Using artificial intelligence (AI) in healthcare may seem new, but the technology has actually been around for decades. Research suggests the earliest incarnation of AI—the simulation of human intelligence in computers—dates back to the 1950s. Granted, the limitations of early models prevented widespread acceptance, not to mention application, in the world of medicine.
By the early 2000s, however, AI began to make good on its early promise. Healthcare workers could use AI to screen for diseases ranging from diabetic retinopathy to skin cancer with amazing accuracy. And the U.S. Food and Drug Administration (FDA) approved a number of AI firsts: the first AI-assisted device for use in the operating room (OR), the first AI-assisted device for cancer diagnosis, a deep-learning algorithm to interpret brain MRIs. To date, the FDA has authorized more than 1400 AI/machine learning-enabled medical devices in the United States.
“We were early adopters of data science and intelligent automation, which has given us more than a decade of learnings that prepared us to apply the next wave of technologies,” says Jim Swanson, Executive Vice President and Chief Information Officer, Johnson & Johnson. “Our global scale lets us bring innovations—like AI-assisted drug discovery, anatomy mapping and clinical trial recruitment—to patients worldwide. We approach AI with the same focus we bring to all innovation: using it responsibly to solve meaningful healthcare challenges and improve outcomes for patients.”
Here are six ways Johnson & Johnson is using AI to help advance healthcare.
1.
In the OR, surgical video captured during procedures can provide medical professionals with opportunities for education, research methodologies and quality-improvement strategies.
Johnson & Johnson is developing digital solutions for the OR that use AI algorithms to essentially “cut a highlight reel of these videos” in a matter of minutes, says Shan Jegatheeswaran, Global Vice President, MedTech Digital, Johnson & Johnson. That way, surgeons can re-watch significant moments from their procedures. Without AI, this process could take hours—even days—to complete.
“Surgeons are a lot like high-performance athletes,” says Jegatheeswaran. “New and learning surgeons want to see how they performed and learn from their performances and how others performed. But it’s a lot of work to sit and watch hours of footage from the full procedure and cut it down to clips.”
The company’s Polyphonic™ digital ecosystem connects surgical data, video, devices and AI insights into a single surgical intelligence layer so that healthcare teams can collaborate more effectively. Its open approach allows a broad range of partners, applications and innovations to work together, helping break down longstanding silos and enabling insights to flow across the surgical journey.
With Polyphonic, surgeons will also be able to analyze patient data ahead of surgery, help with presurgical planning and provide real-time guidance via telepresence. Post-op, the platform is designed to allow surgeons to track recovery by linking with patients’ electronic health records.
“Once you get enough of these enriched surgical videos, there is potential to explore algorithms on the behaviors, tactics and movements that could in the future provide further information to care teams,” Jegatheeswaran says.
Drug discovery is inherently complex, and many of today’s most promising targets have long been considered ‘hard to drug’ or undruggable. AI gives us a powerful way to unlock these high value opportunities and open new paths for patients.
2.
Not only is AI helping to improve procedures—by extension, it’s improving patient outcomes as well.
Cardiac ablation is one example. This minimally invasive procedure delivers energy through a catheter to correct an irregular heartbeat in patients with heart conditions like atrial fibrillation (AFib). During cardiac ablation, a small flexible tube (a.k.a. catheter) is inserted into a vein in the groin and guided through the veins and/or arteries up into the heart, where the location of the abnormal tissue that is responsible for the arrhythmia is identified.
The company’s CARTO™ 3 System enables electrophysiologists to create 3D maps and navigate inside the heart during AFib catheter ablation procedures. The system’s newest Version 8 software incorporates deep learning: a subset of AI capabilities designed to help doctors deliver effective and efficient procedures and generate valuable case data that may help further improve patient outcomes.
3.
Traditionally, discovering and developing new drugs to treat disease is a long and complex undertaking due to the enormous complexity of biology, and this effort is primed to be transformed by AI.
To develop medicines, researchers need to understand which biological and genetic variations cause diseases. By applying AI to de-identified medical datasets, such as electronic health records, lab results or genetic sequencing data, and datasets from partnered biobanks and scientific consortiums, such as UK Biobank, scientists can shed light on what drives specific diseases.
Once promising disease targets are identified, Johnson & Johnson is then leveraging AI to help discover molecules that could interact with the target. AI can assist in the identification of starting points, design and optimize those molecules into game-changing drug candidates that are fit to combat the disease and minimize side effects. The company is doing this through its own state-of-the-art technologies, as well as in partnership with external collaborators.
“Drug discovery is inherently complex, and many of today’s most promising targets have long been considered ‘hard to drug’ or undruggable. AI gives us a powerful way to unlock these high value opportunities and open new paths for patients,” says Justin Scheer, Vice President, In Silico Discovery, Innovative Medicine R&D, Johnson & Johnson. “By applying AI, we can identify promising leads more quickly and optimize them to improve their chances of success in the clinic. Using our highly sophisticated AI, built on in-house data, we aim to accelerate the delivery of medicines for patients who are waiting.”
4.
One of the biggest challenges when it comes to running clinical trials is quickly and efficiently recruiting and enrolling patients that meet the selection criteria. Adopting AI technology into the process may help solve this problem. It can also help researchers more precisely measure how patients respond to treatment once a trial is underway.
