Artificial Intelligence in Healthcare: Saving Lives Today

In medical emergencies, time is the most critical consideration. The earlier the patient receives appropriate treatment, the greater the chances of recovery.

Can artificial intelligence (AI) literally save lives with its rapid ability to critique data and provide predictive analysis? Well, it’s already happened at least once.

How AI Played a Role in Saving a Patient’s Life

A 50-year-old resident of Nahariya, Israel, arrived at the Galilee Medical Center for a Computed Tomography (CT) scan, complaining of a headache with a CT scan taken, and the patient was sent home to rest.

The results would typically take one to two weeks, standard for the procedure – but possibly deadly when an undetected emergency happens.

This time, however, an AI program scanned the data and alerted the doctors to the outside chance that the patient could be suffering from intracranial bleeding – needing immediate attention.

The AI program is known as Viz.ai, which, along with CT scans, can also analyze echocardiography and other scans.

The doctors immediately sent for the patient and operated — the AI had correctly spotted just how serious the situation was.

The patient’s life was saved, credited to the quick artificial diagnosis, which shaved two weeks between the scan and treatment.

Dr. Don Paz of the Galilee Medical Center said, “Without it, the patient would probably have come to us too late.”

It is not an isolated case, although this next one is a “Dr ChatGPT” example.

A mother struggled for three years to find a solution for her son’s persistent symptoms. After consulting with 17 doctors without success over this period, she eventually sought answers from ChatGPT. Ultimately, it was ChatGPT that provided the much-needed diagnosis — tethered cord syndrome — for her 4-year-old’s chronic pain.

Diagnosis: Success

The incident at the Galilee Medical Center leaves us with specific lessons.

• No matter how much medical technology has improved, on-time diagnosis and treatment are still challenging. Without AI, it may not be possible to expedite diagnosis and treatment significantly.
• AI has arrived in healthcare: Yes, it is still a work-in-progress technology, and there are skeptics, but delayed adoption may not be in the interest of better treatment.

Karim Karti, the CEO of RapidAI, which applies AI to quickly diagnose cases of strokes, aneurysms, pulmonary embolisms, and other medical issues, described the role of AI in saving lives in medical emergencies in an interview with Healthcare IT News. The main excerpts from the interview are given.

• Dr. Greg Albers, the founder of RapidAI, and his colleague Dr. Roland Bammer developed a fully automated AI-powered image-processing software for CT scans and MRIs. Such software can rapidly and accurately generate images of the brain and heart that can contribute to quick decisions and appropriate treatment.
• AI can help clinicians triage emergency patients quickly with rapid and accurate assessments so the right team can be assembled and the proper treatment is given. For example, when a patient presents with the symptoms of a stroke, AI can automate the process of high-quality scan image generation and deliver the images to the mobile devices of doctors or paramedics, send alerts to appropriate medical personnel, and also pull up the patient’s medical history. AI can perform the whole workflow in a matter of a few seconds.
• An important use case of AI is saving time in case of strokes. When a stroke occurs, the human brain loses brain tissue because it doesn’t receive oxygen. Delays can lead to severe consequences such as death or paralysis. AI workflows can save precious time by enabling the medical team to deliver the proper treatment at the right time.
• Doctors can find problems in studying brain aneurysms in CT scans because of the uneven shapes of the aneurysms. AI can provide more precise and detailed pictures, enabling better and quicker decisions.

What is stopping the Wider Adoption of AI?

Despite AI’s proven advantages, adoption in the healthcare industry has been lagging because of certain factors.

According to?Sachin Patel, the CEO of Apixio, which delivers AI-powered insights on healthcare, “The most prominent barrier holding back AI adoption in highly regulated industries like healthcare is the highest level of accuracy necessary to achieve adoption.”

With a streaming service, for example, “if their algorithms are inaccurate, the worst-case scenario is that a customer dislikes the recommended movie or show. Though this may be inconvenient or even annoying, it is not life-threatening.

“However, if AI in healthcare isn’t accurate enough, it may result in an ineffective treatment plan that puts the patient’s health at risk.”

There are also trust issues: the idea of software providing insights and images on a patient’s body, especially in an emergency, is still hard to accept. The patient’s life is at stake, and AI is still not trusted enough, at least not universally.

According to David Friede, vice president of strategic partnerships for DrOwl, “It can also be difficult to trust a machine to make a prediction — this is frequently an issue in the healthcare industry. Doctors are accustomed to relying on human interactions and specifics to determine the best possible course of action, which involves evaluating a patient’s mental state, expectations, medical history, and more.”

The Bottom Line

History is witness to the fact that change always meets resistance and doubts but eventually finds its way through.

AI is still a relatively new phenomenon, though it is arriving fast, especially in the healthcare sector, because it saves lives.

AI has made a difference in pockets, and it will take time to prove its credentials universally in the medical fraternity.

But eventually, the need for better preparedness and handling of emergency care will usher in AI. The change will not be smooth or sequential.

But if it moves the dial on the Hippocratic Oath — “Do No Harm” — then it is likely to become part of the standard medical kit in time.