Technological advances in artificial intelligence (AI), automation, and robotics—once only dreamed of—are beginning to take shape and are promising to revolutionize healthcare and biotechnology. By 2021, AI alone could generate $6.7 billion in revenue from healthcare globally according to Frost & Sullivan.

Jonathan Linkous
Jonathan Linkous, Founding CEO, American Telemedicine Association

Some of the frontrunning scientists and physicians, who have worked toward AI and robotics advances for decades, spoke at the recent first annual Partnership for Artificial Intelligence, Automation, and Robotics in Healthcare (PATH) Summit, offering a distinct message: As these advanced technologies become integrated into healthcare and biotechnology there will be a significant change in the development and delivery of medicine resulting in improved outcomes, increased productivity, and wider dissemination. It is clear that doctors, health systems, and research institutes who don’t embrace this new wave of technology will be left behind.

PATH—a mission-driven, membership-based group launched by Jonathan Linkous, founding CEO of the American Telemedicine Association (ATA), and Mary Ann Liebert, publisher & CEO of Genetic Engineering & Biotechnology News (GEN), is dedicated to bringing together all stakeholders involved in the advancement of AI, automation, and robotics, to help guide policy related to its use and to promote its integration into health systems.

Mary Ann Liebert
Mary Ann Liebert, Publisher & CEO, Genetic Engineering & Biotechnology News

PATH grew out of the work of Liebert and Linkous. As a leading publisher of authoritative peer-reviewed journals in cutting-edge fields of health and medicine, including The CRISPR Journal, and a trailblazer in guiding the use of telecommunications to provide healthcare services, respectively, both have teamed up to bring together leaders in technology, medicine, and related disciplines to forge the future of healthcare.

“It is,” says Liebert, “the future of healthcare.  The individual disciplines of artificial intelligence, automation, and robotics must converge to ensure that knowledge and experience from each complements and moves forward new and meaningful advances and outcomes in healthcare. It is imperative that these exciting technologies be considered together as they have cross-cutting features that are merging together as new delivery systems are being implemented.”

Healthcare Transformation cover image

According to Linkous, “These advanced technologies are at a similar stage as telemedicine was 25 years ago.  Just as ATA did, our goals for PATH are to accelerate adoption and integration, reduce government and professional regulatory barriers, align payment policies and incentives, promote partnership in developing ethical applications and advance public understanding.”

Healthcare Transformation, the official journal of PATH, focuses on the most critical technology innovations in AI, automation, and robotics, moving the field beyond research to enable the most effective use in the worldwide ecosystem of medicine. Linkous serves as editor in chief of the Journal; Founding editor in chief Stephen Klasko, MD, MBA, CEO of Jefferson Health in Philadelphia, is senior advisor to PATH as well.

The future for the medical profession

“There isn’t a single provider who today can embrace and remember all the information they need to,” Jay Sanders, MD, CEO of The Global Telemedicine Group and adjunct professor of medicine at Johns Hopkins University said at the PATH Summit keynote.

Assistance from computers can help doctors prevent medical errors, recently cited by a Johns Hopkins study as being the third leading cause of death in U.S. hospitals.  Sanders foresees AI and other forms of automation building on the connections made by telemedicine and the data within a patient’s electronic health record to provide timely differential diagnoses and treatment plans based on the wealth of information contained in the world’s most recent medical literature.

According to John Sterling, editor in chief of GEN, “The increasing utilization of proteomics and genomics data by clinicians is also presenting physicians and other medical professionals with growing complexity regarding their decisions on how to best treat their patients.” He Continues “This is all taking place within a healthcare environment moving toward precision medicine and the adoption of immunotherapy. Now is the perfect time for PATH to suggest and provide solutions that will improve patient care while reducing ever rising costs. Data and clinical results that are obtained from the application of PATH-inspired approaches can then be translated into ideas for improving R&D in the healthcare sector and for suggesting new directions in basic research.”

The cloud expands AI capability beyond geographic boundaries

Yulan Wang, PhD, founder, chairman, and chief innovation officer of InTouch Health, and also a keynote speaker at the PATH Summit, emphasizes that the point of the in-roads being made in AI and robotics is to help provide “high-quality healthcare to everybody, at lower costs.”

