Category Archives: Hype

AI in medicine: 3 easy questions to separate hype from reality

Posted on by • Posted in Artificial Intelligence, Artificial Intelligence Generative, Clinical Research, Genetics, Genomic Medicine, Hype, Reality, Results, Transformation • Leave a comment

https://www.linkedin.com/pulse/ai-medicine-3-easy-questions-separate-hype-from-robert-pearl-m-d–ctznc/

Artificial intelligence has long been heralded as a transformative force in medicine. Yet, until recently, its potential has remained largely unfulfilled.

Consider the story of MYCIN, a “rule-based” AI system developed in the 1970s at Stanford University to help diagnose infections and recommend antibiotics. Though MYCIN showed early promise, it relied on rigid, predetermined rules and lacked the flexibility to handle unexpected or complex cases that arise in real-world medicine. Ultimately, the technology of the time couldn’t match the nuanced judgment of skilled clinicians, and MYCIN never achieved widespread clinical use.

Fast forward to 2011, when IBM’s Watson gained global notoriety by besting renowned Jeopardy! champions Ken Jennings and Brad Rutter. Soon after, IBM applied Watson’s vast computing power to healthcare, envisioning it as a gamechanger in oncology. Tasked with synthesizing data from medical literature and patient records at Memorial Sloan Kettering, Watson aimed to recommend tailored cancer treatments.

However, the AI struggled to provide reliable, relevant recommendations—not because of any computational shortcoming but due to inconsistent, often incomplete, data sources. These included imprecise electronic health record entries and research articles that leaned too heavily toward favorable conclusions, failing to hold up in real-world clinical settings. IBM shut down the project in 2020.

Today, healthcare and tech leaders question whether the latest wave of AI tools—including much-heralded generative artificial intelligence models—will deliver on their promise in medicine or become footnotes in history like MYCIN and Watson.

Anthropic CEO Dario Amodei is among the AI optimists. Last month, in a sprawling 15,000-word essay, he predicted that AI would soon reshape humanity’s future. He claimed that by 2026, AI tools (presumably including Anthropic’s Claude) will become “smarter than a Nobel Prize winner.”

Specific to human health, Amodei touted AI’s ability to eliminate infectious diseases, prevent genetic disorders and double life expectancy to 150 years—all within the next decade.

While I admire parts of Amodei’s vision, my technological and medical background makes me question some of his most ambitious predictions.

When people ask me how to separate AI hype from reality in medicine, I suggest starting with three critical questions:

Question 1: Will the AI solution speed up a process or task that humans could eventually complete on their own?

Sometimes, scientists have the knowledge and expertise to solve complex medical problems but are limited by time and cost. In these situations, AI tools can deliver remarkable breakthroughs.

Consider AlphaFold2, a system developed by Google DeepMind to predict how proteins fold into their three-dimensional structures. For decades, researchers struggled to map these large, intricate molecules—the exact shape of each protein requiring years and millions of dollars to decipher. Yet, understanding these structures is invaluable, as they reveal how proteins function, interact and contribute to diseases.

With deep learning and massive datasets, AlphaFold2 accomplished in days what would have taken labs decades, predicting hundreds of proteins’ structures. Within four years, it mapped all known proteins—a feat that won DeepMind researchers a Nobel Prize in Chemistry and is now accelerating drug discovery and medical research.

Another example is a collaborative project between the University of Pittsburgh and Carnegie Mellon, where AI analyzed electronic health records to identify adverse drug interactions. Traditionally, this process took months of manual review to uncover just a few risks. With AI, researchers were able to examine thousands of medications in days, drastically improving speed and accuracy.

These achievements show that when science has a clear path but lacks the speed, tools and scale for execution, AI can bridge the gap. In fact, if today’s generative AI technology existed in the 1990s, ChatGPT estimates it could have sequenced the entire human genome in less than a year—a project that originally took 13 years and $2.7 billion.

Applying this criterion to Amodei’s assertion that AI will soon eliminate most infectious diseases, I believe this goal is realistic. Today’s AI technology already analyzes vast amounts of data on drug efficacy and side effects, discovering new uses for existing medications. AI is also proving effective in guiding the development of new drugs and may help address the growing issue of antibiotic resistance. I agree with Amodei that AI will be able to accomplish in a few years what otherwise would have taken scientists decades, offering fresh hope in the fight against human pathogens.

