AI diagnosis to make medical decisions is just about here


AP Photo/M. Spencer Green

❝ The US Food and Drug Administration approved this week the first software powered by artificial intelligence that replaces the need for a specialized doctor to interpret medical imagery.

The software is called IDx-DR, made by diagnostic AI startup IDx, and specifically analyzes images of the retina to detect whether a person with diabetes has a complication from the disease called diabetic retinopathy…

❝ Diabetic retinopathy is a complication of diabetes where blood sugar damages the back of the eye, according to the FDA, and is the main cause of the loss of vision for those with diabetes…

By allowing this software to be marketed in the US, the FDA is setting a bar for the accuracy needed in order for AI to take over for human doctors. When validating that the AI system worked, the FDA used images from 900 US patients. The software correctly detected more than mild diabetic retinopathy 87.4% of the time, and identified when patients did not have more than mild retinopathy 89.5% of the time. Accuracy for humans naturally varies from doctor to doctor, but for the FDA to approve the technology it “must provide for more effective treatment or diagnosis of a life-threatening or irreversibly debilitating disease or condition.”

No doubt a predictable percentage of Americans will demonstrate fear of this technology to a greater degree than any other educated nation. Part of that education and, more important, political processes, electoral politics, religious folderol, come together to work harder than anywhere else – to keep citizens from modernizing their lives and thinking. Why – we might even question authority.

Are doctors ready for body cams?


Google Glass makes it into the surveillance society

Maybe you remember the famous video by Simons and Chabris. Two groups of students, one in white shirts and the other in black shirts, are passing a basketball around. You are asked to watch the video and count the number of passes made by one of the teams. You proudly count 13 (the actual number is 18). But what you didn’t notice, during all of your counting, was that midway through the video, a gorilla walked straight through the middle of the scene. Indeed about half of individuals tested in the original study missed the gorilla.

A red trauma victim is brought into the ED trauma bay by EMS. The lead paramedic provides details about the crash scene, the patient’s health status, and gives a point-by-point report about the prehospital care. Too bad that only 36% of the key information was accurately remembered by the receiving ED group.

What’s happening here?

These two examples highlight how medical care can be perceived differently, and maybe even contradicted, by doctors and patients. We aren’t aware of something we have missed — like the gorilla. You only see things you are focusing attention on. Have you ever had a patient complain “the doctor didn’t even examine my stomach” when you have performed, and documented, several serial exams? How many times have you been asked by a patient “When am I going to see the doctor?” when you’ve already had several conversations and introduced yourself as THE DOCTOR. Or, are perplexed by a family display of great disbelief when informed that their loved one is sliding towards the end of life.

We think we perceive and remember more of the world than we actually do, and different people experience the same inputs differently. We don’t see, hear, and remember alike. Hearing is passive, but listening requires concentration and focus to understand the meaning of another’s words.

Jeremy Brown has identified lots of examples where a med-cam can provide an objective view of medical reality — a sort of enhanced photojournalism — where the picture tells the truth. But we need to be ready to have our own behaviors and communications on display. After all, what’s good for the patient should be good for the doctor, too.

Questions of what’s private and what isn’t used to be decided essentially by what’s public and what isn’t. Starting with the obvious – like body cams for coppers – I can see where record-keeping is going. Not only for accurate answers to recurrent questions in an ER; but, the lawyers on retainer for the hospital are going to want this kind of recorded observation to keep a handle on liability.

The feeling changes. Your relationship with your employer changes. Doctors especially feel they’re part of the management team – even in large-economy-size urban hospital complexes. That feeling will change under observation. As much as useful qualities like those described in this article may be – in our society it’s easy to worry about office politics, petty jealousy becoming equally important to some administrator you consider to be an ignorant ass.

Dr. Google is as popular as ever — can real doctors adapt?

It’s easy to dismiss the Internet as a risky place to look for health information. As HealthTap founder Ron Gutman joked the first time we met, ”On the Internet, every headache becomes a brain tumor in four clicks or less.”

If you’ve ever done an online search for an unfamiliar ache, you can probably relate: That weird pain in your side could mean appendicitis, food poisoning or pregnancy. That nasty rash on your arm could be poison ivy, a spider bite or cancer.

But despite “Dr. Google’s” shortcomings and concerns about so-called cyberchondria, the Web – and search engines in particular – remains a top destination for people seeking out health information. The Pew Internet & American Life Project this week reported that about a third of U.S. adults have gone online to look for health information. And, eight in 10 Internet users say their last health-related search began with a search engine – a figure that has not changed since Pew last asked that question in 2000, despite the rise of social media, health-specific content sites and startups.

The report also found that those health searchers are reaching diagnoses that their doctors disagree with about one-fifth of the time…

A few studies have attempted to evaluate the reliability of search engines but with mixed conclusions…On one hand, the Web can help direct people to valuable information and studies that even their doctors may not be aware of. But search engines alone don’t give people enough ways of gauging a source’s reliability or providing the context they may need to make the most of sources that are actually good…

Tools that connect doctors with patients in HIPAA-compliant digital environments are growing – HealthTap, for example, helps patients directly ask doctors questions online, and Ringadoc lets people consult physicians via video conference. But they’re just beginning to appeal to doctors who are willing to define their roles and organize their time differently.

