AI holds promise for fast-tracking treatment and personalizing medicine — but it may not come to your clinic anytime soon

  • AI is bringing change to medical-imaging analysis and helping to speed up diagnoses. 
  • The technology also has the potential to sort through vast amounts of genetic information.
  • Experts predict AI could integrate image-based diagnosis and DNA-based personalized medicine.
  • This article is part of “Big Trends in Healthcare,” a series exploring the top trends shaping the future of the industry. 

LoadingSomething is loading.

Thanks for signing up!

Access your favorite topics in a personalized feed while you’re on the go.

As artificial intelligence becomes more common in healthcare, Dr. James C. Tsai anticipates the technology will help him and other clinicians diagnose patients more quickly and accurately, allowing them to fast-track treatment. 

“Ophthalmic AI has a lot of implications for human health,” Tsai, an ophthalmologist and the founding director of the Center for Ophthalmic Artificial Intelligence and Human Health at the Icahn School of Medicine at Mount Sinai, told Insider. “We’re in the early days, but AI will help me to better consider and evaluate a diagnosis I wouldn’t have before.” 

AI already helps doctors understand complicated images and streamline diagnoses. In fact, the US Food and Drug Administration has approved more than 500 medical AI algorithms, with the majority in the radiology field.

Tsai predicts that AI will one day help doctors analyze complicated genetic data — a hallmark of precision medicine.

Together, AI imaging and genetic analysis may help doctors rapidly pinpoint a diagnosis and create a highly personalized treatment plan, thus improving a patient’s care.

AI already has a heavy hand in imaging 

AI is largely being studied for analyzing complex medical images to diagnose patients, said Dr. Peter D. Chang, the director at the Center for Applied AI Research at the University of California, Irvine.  

“AI is really in every facet of that process now,” he said. “Everything from making the images themselves clearer to speeding up how fast physicians can make a diagnosis.”  

For example, Chang and his colleagues at UCI’s Comprehensive Stroke and Cerebrovascular Center use AI-based tools to analyze brain scans and triage patients, accelerating the diagnosis of stroke and vascular conditions such as tears in the aorta, lung clots, or trauma-related bleeding. 

For Tsai, AI is instrumental in analyzing highly detailed images of the eye and can help narrow down what may be ailing a patient beyond ocular conditions like glaucoma and macular degeneration. 

AI analysis of eye images can help predict what conditions a patient is at risk of developing, including stroke, diabetes, cardiovascular disease, mild cognitive impairment, or even the risk of Alzheimer’s disease and Parkinson’s disease, he said. 

“The eye tells us so much about the cardiovascular system through the blood vessels in the retina and neurological health through the optic nerve,” he said. 

AI and the potential for genetic analysis

In addition to medical imaging, AI could one day comb through large amounts of genetic information, a challenging task for researchers.

Through the analysis of the genome — or the complete set of genetic instructions for an individual — artificial intelligence may help doctors and researchers understand why some patients respond to certain treatments and others don’t, Dr. Zhenghe J. Wang, the chair of the department of genetics and genome sciences and the co-leader of the cancer genome and epigenome program at Case Western Reserve University, told Insider. 

“We have a lot of genomic data, but to make sense of it can be really challenging,” he said. “AI will be a way for us to tease out important information that the human brain can’t. It’s an exciting area of study.”

Notably, AI may be able to help researchers identify biomarkers in cancer patients that show whether they will respond to a particular medication, Wang said. In addition, the technology may be able to sort through the results of genetic testing. Some genetic mutations cause diseases while others fall into a gray area. 

“We don’t know if some genetic variants actually cause disease or not,” he said, adding that AI may be able to help doctors sort through these nuances to determine if a patient is at risk of developing a disease. 

A possible convergence in the clinic 

Wang does not see combining AI imaging and AI genetic analysis at the doctor’s office happening within the next couple of years. “One day, maybe,” he said. “We are still at an early stage.”

Pathology reports, or the study of human tissue samples, are still critical in diagnoses and treatment decisions and at this point may be more accurate than using AI tools, he added. 

Research exploring the intersection of genetics and imaging is ongoing. Chang and his colleagues are investigating how AI analysis of brain scans can predict genetic mutations in brain tumors. Information about a tumor’s genetic mutations, which guide treatment decisions about radiation therapy and chemotherapy, is traditionally only available after the tumor has been surgically removed. He said that using AI to predict these genetic factors through noninvasive imaging can streamline and optimize patient care.

Tsai anticipates a day when AI can analyze vast amounts of genetic data in addition to imaging scans to help him create personalized treatment plans for his patients. 

“While a lot of AI in medicine is now more image-based, and precision medicine is more DNA-based, I see a future where they’ll likely intersect,” said Tsai.

Read More

Heather Lindsey