Moving radiology into the age of artificial intelligence

Radiology plays a pivotal role in health care, enabling doctors to visualise the body's internal structures, detect diseases, and guide surgical procedures. The interpretation of these complex medical images can be time-consuming, often requiring the expertise of highly trained radiologists. Over the past few years, the field of radiology has been undergoing a transformation as advancements in artificial intelligence (AI) algorithms can recognise patterns and abnormalities in medical images, thus speeding the process and manpower needed to interpret images. Therefore, the intersection of radiology and generative AI holds significant promise for enhancing healthcare and improving patient outcomes. 

Recent releases of generative AI from US-based technology companies has accelerated and enhanced the radiologist’s workflow, increasing diagnostic accuracy and improving patient outcomes. For the field of radiology, this means a greater opportunity for efficiency in value-based care and patient-centred outcomes. 

In an interview with Omnia Health, Raymond Liu, MD, Vice President for Mass General Brigham’s Global Advisory division and Associate Radiologist at the Division of Interventional Radiology at Massachusetts General Hospital, believes the depth and breadth of technological advances like generative AI have the power to transform not just radiology, but the entirety of global healthcare in the coming decade.  

Related: Beyond the hype: Generative AI and healthcare in the Middle East

Generative AI has already impacted pattern recognition, and with its ability to leverage enormous amounts of data. It has further promise in streamlining the entire care continuum even before the radiologist sees an image on their screens. 

Liu highlighted a recent study conducted by Mass General Brigham’s Marc D. Succi, MD, examining patient vignettes fed into Chat GPT. The study found Chat GPT was about 72 percent accurate in making the final diagnosis. “AI has the ability to support the workload of the radiologist, especially with image processing. But AI also has the opportunity to affect the field right from before the radiologists start working on the image to after they finish with the interpretation,” he highlighted. (Who? Ray or Marc?) 

Liu also points out how, in the future, generative AI will be able to help with recommending the right imaging study for a specific diagnostic dilemma and the best type of image and protocols, resulting in valuable cost- and time-saving for radiologists.   

Another area where innovation is occurring is in the medical device field, both diagnostic and interventional. Some of the scanners in the market today can make data acquisitions faster and more efficiently.  

“There is also a lot of excitement in the field of interventional radiology,” said Liu. “The new devices allow us to perform treatments in the fields of oncology and combine them with other therapies, such as immunotherapy or external radiation. This allows for the breaking down the silos between the fields, providing multidisciplinary care, and creating new opportunities for radiology as the technology advances.” 

However, he stressed that there is a need to balance the rewards with risks. For instance, generative AI won't replace radiologists but could replace radiologists who don't use AI to their advantage. “Radiologists need to have AI technologies at their side to perform at top calibre,” he added. 

Additionally, Liu emphasised that some obstacles to what to watch out for are the risks in the regulatory environment, which needs to keep pace with technological advancements. “Often, the regulatory environment works behind, so we need to have safe guardrails around patient privacy and data integrity to mitigate risks. Also, physicians and the workforce need to be comfortable using AI as a supportive friend, so there needs to be a culture and mindset shift,” he said. 

Sharing expertise in Saudi Arabia 

As one of the leading nations in the Middle East, what happens in Saudi Arabia impacts the entire region. Saudi Arabia’s Vision 2030 is one of the most important drivers of the healthcare industry that is accelerating the country’s thriving economy and providing for a vibrant society, both in public health and medical innovation.  

Dr. Liu shared that Mass General Brigham is excited to support healthcare initiatives who share the same patient-centred vision. “What is happening in Saudi Arabia is unique as all the different elements from government leadership to funding sources to the private sector and international collaborations are all coming together and dovetailing to push forward and change how healthcare is being delivered in the country,” he said. 

He added that the MGH Institute’s Health Professions Education Department, part of Mass General Brigham, has launched a master’s degree in partnership with Saudi’s Alfaisal University. “We are excited to work in the Kingdom and share our own learnings as true long-term partners for both public and private healthcare sectors.” 

Related: Lessons from Canada for Saudi Arabia's Vision 2030

As one of the speakers at the Global Health Exhibition taking place in Riyadh, Dr. Liu will discuss ‘Medical imaging in the age of generative AI’ as part of the Future of Radiology conference. He shared: “I will shed light on what the future of innovation looks like for radiology and how generative AI, among other important emerging technologies, will change radiology forever. Almost 20 years ago, the digital revolution moved imaging into the information age, which saw the digitisation of images that used to be on hard films. But today, imaging is gravitating towards the “age of intelligence.” During the discussion, I will review the recent history of technological advances and identify the opportunities and risks for radiology and the entire healthcare system.” 

Dr. Liu will discuss ‘Medical Imaging in the Era of Generative AI (Chat-GPT)’ at 12:35pm on Sunday, October 29, at the Future of Radiology Conference at the Global Health Exhibition.