Question 2: Does the complexity of human genetics make the problem unsolvable, no matter how smart the technology?

Imagine searching for a needle in a giant haystack. When a single answer is hidden within mountains of data, AI can find it much faster than humans alone. But if that “needle” is metallic dust, scattered across multiple haystacks, the challenge becomes insurmountable, even for AI.

This analogy captures why certain medical problems remain beyond AI’s reach. In his essay, Amodei predicts that generative AI will eliminate most genetic disorders, cure cancer and prevent Alzheimer’s within a decade.

While AI will undoubtedly deepen our understanding of the human genome, many of the diseases Amodei highlights as curable are “multifactorial,” meaning they result from the combined impact of dozens of genes, plus environmental and lifestyle factors. To better understand why this complexity limits AI’s reach, let’s first examine simpler, single-gene disorders, where the potential for AI-driven treatment is more promising.

For certain genetic disorders, like BRCA-linked cancers or sickle cell disease that result from a single-gene abnormality, AI can play a valuable role by helping researchers identify and potentially use CRISPR, an advanced gene-editing tool, to directly edit these mutations to reduce disease risk.

Yet even with single-gene conditions, treatment is complex. CRISPR-based therapies for sickle cell, for example, require harvesting stem cells, editing them in a lab and reinfusing them after risky conditioning treatments that pose significant health threats to patients.

Knowing this, it’s evident that the complications would only multiply when editing multifactorial congenital diseases like cleft lip and palate—or complex diseases that manifest later in life, including cardiovascular disease and cancer.

Put simply, editing dozens of genes simultaneously would introduce severe threats to health, most likely exceeding the benefits. Whereas generative AI’s capabilities are accelerating at an exponential rate, gene-editing technologies like CRISPR face strict limitations in human biology. Our bodies have intricate, interdependent functions. This means correcting multiple genetic issues in tandem would disrupt essential biological functions in unpredictable, probably fatal ways.

No matter how advanced an AI tool may become in identifying genetic patterns, inherent biological constraints mean that multifactorial diseases will remain unsolvable. In this respect, Amodei’s prediction about curing genetic diseases will prove only partially correct.

Question 3: Will the AI’s success depend on people changing their behaviors?

One of the greatest challenges for AI applications in medicine isn’t technological but psychological: it’s about navigating human behavior and our tendency toward illogical or biased decisions. While we might assume that people will do everything they can to prolong their lives, human emotions and habits tell a different story.

Consider the management of chronic diseases like hypertension and diabetes. In this battle, technology can be a strong ally. Advanced home monitoring and wearable devices currently track blood pressure, glucose and oxygen levels with impressive accuracy. Soon, AI systems will analyze these readings, recommend diet and exercise adjustments and alert patients and clinicians when medication changes are needed.

But even the most sophisticated AI tools can’t force patients to reliably follow medical advice—or ensure that doctors will respond to every alert.

Humans are flawed, forgetful and fallible. Patients skip doses, ignore dietary recommendations and abandon exercise goals. On the clinician side, busy schedules, burnout and competing priorities often lead to missed opportunities for timely interventions. These behavioral factors add layers of unpredictability and unresponsiveness that even the most accurate AI systems cannot overcome.

And in addition to behavioral challenges, there are biological issues that limit the human lifespan. As we grow older, the protective caps on our chromosomes wear down, causing cells to stop functioning. Our cells’ energy sources, called mitochondria, gradually fail, weakening our bodies until vital organs cease to function. Short of replacing every cell and tissue in our bodies, our organs will eventually give out. And even if generative AI could tell us exactly what we needed to do to prevent these failings, it is unlikely people would consistently follow the recommendations.

For these reasons, Amodei’s boldest prediction—that longevity will double to 150 years within a decade—won’t happen. AI offers remarkable tools and intelligence. It will expand our knowledge far beyond anything we can imagine today. But ultimately, it cannot override the natural and complex limitations of human life: aging parts and illogical behaviors.

