove how deep learning systems can be used to enrich clinicians’ understandings of the inner workings of AI and other advanced technology in healthcare.
“The availability of automated deep learning might be a cornerstone for the democratization of sophisticated algorithmic modeling in healthcare,” they wrote, while warning,
“The translation of this technological success to meaningful clinical effect requires concerted efforts and a careful stepwise approach to avoid biasing the results. Deep learning experts and clinicians will need to collaborate in ensuring the safe implementation of artificial intelligence.”
——————————————————
Photo courtesy of: Becker’s Hospital Review
Originally Published On: Becker’s Hospital Review
Follow Medical Coding Pro on Twitter: www.Twitter.com/CodingPro1
Like Us On Facebook: www.Facebook.com/MedicalCodingPro