At Johnson & Johnson, for example, researchers are applying AI and machine learning (ML) algorithms to large de-identified datasets to locate clinical research sites with patients who could potentially benefit from the medicines being studied. The clinical trial operations team can then work to determine the likelihood of enrolling the newly identified sites into their trials.
Treatment for oncology patients is becoming increasingly more personalized. Our goal is to leverage AI to find ways to help ensure patients can be connected with these tailored treatments.
“At Johnson & Johnson, we leverage AI to optimize trial inclusion criteria—broadening eligibility while maintaining scientific rigor—so our protocols can reach the largest number of patients,” says John Yannone, Director of Global Feasibility, Johnson & Johnson. “By using AI and advanced analysis to identify sites with the greatest eligible patient pools, we accelerate enrollment and improve study representativeness.”
Data and AI are also helping researchers make clinical trials more representative, as advanced analytics are finding locations and healthcare institutions where patients from different backgrounds are more likely to be treated. Researchers can then prioritize recruiting eligible patients from those study sites into clinical trials. This is critical, given the importance of ensuring that medicines are studied in patient populations representative of those impacted by diseases.
Beyond helping researchers identify and enroll patients, AI can also improve how disease activity and treatment response are assessed during a clinical trial. For example, in inflammatory bowel disease, researchers often assess disease severity by examining the degree of inflammation and damage visible during an endoscopy. With the help of AI, researchers can more accurately measure the extent of damage in the large intestine and detect subtle changes that might otherwise be missed. This can provide a more consistent, detailed and objective assessment of disease activity, giving researchers a clearer picture of a patient’s condition and response to treatment over time.
5.
When it comes to more precisely understanding disease to deliver more personalized care, AI can be a real game changer. By applying AI to clinical data derived from or generated by common diagnostic tests such as CT scans, MRIs, electrocardiograms and echocardiograms, AI can enable more personalized targeted treatment and accelerate clinical trial enrollment.
Take, for example, bladder cancer, which is the ninth most common cancer worldwide. Despite the existence of approved treatments, and others under clinical evaluation now for patients whose tumors have specific alterations in FGFR genes, the uptake of testing to identify patients with these alterations is often limited, due to various challenges. As a result, patients with FGFR-positive tumors don’t always receive the targeted, personalized treatment that could significantly impact their care and outcomes.
To help address this challenge, Johnson & Johnson has developed an AI-assisted biomarker test that, when applied to digitized images of biopsies, can detect whether patients’ tumors are likely to have FGFR alterations. The company has leveraged this test in prior screening and is further evaluating the clinical utility of this technology for future use in trials and, eventually, in routine care.
“Treatment for oncology patients is becoming increasingly more personalized, thanks to big advances in both science and technology,” says Kris Standish, Senior Director, AI/ML and Computer Vision for R&D, Johnson & Johnson Innovative Medicine. “Our goal is to leverage AI to find ways to help ensure patients can be connected with these tailored treatments.”
6.
First comes the discovery and development of therapies, medications and other healthcare products. The next critical juncture? Making sure these products reach patients. AI can help with that, too.
Stocking products in hospitals, pharmacies, clinics and other healthcare facilities where they’re needed requires an accurate prediction of supply and demand. This can be challenging, as a wide range of factors can affect the supply chain, including market trends, economic disruptions, supplier issues and more.
AI-assisted solutions can help prioritize which locations will be greatly impacted in order to quickly respond to risk factors that may otherwise affect the ability to deliver products to people who count on them.
“It’s our responsibility to make sure patients and customers have reliable access to the transformational therapies our company creates,” says Steven Saus, Vice President, Innovative Medicine Supply Chain Strategic Operations, Johnson & Johnson. “AI is helping us build a stable, efficient and resilient supply chain so we can deliver on that obligation.”
When we use AI, it’s always with a purpose. Our Credo states that our patients and customers come first, and that will continue as we move forward with this technology.
For example, Johnson & Johnson is using advanced ML algorithms to sift through and analyze large amounts of data, including demand fluctuations and supplier performance, to help predict the impact of real-time events that may disrupt the supply chain (think severe weather events and economic disruptions).
Ensuring patients have access to the right products also requires timely communication with healthcare providers.
“At Johnson & Johnson, we’re using a range of AI approaches, such as generative AI and AI agents, to help us develop deeper insights about disease and treatment and translate them into meaningful action for healthcare providers and patients,” says Ajit Menon, Chief Global Commercial Data and AI Officer, Innovative Medicine Global Commercial Strategy Organization.
This application is part of a suite of Johnson & Johnson global AI capabilities known as Engagement.ai, designed to guide the company’s engagement with healthcare professionals and patients. Engagement.ai is powered by AI that is trained on extensive de-identified datasets to provide insights that maximize the company’s ability to support providers and their patients.
“Engagement.ai helps provide us with insights about when and how we should connect with healthcare providers and patients across all channels–helping equip them with relevant education, information, and services that can continually support treatment decisions,” says Menon.
Ultimately, that’s what AI in healthcare is all about: better serving patients. “When we use AI, it’s always with a purpose,” says Swanson. “Our Credo states that our patients and customers come first, and that will continue as we move forward with this technology.”
This story, originally published on October 10, 2024, has been updated.
