You’re probably familiar with IBM’s supercomputer program Watson from its stint on Jeopardy! about four years ago when it defeated past champions Ken Jennings and Brad Rutter. Described on the company’s website as “a technology platform that uses natural language processing and machine learning to reveal insights from large amounts of unstructured data,” it was this combination of intelligence that allowed it to understand and answer the Jeopardy! questions correctly. Since the show, however, IBM has used Watson more usefully; here are five ways in which the program isn’t only helping the medical field but changing it altogether.
Interpreting Medical Images
A few months ago, IBM bought Merge Healthcare for $1 billion. The acquisition gave IBM access to more than 30 billion medical images from over 7,500 hospitals and clinics in the U.S. that used to belong to Merge. Though most people in the world would have no use for these images, they were a gold mine for Watson.
IBM will feed Watson these images in hopes that it’ll spot important information physicians might miss. It’ll also have the help of a new technology it’s been upgraded with, called deep learning, which allows it to spot patterns in data involving large amounts of information.
With deep learning and all the medical images, Watson will be used to study and diagnose melanoma, a difficult form of skin cancer to spot due to the way in which it manifests differently in each individual. Using Watson’s deep learning abilities, it might be possible for the software to recognize important, but sometimes missed, features of the disease.
For example, doctors can feed an image of a newly diagnosed melanoma to Watson, which then adds the image to its databases, so it can be compared to thousands of other images. By compiling this information, it allows doctors to have an overall understanding of how to diagnose future potential melanoma patients, as well as a list of treatment options.
“Combining Merge’s leading medical imaging solutions with the world-class machine learning and analytics capabilities of IBM’s Watson Health is the future of health care technology,” said Michael W. Ferro Jr., Merge’s chairman, in a press release. “Merge’s leading technology and proven expertise represent a unique combination of assets that will deliver unparalleled value to Watson Health clients. Together, we will unlock unprecedented new opportunities to improve patient diagnostics and deliver enhanced care.”
Treating Rare Forms of Childhood Diseases
One in 10 Americans suffer from rare diseases, according to the Global Genes Project, and half of them are children. So, IBM and Boston Children’s Hospital teamed up in hopes of using Watson’s cognitive platform to help identify, diagnose, and treat rare pediatric diseases.
Initially, the partnership will focus on steroid-resistant nephrotic syndrome (SRNS), a rare genetic form of kidney disease that affects children as young as 2 years old. Watson will learn nephrology — a form of study that focuses on kidney function — by reading medical literature on the subject and pulling together more information on mutations for SRNS. Experts from the Boston Children’s Hospital will then give Watson genomic information from previous patients to further enhance its knowledge of the disease.
"One of Watson's talents is quickly finding hidden insights and connecting patterns in massive volumes of data," Deborah DiSanzo, general manager at IBM Watson Health said in a press release. "For the kids and their families suffering without a diagnosis, our goal is to team with the world's leading experts to create a cognitive tool that will make it easier for doctors to find the needle in the haystack, uncovering all relevant medical advances to support effective care for the child."
Once Watson finishes focusing on SRNS, it will analyze the scientific literature and clinical databases found in the Watson Health Cloud in order to match genetic mutations to other diseases and find better ways for doctors to pinpoint treatment options. The purpose of the entire project is to provide doctors with a system that can interpret a child’s genome sequence, which, when coupled with medical literature, will allow them to properly identify rare diseases faster than ever before.
Treating Cancer Patients Where Doctors Are Scarce
In places where patients far outnumber doctors — such as India, which barely has one doctor per 1,700 patients — getting the required treatment can be a monumental task. But recently, IBM and Manipal Hospital, the third-largest health care network in India, have partnered up to to use Watson for diagnosing and treating cancer throughout Manipal’s 16 facilities and academic centers, which treat some 200,000 cancer patients each year.
There are more than one million new cancer cases diagnosed annually in India, where a shortage in doctors is about 16 times greater than in the US. “We are desperately short on doctors,” Dr. Ajay Bakshi, chief executive officer of Manipal Hospitals, told Forbes. “This will ensure my patients are getting the best possible care.”
Called Watson for Oncology, the new software was designed in partnership with Memorial Sloan Kettering Cancer Center in New York. Similar to what it’s doing at Boston Children’s Hospital, Watson will ingest medical information from journals and textbooks, and then put this information into terms anyone could use to find different treatment options, analyze specific patient information, and tailor the evidence it finds to meet their health needs. It will also use deep learning to continuously update its databases with information, which means it’ll constantly be learning more about oncology as more textbooks and journals are published.
