How Artificial Intelligence can help clinicians focus
Literature abounds on the use of ultrasound. The technology is applied to everything from peering at a fetus inside a body to helping to diagnose shock. It’s fast and generally affordable.
However, it demands a fair amount of expertise and concentration to obtain correct measurements from ultrasound images and to interpret those images.
Today, artificial intelligence (AI) adds to ultrasound a powerful layer that lets users gain critically needed information – especially when time is of the essence.
When a patient arrives in the ER with symptoms characteristic of shock, the attending physician needs to act quickly. The Venue ultrasound platform offers a Shock Toolkit of automated tools – developed on principles of machine learning — to get information about the heart, lungs and inferior vena cava, automating the parts of shock evaluation that are the most tedious or difficult.
“With the AI-powered Venue ultrasound system, users with basic ultrasound skills can use the AI tool and obtain cardiac measurements rapidly,” said Dr. Srikar Adhikari with the College of Medicine at the University of Arizona at Tucson. “Among clinicians who use Point of Care ultrasound, expertise is variable, so this machine can help to validate important decisions.”
AI in ultrasound helps to create consistency, whether there’s a different user 10 minutes later or the same user the next day. One user may be very liberal with his measurements, another very conservative with hers. Consistency enables comparison of exams and helps to improve documentation of exams, which facilitates credentialing.
The Venue ultrasound platform employs algorithms developed by machine learning; those algorithms provide the calculations that the human user would otherwise have to make himself, saving him precious time. And, if he makes an error in his calculations the next day, he may erroneously believe his patient’s health has changed.
“In the ER, where there are constant distractions, minutes saved can improve patient care,” said Dr. Adhikari. “If I can get measurements on a hypotensive patient in five or 10 minutes, we can call in a needed prescription right away, and everything downstream goes faster.”
AI frees the user to focus on the patient and not the calculations. In other words, the practitioner can do what she does best: practice medicine.
“AI will augment what clinicians do by increasing their workflow efficiency and minimizing errors and variability,” explained Dr. Adhikari. “For example, I don’t believe AI will ever replace radiologists because they do a lot more than just interpreting images; in fact, AI will empower radiologists to do more meaningful work that calls upon their experience, training and skill set.”