As mobile adoption grows and digital natives age, a doctor willing to email you and curate online information isn’t just going to be a nice to have — for many, it will be a need to have.

That’s Heussner’s conclusion. The majority of doctors I’ve run into in recent years have been skeptical of absolutely anything I might learn online. With two exceptions. One is a dentist I’ve been seeing as needed for over a couple of decades. He knows me pretty well, knows my interest in science and medicine and the sources I read. The other is more recent – my dermatologist. But, he’s as much of a geek as I am and we share discussions on new findings in medicine in general. Along with geek topics, politics and the future of humankind. A delight.

The rest of the time? The response usually runs 110% negative.

More stringent definition of autism may exclude many

Proposed changes in the definition of autism would sharply reduce the skyrocketing rate at which the disorder is diagnosed and might make it harder for many people who would no longer meet the criteria to get health, educational and social services…

The definition is now being reassessed by an expert panel appointed by the American Psychiatric Association, which is completing work on the fifth edition of its Diagnostic and Statistical Manual of Mental Disorders, the first major revision in 17 years. The D.S.M., as the manual is known, is the standard reference for mental disorders, driving research, treatment and insurance decisions. Most experts expect that the new manual will narrow the criteria for autism; the question is how sharply.

The results of the new analysis are preliminary, but they offer the most drastic estimate of how tightening the criteria for autism could affect the rate of diagnosis. For years, many experts have privately contended that the vagueness of the current criteria for autism and related disorders like Asperger syndrome was contributing to the increase in the rate of diagnoses — which has ballooned to one child in 100, according to some estimates.

The psychiatrists’ association is wrestling with one of the most agonizing questions in mental health — where to draw the line between unusual and abnormal — and its decisions are sure to be wrenching for some families. At a time when school budgets for special education are stretched, the new diagnosis could herald more pitched battles. Tens of thousands of people receive state-backed services to help offset the disorders’ disabling effects, which include sometimes severe learning and social problems, and the diagnosis is in many ways central to their lives. Close networks of parents have bonded over common experiences with children; and the children, too, may grow to find a sense of their own identity in their struggle with the disorder…

The new analysis, presented Thursday at a meeting of the Icelandic Medical Association, opens a debate about just how many people the proposed diagnosis would affect…

That the qualitative expansion in diagnoses of autism mostly reflects doctors and parents and a culture growing in parallel that found itself able to get federal and public support for difficult behavioral problems – by defining those problems as autism – ain’t new. That quantifying levels of impairment and reviewing the premises of diagnosis is finally happening is the only surprise.

Doesn’t make the needs of parents and offspring any less or diminish society’s responsibility. It’s just that that, too, has become a political question in 21st Century America.

New technique for diagnosing Alzheimer’s

It may soon be possible to obtain a highly accurate diagnosis of Alzheimer’s disease by analyzing a sample of spinal fluid. A study released Monday found that a constellation of three substances in the cerebrospinal fluid was present in 90% of people who had been diagnosed with Alzheimer’s.

The test also showed the same markers were found in 72% of people with mild cognitive impairment, considered an early stage of the disease, and in one-third of adults who had no cognitive problems.

Many experts believe that biomarkers in spinal fluid may emerge as the most accurate diagnosis of Alzheimer’s disease. At present, the disease is diagnosed using pencil-and-paper cognitive tests, which are subjective and may be inaccurate. The diagnosis can only be confirmed by examining brain tissue at an autopsy…

In an editorial accompanying the paper, two U.S. experts in the disease said that cerebrospinal fluid analysis should be put into wider practice. “There is now ample evidence that these measurements have value; physicians need to formulate when and how to incorporate cerebrospinal fluid measurements into practice,” they wrote.

Hopefully, the indicators will provide a pathway to a cure – at least symptomatic relief.

Genomic revolution only just starting

The 10-year-old Human Genome Project has only just begun to bring to fruition its promise to transform medicine…

Francis Collins, who led the U.S. component of the project and is now director of the U.S. National Institutes of Health, said that although it may seem that the revolution promised with the publication of the first draft in 2000 is slow in coming, many early predictions had been prematurely hyped.

Isn’t that a pundit-based specialty?

Scientists have barely scratched below the surface of the possibilities opened up by having access to the whole human gene map, he said, and when they do, their results will determine the way all people are diagnosed and treated for diseases.

“It’s fair to say that most people have not yet had the experience of having their personal medical care directly affected by the sequencing of the human genome,” Francis told a briefing in London marking the project’s 10-year anniversary.

“So while one might argue that the consequences have not come across in the first 10 years in the most dramatic form that some predictors put forward in the year 2000, I think the predictions … were probably a bit overblown.”

Mike Stratton, another of the project’s founders and now director of Britain’s influential Sanger Institute, pointed to several areas of disease where big medical advances had already come about thanks to the ability to read the map of human life.

