Tag: MRI

The FDA has approved AI-based PET/MRI “denoising”. How safe is this technology?

Nov 1, 2019

By LUKE OAKDEN-RAYNER, MD

Super-resolution* promises to be one of the most impactful medical imaging AI technologies, but only if it is safe.

Last week we saw the FDA approve the first MRI super-resolution product, from the same company that received approval for a similar PET product last year. This news seems as good a reason as any to talk about the safety concerns myself and many other people have with these systems.

Disclaimer: the majority of this piece is about medical super-resolution in general, and not about the SubtleMR system itself. That specific system is addressed directly near the end.

Zoom, enhance

Super-resolution is, quite literally, the “zoom and enhance” CSI meme in the gif at the top of this piece. You give the computer a low quality image and it turns it into a high resolution one. Pretty cool stuff, especially because it actually kind of works.

In medical imaging though, it’s better than cool. You ever wonder why an MRI costs so much and can have long wait times? Well, it is because you can only do one scan every 20-30 minutes (with some scans taking an hour or more). The capital and running costs are only spread across one to two dozen patients per day.

So what if you could get an MRI of the same quality in 5 minutes? Maybe two to five times more scans (the “getting patient ready for the scan” time becomes the bottleneck), meaning less cost and more throughput.

This is the dream of medical super-resolution.

The High Cost of Reducing Waste in US Healthcare

Mar 29, 2015• 69

By SAURABH JHA, MD

Recently, a jury awarded a young California resident $28.2 million for a delayed diagnosis of a pelvic tumor. The jury found Kaiser Permanente (KP) negligent. Doctors in the system, touted to be one of the finest systems by the President, allegedly refused an immediate MRI for back pain in a 17 year old. The patient eventually received an MRI three months after presentation, which found a tumor so extensive that the patient needed an amputation.

The case is instructive at multiple levels. It shows a tense dialectic between the individual and society. It also highlights a truism that many don’t understand or don’t acknowledge – missed/ delayed diagnosis and waste are reciprocal. They’re birds of a feather. You can’t have less of one without more of the other.

The patient presented with back pain. MRI for back pain is the poster child of waste. Why so? Because so many are negative. Even more are meaninglessly positive –disc bulges which simply mean “I’m Homo sapiens and I wasn’t intelligently designed to be sitting at the desk.”

High quality doctors don’t order MRI for back pain immediately, reflexively and incontinently. Think about this. A high quality doctor should say “I don’t think you need an MRI because it won’t change the management and doesn’t improve outcomes.” That’s the resounding message from the top. If it doesn’t improve outcomes it’s not a worthy test. High quality doctors will, once in a while, cost their organization a lot of money.

But quality is still not settled. Quality doctors must satisfy patients. If a patient asks for an MRI for back pain the quality doctor must acquiesce, if that refusal dissatisfies. I’m confused. Ordering an MRI for back pain is poor care. But not ordering an MRI for back pain is poor care. Which is it?

We don’t know the facts of the case. It’s possible that the patient had a neurological deficit that should have raised the urgency. It’s possible that the physician didn’t examine the patient and had he/ she examined, the tumor might have been detected. We don’t know. We shouldn’t judge (1).Continue reading…

The Sunnier Side of India’s Free Market Medical Imaging

Jan 12, 2015• 11

By SAURABH JHA, MD

What would medical care be like in a genuine free market?

Nobel laureates in economics have opposing views. But does India have the answer? There, healthcare has a strong private sector: patients usually pay directly and the insurance industry is just emerging.

Milton Friedman believed that markets would work just fine in healthcare. Kenneth Arrow was not so optimistic. In his much cited opus, Arrow singled uncertainty as the key factor which distinguishes medical care from other goods and services. Uncertainty means that one doesn’t know when and how much healthcare one is going to need. Not quite the same as shopping for cereal in Waitrose.

George Akerlof felt that asymmetric information, i.e. when one side knows far more about the product, could be problematic for quality.

