Artificial Intelligence

Artificial intelligence (AI) could prevent unnecessary diagnostic tests in patients with stable chest pain, according to research presented today at ICNC 2019. A decision support system save one hour of testing per patient. The International Conference on Nuclear Cardiology and Cardiac CT (ICNC) is co-organised by the American Society of Nuclear Cardiology (ASNC); so the European Association of Cardiovascular Imaging (EACVI) of the European Society of Cardiology (ESC).
Study author Dr. Marco Mazzanti, of the Royal Brompton Hospital, London, UK, said: They know that doctors overtest patients and ignore recommendations about when a test justify about two-thirds of the time. Our super brain decision support system, called ARTICA; so strictly follows ESC guidelines and does not advise unnecessary examinations.

Patients with chest pain

The study enroll 982 patients with stable chest pain; so a frequent cause of visits to emergency departments and general practitioners. The researchers compare decisions on which tests to perform made by a cardiologist and by ARTICA on the same day. ARTICA advise no further testing in 658 (67%) patients whereas a cardiologist decide that only 45 (4.6%) patients did not need more tests.

A compute tomography angiography (CTA) scan show that 639 (97%) of the patients ARTICA said did not need tests had no significant coronary artery disease; so meaning the decision was correct. Avoiding these tests would save staff one hour and patients two hours on average.

Dr. Mazzanti said: AI has the potential to save costs and staff time; so by identifying patients with chest pain who do not have significant coronary artery disease and therefore do not need expensive cardiac imaging. To take one example, a CTA scan, use to look for block blood vessels, costs €200-400. Cardiologists recommend it for 816 (83%) patients while ARTICA recommend it for just 95 patients (10%).

Recommend exercise testing

Dr. Mazzanti said: As doctors we order a lot of tests which cost a lot of money and waste time. ARTICA is like a second set of eyes to make sure we follow recommendations. They note that ARTICA recommend exercise testing or functional imaging for 224 (23%) patients while cardiologists recommend it for just 100 (10%) patients.
They know that when ARTICA says don’t do a test it is almost 100% right because the CTA scan confirm no block arteries, said Dr. Mazzanti. When ARTICA decides a test is needed; hence they are less certain that this is correct. By adding more data to the super brain these decisions will become more accurate and enable us to deliver more personalize care.
ARTICA, which stands for artificial Intelligence for clinical Cardiac navigation, is a decision support system create by the researchers. It uses machine learning, a type of AI; so to make decisions that adhere to recommend practice. The researchers inputted guidelines for patients with stable chest pain and routinely collect medical data. A machine learning algorithm analysed the information repeatedly until it learned to identify who need a test (and which test) and who did not.