A patient vividly demonstrates how artificial intelligence (AI) can potentially enhance patient care in complex health systems by catching missed care opportunities. After entering the Emergency Department (ED) with acute abdominal pain, they are quickly diagnosed via a CT scan with diverticulitis involving a perforation, swiftly treated and sent home pain-free. However, buried in the lengthy CT report was a mention of a small aortic aneurysm (AA) which was not the focus of the immediate health crisis, and therefore did not prompt any immediate action from the ED or the patient’s primary care physician.
A lack of symptoms associated with smaller AAs, unfortunately, often result in a lack of patient follow-up. Several years down the line the patient’s once-small AA could now be at risk for rupture, a potentially fatal event, due to its predictable growth rate. With an exponential increase of patient data flowing through health systems, from lab results to imaging reports and test records, ensuring accurate AA diagnoses, adequate patient management, and timely interventions is becoming increasingly challenging.
Artificial intelligence, specifically Natural Language Processing (NLP), could provide a solution. In a world where PCPs and ED physicians are becoming increasingly stretched for time, AI can offer support. NLP can interpret human language either spoken or written, in this case from radiology reports, identifying vital clinical data about AAs. Importantly, this can then be passed onto relevant clinicians who can decide on the necessary course of action. AI has the potential to flag up AAs that are initially found from scans for unrelated issues, and ensure that they are reviewed by the appropriate specialists, ensuring reliable follow-up care.
Currently, aortic aneurysm care is considered to be at a very good standard with physicians able to monitor AA patients for years without intervention being necessary, particularly if they are not symptomatic or urgent. Treatments, when required, can include minimally invasive EVAR (endovascular aneurysm repair), as well as lifestyle modifications like quitting smoking, dietary changes, exercise, and medicinal therapy. However, without accurate tracking and management, patients can still be lost in the follow-up process.
While the AA is typically not an urgent condition, as it grows, so does the risk of rupture, which can be life-threatening. Given that most AAs do not exhibit symptoms until they rupture or have grown significantly, follow-up and suitable management are crucial. However, this follow-up is often lost due to the overwhelming volume of data healthcare providers need to comb through daily, leading to the need for AI assistance.
Using AI tools such as Aidoc’s, health systems, practitioners, and patients can remain connected and drive optimal care decisions. By employing AI to analyze scanning reports, it is possible to prevent the oversight of conditions such as AA and provide patients with more proactive, effective care. Therefore, AI has the potential to significantly impact the healthcare industry by increasing efficiency and improving patient outcomes.