Telehealth is already helping to achieve this by making medical care available more universally, says Wang, who performed the first transatlantic surgery. “When a clinician interacts with the patient, it doesn’t matter how informed you make that clinician if that clinician is only able to provide their skillset to the immediate patients around them,” he says. “We need to change that to impact a greater number of patients. This is especially true as medicine is getting more and more complex. If people are no longer constrained by geographic boundaries, then we have improved the delivery system in a profound way.”

Another benefit of the advances in health IT: As data is being stored and routed through the cloud, data analytics and AI can be applied.

“Today, AI machine learning is deep learning,” he adds. One of the ways this works well is with telehealth visits of stroke victims, where the system can provide
real-time feedback to help clinicians make the right diagnoses. “I’m a computer robotics engineer by training,” Wang explains. “The convergence of telehealth, with AI-based virtual care delivery ‘built on top,’ is going to allow the kind of transformation that will allow for better care for more people, at lower costs.”

Computer-aided diagnosis improves prostate cancer detection

Prostate cancer is the second leading cause of death for men, but diagnosing it is still difficult, says Peter Choyke, program director of the Molecular Imaging Program at the National Cancer Institute (NCI).  Choyke was part of a PATH Summit panel titled, “Successful AI Applications: What Have We Learned from Imaging AI in Radiology?” “One of the problems with prostate MRI readings is the subjectivity involved. There is clearly a need for computer-aided diagnosis (CAD),” he says.

Around 2010, NCI found a way to fuse MRI with ultrasound to ease the biopsy process, but, because of this new technology, many more radiologists needed to be trained to read the images, and they were very difficult to interpret, Choyke explains. At the same time, NCI began to use its own store of 8000 MRI prostate screenings to train a CAD, combined with the much denser pathology data, and the first attempt resulted in 90% specificity, he says. NCI is continuing to upgrade the CAD’s capability—now on its fourth version—and the goal is to add 10,000 cases to its database. “It gets better the more data you add,” Choyke says.

“From our experience, it helps to start with a stable, reliable dataset so you know where you’re going, and that really gets you into the ballpark; we’ve really gone very far on that data,” he explains. “However, you have to introduce more variety into the dataset and more adversity and variation, with careful curation.” There still is a lot of reader variability and some skepticism, but NCI is keeping its international team of radiology readers together to continue to refine the work. “Our CAD, with its deep learning, has the potential to revolutionize the diagnosis of prostate cancer,” Choyke says.

Despite being new, PATH is growing rapidly and is already bringing about change.  Senior advisors to PATH include the heads of Jefferson Health, QUALCOMM Life, InTouch Health, Syneos Health, GE Ventures, and ViTel Net. The next PATH summit will be held
September 30–October 2, 2019 in  Washington, DC. 

More information about PATH, including membership, is available at:

Translational Informatics Offers Potential

Moving from the initial stages of target identification to an approved drug remains a significant challenge for biopharmaceutical companies.

According to the Tufts Center for the Study of Drug Development, pharmaceutical companies have nearly doubled their R&D investment in personalized medicines over the past five years, and expect to increase it by an additional 33% over the next five years. Investment in translational research has become central to the industry’s strategy of accelerating and optimizing the drug discovery and development process.

To ensure a satisfactory return on investment, translational research teams need access to massive amounts of combined clinical and genomic data, allied to a robust informatics strategy for interrogating that data as they hunt for targets and biomarkers of disease progression and therapy response.

Given the huge volume of disease-specific and population-level genetic data drawn from public datasets, as well as patient-level data collected from electronic health records and clinical trials, a reliable and secure informatics infrastructure enables research teams to discover actionable insights contained within these data assets.

New tools, methods, and visualizations are required to share and use petabyte-sized clinical and genetic data securely and efficiently in an environment compliant with globally emerging patient protection requirements.

At DNAnexus, says Megan Laurance, PhD, the company’s director of data services, a translational informatics solution has been developed that provides researchers with a compliant platform and interactive tools to harmonize, explore, analyze, and visualize multi-omic and clinical datasets at scale.

“The result is accelerated and improved decision support for identification of drug targets and markers of disease progression and drug response,” she says. “The informatics environment enables researchers to interrogate clinico-genomic datasets across multiple stages of the drug discovery and development process, such as indication discovery, derisking, biomarker identification, evidence generation, and drug target discovery.