In the end, you should embrace AI promises when they build on scientific research. But when they violate biological or psychological principles, don’t believe the hype.

C.D.C. Tells States How to Prepare for Covid-19 Vaccine by Early November

Posted on by • Posted in 2020 Election Issues, Campaign, Centers for Disease Control & Prevention (CDC), Coronavirus, Credibility, Crisis, Crisis Management, Distribution, Election, Evidence-based Leadership, Evidence-Based Plan, Evidence-Based Policy, Expert, Expert Opinion, Health Policy, Hype, Mortality, Pandemic, Risk, Risk Assessment, Risk Management, Risk Tolerance, Scientists, Supply Chain, Transparency, Trust, Truth, Vaccine, Vaccine Distribution, Wishful Thinking • Leave a comment

As President Trump pushes the possibility of a vaccine this year, the C.D.C. has outlined technical scenarios to state public health officials for an unidentified Vaccine A and Vaccine B.

The Centers for Disease Control and Prevention has notified public health officials in all 50 states and five large cities to prepare to distribute a coronavirus vaccine to health care workers and other high-risk groups as soon as late October or early November.

The new C.D.C. guidance is the latest sign of an accelerating race for a vaccine to ease a pandemic that has killed more than 184,000 Americans. The documents were sent out on the same day that President Trump told the nation in his speech to the Republican National Convention that a vaccine might arrive before the end of the year.

Over the past week, both Dr. Anthony S. Fauci, the country’s top infectious disease expert, and Dr. Stephen Hahn, who heads the Food and Drug Administration, have said in interviews with news organizations that a vaccine may be available for certain groups before clinical trials have been completed, if the data is overwhelmingly positive.

Public health experts agree that agencies at all levels of government should urgently prepare for what will eventually be a vast, complex effort to vaccinate hundreds of millions of Americans. But the possibility of a rollout in late October or early November has heightened concerns that the Trump administration is seeking to rush the distribution of a vaccine — or simply to hype that one is possible — before Election Day on Nov. 3.

For an administration that has struggled with the logistical challenges of containing the coronavirus, the distribution of millions of vaccines that must be stored in subzero temperatures and provided first to high-risk groups through America’s flawed, fragmented health care system would be a daunting challenge. Even the C.D.C.’s guidance acknowledged that its plan was hypothetical and based on the need to immediately begin organizing the gigantic effort that would be required if the F.D.A. were to allow the use of a vaccine or two this year.

The C.D.C. plans lay out technical specifications for two candidates described as Vaccine A and Vaccine B, including requirements for shipping, mixing, storage and administration. The details seem to match the products developed by Pfizer and Moderna, which are the furthest along in late-stage clinical trials. On Aug. 20, Pfizer said it was “on track” for seeking government review “as early as October 2020.”

Credit…

“This timeline of the initial deployment at the end of October is deeply worrisome for the politicization of public health and the potential safety ramifications,” said Saskia Popescu, an infection prevention epidemiologist based in Arizona. “It’s hard not to see this as a push for a pre-election vaccine.”

Three documents were sent to public health officials in all states and territories as well as officials in New York, Chicago, Philadelphia, Houston and San Antonio on Aug. 27. They outlined detailed scenarios for distributing two unidentified vaccine candidates, each requiring two doses a few weeks apart, at hospitals, mobile clinics and other facilities offering easy access to the first targeted recipients.

The guidance noted that health care professionals, including long-term care employees, would be among the first to receive the product, along with other essential workers and national security employees. People 65 or older, as well as Native Americans and those who are from “racial and ethnic minority populations” or incarcerated — all communities known to be at greater risk of contracting the virus and experiencing severe disease — were also prioritized in the documents.

That’s a positive development, “so it doesn’t just all wind up in high-income, affluent suburbs,” said Dr. Cedric Dark, an emergency medicine physician at Baylor College of Medicine in Texas.

The C.D.C. noted in its guidance that “limited Covid-19 vaccine doses may be available by early November 2020.” The documents were dispatched the same day that Dr. Robert Redfield, director of the C.D.C., sent a letter to governors asking them to prepare vaccine distribution sites by Nov. 1, as McClatchy reported.

