AI and Public Health Series: Part 2: How AI Can Help Tackle Major Causes of Suffering and Death
Artificial intelligence (AI) and machine learning (ML) drive many major medical advances today. Part 1 of this series identified how algorithmic systems are helping us expand and better understand the growing corpus of medical knowledge. It also explained how AI is helping doctors and nurses improve the quality of care they deliver to patients.
Part 2 explores how data science, algorithmic systems, and advanced robotics are facilitating earlier detection and treatment of specific ailments and diseases that cause considerable suffering and death each year in America.
How AI Can Help Address the Major Killers
The U.S. Centers for Disease Control and Prevention (CDC) publishes annual data for the leading causes of death among Americans. The following table shows data for 2022.
Medical professionals are already tapping the power of AI, ML, and robotic technologies to fight these killers. For example:
- Heart attack detection and treatment: After decades of steady decline, heart failure mortality rates leveled off and are now worsening. According to the CDC, just over 700,000 Americans died from heart diseases in 2022. AI and ML tools are being developed to help detect and treat heart disease and heart attacks. In 2022, scientists at Cedars-Sinai developed an algorithmic tool that can quantify coronary plaque buildup in five to six seconds compared to (at least) 25 to 30 minutes before. This will greatly improve the ability to predict who will have a heart attack. Other researchers have developed AI tools to improve personalized treatment for women who have had heart attacks. Women who suffer a heart attack have a higher mortality rate than men, often because their symptoms are misunderstood or misdiagnosed. Meanwhile, the British National Health Service recently started using a new AI tool that can detect heart disease in just 20 seconds as patients undergo magnetic resonance imaging, compared with the 13-plus minutes it usually takes to analyze images manually after performing a scan.
- Cancers: President Richard Nixon declared a national “war on cancer” over 50 years ago. More recently, the Obama and Biden administrations pushed for a “cancer moonshot.” Unfortunately, cancer remains the second leading cause of death in the United States, claiming over 600,000 lives in 2022. Pancreatic cancer is the third leading cause of cancer deaths, claiming 49,830 lives in 2022. AI/ML-enabled technologies are poised to help reduce this staggering death toll. Mayo Clinic researchers have shown how ML models can help diagnose and treat pancreatic cancer at an earlier stage. Other ML models can already identify high-risk cancer patients years in advance. British scientists recently reported on new AI software that can spot signs of pre-cancer during endoscopies in 92 percent of patients, which could significantly lower deaths from esophageal cancer. Aided by increasingly personalized screening techniques, AI/ML technologies are also involved in the early detection and treatment of lung cancer, breast cancer, brain cancer, cervical cancer, and many other types of cancer (including undiagnosable cancers). The U.S. Food and Drug Administration is approving more AI-powered medical devices to help facilitate early detection of the most prevalent cancers. AI-enabled cancer detection tools can also help reduce the workload faced by human radiologists and cancer doctors. Major AI model developers like Google, Microsoft, and OpenAI have recently developed multimodal generative AI systems to advance cancer detection and treatment.
- Sepsis and superbugs: Recent medical studies document how AI-powered monitoring systems help detect and address antimicrobial resistance (AMR) in “superbugs” and sepsis (when the body reacts extremely negatively to infection). Roughly 1.7 million adults develop sepsis each year in the United States, and 350,000 of them die. Meanwhile, researchers find that AI “dramatically cuts the time it takes to sort through thousands of promising compounds” to fight drug-resistant pathogens. A group of Australian researchers recently used ML techniques to identify over 860,000 promising antimicrobial peptides—naturally occurring antibiotic sources of immunity that could help kill or block the growth of bacteria that cause dangerous and difficult-to-treat infections. According to a recent study, AMR could become the primary cause of death around the world by 2050, leading to approximately 10 million deaths annually.
- Strokes and brain diseases: The CDC reports that someone has a stroke every 40 seconds in the United States and that more than 795,000 Americans suffer a stroke each year. Scientists are developing AI/ML-powered systems to detect strokes quicker and to diagnose degenerative brain diseases like Alzheimer’s, dementia, and Parkinson’s. One ML model helped researchers accurately identify Parkinson’s disease in 79 percent of individuals up to seven years before symptom onset.
