AI Tool Could Identify Serious Illnesses through Photographs
An AI tool might be the new way to identify cancer types and genetic changes in lung tumors, according to a recent study by researchers at NYU School of Medicine. The tool can use just a photograph to identify serious diseases.
A new AI technology can identify cancer types and genetic changes in lung tumors, potentially speeding up the diagnosis and treatment process.
According to ANSI, this tool can use machine learning to analyze images of patients’ lung tumors and differentiate adenocarcinoma and squamous cell carcinoma with 97 percent accuracy. Interestingly enough, the study that tested this technology showed that the patients that the AI often mistyped were the ones that pathologists had also misdiagnosed, showing the difficulty in differentiating between these two lung cancer types.
With slightly less accurate, but still impressive accuracy, it was also able to identify the presence of abnormal versions of six genes linked to lung cancer in cells with 73 to 86 percent accuracy. The computer technology used the same technique — machine learning. This could significantly speed up the time before diagnosis, as current genetic tests can take weeks to return results.
AI is continuing to become more accepted in the medical field, as it creates opportunities for the medical field to be expand to underdeveloped areas and potentially improve outcomes of patients. This acceptance is relatively new, with the FDA issued their first approval of marketing a technology that uses AI to detect diabetes-related eye problems in 2018.
After the adoption of a new policy on the use of AI in healthcare, the American Medical Association said the policy aligns with their main principal – advancing the quadruple aim. This means it should “enhance the patient experience of care and outcomes, improve population health, reduce overall costs for the health care system while increasing value, and support the professional satisfaction of physicians and the health care team.”
While general AI standards have been around for a while, standards for AI in healthcare are in recent development along with the technology. AAMI recently released a position paper on the importance of standardization of AI in healthcare. “The emergence of artificial and machine learning in healthcare: Recommendations to support governance and regulation” analyzes AI standardization in healthcare techniques and provides recommendations for future standardization activities.