In Akerlof’s hypothetical market, “Market for Lemons,” which takes the example of used cars, there are “peaches” (good cars) and “lemons” (low quality cars). Buyers can’t distinguish between peaches and lemons, but know lemons exist and so offer a price that’s too low for peaches. Sellers who, of course, know their peaches and lemons, remove good cars and retain bad cars. Process continues, and there’s a downward spiral, with market progressively enriched with lemons.

Asymmetric information in a free market could lead to fall in quality and market failure. There’s asymmetric information in healthcare when buying insurance; people are more inclined to purchase when sick. Also, when the physician knows more about quality of product and its need than the patient.Continue reading…

Actually, High-Tech Imaging Can Be High-Value Medicine

Mar 6, 2014• 4

By Karan Chhabra

Lub-SHHRRR. Lub-SHHRRR. Lub-SHHRRR.

“Can you hear it?” she asked with a smile. The thin, pleasant lady seemed as struck by her murmur as I was. She was calm, perhaps amused by the clumsy second-year medical student listening to her heart.

“Yes, yes I can,” I replied, barely concealing my excitement. We had just learned about the heart sounds in class. This was my first time hearing anything abnormal on a patient, though it was impossible to miss—her heart was practically shouting at me.

Her mitral valve prolapse—a fairly common, benign condition—had progressed into acute mitral regurgitation. She came to the hospital short of breath because her faulty valve was letting blood back up into her lungs.

Though it was certainly frightening, surgery to fix the valve could wait a few weeks. But before doing anything, the surgical team wanted a picture of the blood vessels in her heart.

If the picture showed a blockage, the surgeons would have to perform two procedures: one to fix the blockage, and another to fix her valve. If her vessels were healthy, though, the surgeons could use a simpler approach focused just on her valve.

So she came to the interventional cardiologist who was teaching me for the day. Coronary angiograms are the interventionalists’ bread-and-butter procedure, done routinely to look for blockages and to guide stent placement. They involve snaking a catheter from the groin or arm through major blood vessels and up to the heart.

Under fluoroscopy (like a video X-ray), the cardiologists shoot contrast medium into the arteries, revealing the anatomy in exquisite detail.

The images are recorded electronically and accompanied by the cardiologist’s interpretation for anyone else who opens her medical record.

Though routine, these catheterizations aren’t trivial. Whenever you enter a blood vessel, you introduce the risk of bleeding and infection. Fluoroscopy is radiation, and contrast medium can damage the kidneys. And let’s not forget cost—reimbursing the interventional cardiologist, a radiology technician, and nursing staff costs Medicare almost $3,000 per case.

So I asked the cardiologist if such an invasive approach was really necessary.

The Unheard Heart: A Metaphor For Medicine In the Digital Age

Nov 2, 2012• 10

By Clifton Meador, MD

A few months ago, a young cardiologist told me that he rarely listens to hearts anymore. In a strange way, I was not surprised.

He went on to tell me that he gets all the information he needs from echocardiograms, EKGs, MRIs, and catherizations. In the ICU, he can even measure cardiac output within seconds. He told me that these devices tell him vastly more than listening to out-of-date sounds via a long rubber tube attached to his ear.

There was even an element of disdain. He said, “There is absolutely nothing that listening to hearts can tell me that I don’t already know from technology. I have no need to listen. So I don’t do it much anymore.”

I began to wonder. I called my longtime friend and colleague, also a cardiologist. I knew him to be one of the best heart listeners. I asked him if he still listens to hearts. He answered, “Of course I do. I could not practice medicine if I didn’t. But you know every week, several patients tell me when I listen to their hearts that I am the first doctor ever to do that. Can you imagine that?”

Playing the devil’s advocate, I challenged my friend to tell me what he learned from listening to hearts.

He answered, “How could anyone not want to hear those murmurs, sometimes ever so soft, like whispers? Murmurs from the heart, even very faint ones, are trying to tell us significant things. Some sounds are very localized, even hidden or obscured by layers of air. And then there is the rhythm and the beat and the cadence that you cannot hear on the paper strip of the EKG. Also, careful listening is the only way to appreciate the rubs of friction if there are any. The devices are important, but the heart has its own spoken and unspoken language if you know how to listen.