Healthcare AI company launches radiology-specific vision language model
Harrison.ai is launching a vision language model specific to radiology, the healthcare AI technology company announced September 4.
Harrison.rad.1 can conduct chats related to imaging, identify and localize X-ray findings, generate reports and provide its reasoning based on patient history and clinical context. Unlike other generative AI models, the dialogue-based model has been trained on real-world, proprietary radiology data consisting of millions of clinical radiology images, studies and reports.
"The Harrison.rad.1 model is transformative and an exciting next step for the company,” Robyn Denholm, Harrison.ai board director, said in the announcement. “Harrison.ai is delivering on the promise of helping solve real-world problems more effectively and reliably and helping to save lives."
When tested, the model has yielded impressive performance on the Fellowship of the Royal College of Radiologists (FRCR) 2B Rapids examination, scoring on par with the 40%-59% of radiologists who pass the exam on their first attempt.
Harrison.rad.1 scored a 51.4 out of 60 on its first attempt—higher than the average 50.88 out of 60 radiologists score on their second attempt of the exam. In comparison, other models, like GPT-4, Gemini-1.5 Pro, and LLaVA-Med scored below 30 on average.
Co-founder and CEO of Harrison.ai Aengus Tran, MD, described the model as “groundbreaking” for the field of radiology.
"We are already excited by the performance of the model to date. It outperforms major LLMs in the Royal College of Radiologists' 2B exam by approximately 2x,” Tran said. “The launch of this model and our plan to engage in further open and competitive evaluation by professionals underscores our commitment to responsible AI development."
The model is currently being made available to select industry partners, healthcare professionals and regulators to encourage conversations centered on the responsible use of AI in medicine.
Tran added that the company “is committed to being a leading global voice in helping inform and contribute to an important conversation on the future of AI in healthcare.”
Learn more about Harrison.rad.1 here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She joined Innovate Healthcare in 2021 and has since put her unique expertise to use in her editorial role with Health Imaging.