The agency also said its plans were as yet hypothetical, noting, “The Covid-19 vaccine landscape is evolving and uncertain, and these scenarios may evolve as more information is available.” A C.D.C. spokeswoman confirmed that the documents were sent but declined to comment further.

Many of the details listed for the two vaccines — including required storage temperature, the number of days needed between doses, and the type of medical center that can accommodate the product’s storage — match what Pfizer and Moderna have said about their products, which are based on so-called mRNA technology. Neither company responded to requests for comment.

The scenarios, which assume that the two vaccines will demonstrate sufficient safety and effectiveness for an emergency authorization from the F.D.A. by the end of October, noted that Vaccine A, which seems to match Pfizer’s, would have about two million doses ready within this time frame, and that Vaccine B, whose description matches Moderna’s, would have about one million doses ready, with tens of millions of doses of each vaccine ready by the end of the year. Although it’s possible that some promising preliminary data may emerge by the end of October, experts are skeptical.

“The timeline that’s reported seems a bit ambitious to me,” Dr. Dark said. “October’s like 30 days away.”

Trials that test a vaccine’s effectiveness can take years to yield reliable results. It’s possible to draw conclusions sooner “if there is an overwhelming effect” in which vaccinated people appear to be far better protected from disease, said Padmini Pillai, a vaccine researcher and immunologist at M.I.T.

But there can be significant risks in approving a vaccine for broad use in the public before Phase 3 clinical trials involving tens of thousand of participants are completed. Rare but dangerous side effects may only surface over time, after such large numbers of people have received the vaccine.

And data gathered early in a trial might not hold true months down the line. Researchers also need time to test large numbers of people from a variety of backgrounds to determine how well the vaccine works in different populations — including the vulnerable communities identified in the guidelines.

Should any of these snags occur, Dr. Pillai said, “all of this together could diminish public trust in the vaccine.”

James S. Blumenstock, senior vice president of pandemic response and recovery at the Association of State and Territorial Health Officials, confirmed that the three C.D.C. documents were sent to all state and territorial health departments last week. “It is now the time to enhance organizational structure and involve all partners in this planning process going forward,” he said.

Lisa Stromme, a spokeswoman for the Washington State Department of Health, said that her state’s health officials were still at “a very early stage in a planning process,” but were already working toward developing infrastructure that would accommodate the assumptions laid out by the C.D.C.

The C.D.C. documents said that public health administrators should review lessons learned from the 2009 H1N1 pandemic vaccination campaign, which did not have enough doses at the beginning to meet demand.

“It’s good to have a plan out for hospitals and health care systems to prepare” for a potential rollout, said Dr. Taison Bell, a pulmonary and critical care physician at the University of Virginia. But Dr. Bell added that he was concerned that the timeline outlined in the documents “is incredibly ambitious and makes me worry that the administration will prioritize this arbitrary deadline rather than maintaining diligence with following the science.”

The technical comparison of Vaccine A and Vaccine B has some echoes of what was discussed at an Aug. 26 meeting of the Advisory Committee on Immunization Practices of the C.D.C. At the meeting, Dr. Kathleen Dooling, a C.D.C. medical officer, laid out three scenarios: Vaccine A, or the Pfizer vaccine, is approved, Vaccine B, the Moderna vaccine, is approved, or both. The requirement that Pfizer’s vaccine be stored at minus 70 degrees Celsius would mean that it couldn’t be administered at most small sites, she said. The C.D.C. documents noted that orders of Vaccine A would go “to large administration sites only.” The Moderna vaccine requires storage at minus 20 degrees Celsius.

The C.D.C. documents said the vaccine would be free to patients, but that providers might not be reimbursed for administrative costs if the vaccine was given an emergency authorization, rather than a standard approval.

Experts worry that the process is unlikely to go off without a hitch, given the last-minute scramble and the mixed messaging so far. “I think distribution is going to be very tricky for the vaccine, particularly if there is a cold storage requirement,” Dr. Bell said.

There are also likely to be challenges administering both doses of the proposed vaccines, which must be given weeks apart, Dr. Dark said. “How are you going to make sure people get both?”