Cancer drugs, like so-called BRAF inhibitors for malignant melanoma skin cancers — versions of which are being developed by drugmakers including Switzerland’s Roche and Britain’s GlaxoSmithKline — were examples how quickly gene sequencing had given birth to targeted treatments, he said…

The genome founders also noted that scientists had already found more than 800 genetic variants that play a role in risks of common illnesses like heart disease, cancers and diabetes…

“When a truly transformative advance occurs in science, inevitably there will be in the short term an overly optimistic set of predictions,” he said. “But in the long term…the consequences will turn out to have been underestimated. I think that will…be true of the Human Genome Project.

Couldn’t agree more. We always hope for more than just plain good news. In part, let’s face it, because of our mortality.

But, real science requires proofs and testing, peer review and more testing. There is no easy way around sound methods. And, of course, if results aren’t forthcoming quickly enough to satisfy Reality TV and beancounters – whining is the result.

Simple urine test may identify autistic children for early treatment

Children with autism have a different chemical fingerprint in their urine than non-autistic children, according to new research published in the Journal of Proteome Research. The researchers behind the study, from Imperial College London and the University of South Australia, suggest that their findings could ultimately lead to a simple urine test to determine whether or not a young child has autism.

Autism affects an estimated one in every 100 people in the UK. People with autism have a range of different symptoms, but they commonly experience problems with communication and social skills, such as understanding other people’s emotions and making conversation and eye contact.

People with autism are also known to suffer from gastrointestinal disorders and they have a different makeup of bacteria in their guts from non-autistic people.

Today’s research shows that it is possible to distinguish between autistic and non-autistic children by looking at the by-products of gut bacteria and the body’s metabolic processes in the children’s urine. The exact biological significance of gastrointestinal disorders in the development of autism is unknown…

At present, children are assessed for autism through a lengthy process involving a range of tests that explore the child’s social interaction, communication and imaginative skills. Early intervention can greatly improve the progress of children with autism but it is currently difficult to establish a firm diagnosis when children are under 18 months of age, although it is likely that changes may occur much earlier than this.

Bravo! A new beginning – from a new direction.

Another new affordable test that may lessen the economic burden of health care in many nations.

You never realized a salad spinner was a centrifuge?

A simple salad spinner will save lives this summer, if everything goes as planned by two Rice University undergraduates.

The spinner has been turned, so to speak, into a rudimentary centrifuge that medical clinics in developing countries can use to separate blood without electricity.

Rice sophomore Lila Kerr and freshman Lauren Theis will take their Sally Centrifuge abroad for nearly two months this summer as part of Beyond Traditional Borders (BTB), Rice’s global health initiative that brings new ideas and technologies to underdeveloped countries. Kerr will take a spinner to Ecuador in late May, Theis will take one to Swaziland in early June and a third BTB team will take one to Malawi, also in June. Such field testing is important to Rice students as they develop a range of tools to enhance global health.

Kerr and Theis are minoring in global health technologies and took the Introduction to Bioengineering and World Health class taught by Rebecca Richards-Kortum…director of Rice 360˚: Institute for Global Health Technologies…

“We were essentially told we need to find a way to diagnose anemia without power, without it being very costly and with a portable device,” added Theis, a political science major and native of San Antonio, Texas.

They found that a salad spinner met those criteria. When tiny capillary tubes that contain about 15 microliters of blood are spun in the device for 10 minutes, the blood separates into heavier red blood cells and lighter plasma. The hematocrit, or ratio of red blood cells to the total volume, measured with a gauge held up to the tube, can tell clinicians if a patient is anemic. That detail is critical for diagnosing malnutrition, tuberculosis, HIV/AIDS and malaria…

The centrifuge, assembled using plastic lids, cut-up combs, yogurt containers and a hot-glue gun, costs about $30 in parts, including the spinner.

Bravo! Another successful field expedient.

Scientists say MEG scan can detect PTSD

Post-traumatic stress is estimated to afflict more than 300,000 veterans of Iraq and Afghanistan, but until now, it’s been labeled a “soft disorder” — one without an objective biological path to diagnosis.

That may have changed this week, after researchers at the University of Minnesota and the Minneapolis VA Medical Center announced they’d found a distinct pattern of brain activity among PTSD sufferers.

The team used magnetoencephalography (MEG), a brain imaging method that measures how the brain processes information.

They scanned the brains of 74 U.S. veterans with PTSD, and 250 civilians without the disorder, and say that by spotting specific brain biomarkers, they managed to accurately diagnose PTSD sufferers with 90 percent accuracy.

The study could be a breakthrough for the military, which has been scrambling to address a surge in post-traumatic symptoms among newly returning vets. Right now, troops are evaluated by mental health experts, but diagnosis is a crapshoot: Symptoms can take years to show up, and they vary from person to person, even among those exposed to the same traumas.

Of course, a study of 74 vets is only a start. Next up, the researchers want to evaluate 500 vets, alongside 500 civilians, to further validate their findings.

It’s a start. I lived and worked with PTSD-afflicted veterans all the way back to WW2 – through a long period when the Veterans Administration wouldn’t even admit it was an ailment.

Overdue.