Addressing Other Ailments and Diseases
These examples highlight how AI/ML technologies can help address the leading causes of death. But this only scratches the surface in terms of how algorithmic and robotic technologies can address major ailments. Consider some additional examples:
- Paralysis and disabilities: The combination of algorithmic and robotic technologies hold promise for helping paralyzed individuals regain certain motor functions. The Christopher & Dana Reeve Foundation has estimated that nearly one in 50 Americans lives with paralysis. In May 2023, a Dutch man paralyzed from the waist down for more than a decade regained his ability to walk thanks to brain and spine implants and an AI-enabled thought decoder that helped translate electrical brain signals into muscle movement. A paralyzed American man regained his sense of touch and mobility thanks to similar AI-enabled brain implants. AI/ML capabilities power other brain-machine implants that are helping to address disabilities in other ways, such as by restoring a person’s ability to speak after a stroke or other ailments. Rep. Jennifer Wexton (D-Va.)—who lost the use of her voice due to progressive supranuclear palsy, a disease affecting over 30,000 Americans—recently delivered a House floor speech using an AI-enabled text-to-speech tool from ElevenLabs that reproduced her voice from old clips of past speeches. Wexton has stated that AI “can provide new, unimaginable and life-changing opportunities for Americans with disabilities.” The New York Times recently documented how a woman who lost her arm in an accident is able to control her new prosthetic robotic arm thanks to advances in AI and sensors embedded in her body. The transformative potential of brain-machine interfaces and neurotechnology will be explored in a future installment of this series.
- Mental health and drug addiction: AI can help identify and address mental health problems through textual analysis, which can supplement human-based analysis at a time when there is a nationwide shortage of health care workers in this area. A 2023 literature review of studies on ML methods used to help predict schizophrenia and bipolar disorders (BD) found that “ML can precisely predict results and assist in making clinical decisions‐concerning schizophrenia and BD.” AI tools are also being tapped to assist in finding novel drugs that can help counter opioid addiction.
- Ultrasounds, X-rays, and electroencephalograms (EEGs): AI-powered ultrasounds are helping identify mothers with a potential risk of complications during pregnancy and helping save babies through more widespread and cost-effective testing in developing nations. New AI tools are also used to better identify normal versus abnormal chest X-rays, reducing the heavy workload radiologists deal with as demand for medical imaging increases. One doctor and co-author of a 2023 study on how AI improves radiology said, “[W]e could not find a single chest X-ray in our database where the algorithm made a major mistake. Furthermore, the AI tool had a sensitivity overall better than the clinical board-certified radiologists.” Similar advances are happening with AI-enabled EEGs, which improve patient care in areas without enough trained neurology workers or systems to handle the volume of EEG evaluations.
- Organ donation: One of AI’s great success stories in the field of organ donation is paired kidney donation, where the technology helps doctors and patients by “tak[ing] an incredibly complex problem and solv[ing] it faster and with fewer errors than humans can,” saving more lives as a result. According to the U.S. Health Resources and Services Administration, over 100,000 Americans are currently on the national transplant waiting list.
- Eye disease and blindness: AI/ML is powering new types of “assistive technology” for people with disabilities, especially those with impaired vision. It is also helping detect and address eye disease and blindness. In April 2024, the National Institutes of Health announced a breakthrough in AI retinal imaging that produces high-resolution images of cells in the eye 100 times faster and with a 3.5-fold improvement in image contrast. Meanwhile, ChatGPT-4 is able to answer 70 percent of questions correctly on ophthalmic imaging tests, demonstrating that AI systems are getting much better at eye imaging.
As noted in Part 1 of this series, AI/ML is powering other advanced learning capabilities that will help doctors and scientific researchers access and understand massive amounts of patient and health data and put it to even better use. These same capabilities will help innovators create new personalized health monitoring and tracking systems for the public. AI is also facilitating new drug and vaccine discovery, which will be discussed in a future installment of this series. AI will become crucial for various surgeries as well, improving outcomes when operations are necessary (often through robotic-assisted surgery) or, better yet, avoiding the need for invasive procedures altogether.
Conclusion: Embrace Innovation in Both Technology and Policy
These examples illustrate how algorithmic technologies could help society address serious ailments and chronic killers that have proven hard to counter despite decades of effort and considerable spending. These algorithmic capabilities can expand so long as America keeps the door open to ongoing experimentation, innovation, and investment. Balanced public policies will be crucial in accomplishing that goal. If lawmakers hope to facilitate more health-improving and life-saving innovations, they must leave risk-averse regulatory schemes behind and embrace policies that open the door to further experimentation in AI-enabled medicine.