As for how exactly Watson will help doctors deal with India’s major overload of cancer patients, Dr. Bakshi said in a press release, “Watson for Oncology will assist Manipal’s physicians in their goal to provide every cancer patient the most advanced, efficacious, and cost-effective treatment if they are seen by a Manipal oncologist anywhere in the Manipal network.”
Understanding A Person’s Genetic Profile and Offering Personalized Treatment Options
When it comes to personalizing cancer patients’ treatment options, it can sometimes take upward of a few weeks. In collaboration with 14 cancer institutes, IBM hopes Watson can cut this time down to a few minutes.
The institutes, which include the New York Genome Center, the University of Kansas Cancer Center, and the Yale Cancer Center, will use Watson’s deep learning abilities to translate a patient’s DNA into a usable genetic profile. When combined with relevant medical literature, it will allow doctors to provide patients with the most personalized treatment options. Deep learning will also continue to improve Watson’s insights on cancer treatments, giving oncologists bountiful amounts of information they wouldn’t be able to find anywhere else in such a short time.
A single individual’s genome takes up about 100 gigabytes of memory on its own. Analyzing this information in addition to health records, journals, studies, and textbooks would take up countless hours, which doctors could instead use to attempt treating their patients. Watson, on the other hand, can complete this task in a matter of minutes, producing a visualization of the patient’s case and choosing potentially useful drugs based on the patient’s DNA profile.
"When you are dealing with cancer, it is always a race," said Dr. Lukas Wartman, assistant director of cancer genomics at The McDonnell Genome Institute at Washington University in St. Louis, in a May press release. "As a cancer patient myself, I know how important genomic information can be. Unfortunately, translating cancer-sequencing results into potential treatment options often takes weeks with a team of experts to study just one patient’s tumor and provide results to guide treatment decisions. Watson appears to help dramatically reduce that timeline."
In terms of tech years, IBM Watson is still in its infancy. The technology that will allow Watson to perform to its fullest potential will come in the near-future, but the keyword there is future. The evolution of any technology takes time, and as much as the work happening now may raise hopes, Watson can’t yet replace doctors, nurses, or other health care professionals. However, Watson’s promise to scan billions of medical images and limit diagnosis time, provide info for more personalized treatment, and help doctors manage their patient load is still something to look forward to.
Powering An App That Gives Nutrition Advice To Pregnant Women
While most of the aforementioned innovations will be found in hospitals and emergency centers around the world, there is one way in which Watson will be available on your smartphone.
IBM unveiled the Nutrino App Powered by Watson in December. A collaboration with healthy eating app developer Nutrino Inc., the smartphone app is meant to give pregnant women “science-based, personalized, and contextual nutrition advice.” The app will use Watson and Nutrino’s nutrition insights platform to give personalized meal recommendations and 24/7 nutrition support.
“Healthy eating — a challenge at any point in our lives — becomes that much more daunting during pregnancy, when a mom-to-be’s nutritional needs can fluctuate week to week,” said Dr. Yaron Hadad, co-founder and chief science officer at Nutrino, in a press release. “The Nutrino App Powered by Watson helps women navigate through a trove of available nutrition information and offers recommendations responsive to a woman’s changing needs throughout her pregnancy.”
Once a pregnant woman registers with the app, she can input her pregnancy status, dietary needs, health goals, food preferences, eating habits, and even data culled from wearable devices like Fitbit. She can then ask Watson a variety of questions about pregnancy, such as, “Can I drink coffee?,” “Are eggs OK to eat in my first trimester?,” or “What should I eat to help with heartburn in my third trimester?” Watson answers these questions by searching through Nutrino’s database, which contains over 500,000 foods and 100,000 medical sources. Over time, Watson’s deep learning helps tailor the answers and recommendations to the individual, thus providing personalized care inside the home.
As beneficial as Watson will be in hospitals and emergency centers around the world, there is one takeaway Dr. Kyu Rhee, chief health officer for IBM Watson Health, emphasizes: It won’t replace health care teams, just augment them. “No, no. No replacement for doctors, it’s about relationships — relationships between [the patient’s] health care team and the patient, and the [relationship between] the health care team themselves.” Rhee told Medical Daily.
It’s also about doctors and cognitive systems, and augmented intelligence. But primarily, it’s about strengthening relationships in a health care setting. “I can’t know everything about evidence or [a patient’s] chart. To have Watson remind me or prompt me” will be very important in getting the most out of every visit to the doctor, he said.
IBM’s ultimate goal is to “democratize access” to health care. Patients will always have access to their medical records because, “health care is digitized, [and there are plenty of] opportunities to leverage that data to help people in the health care setting.”
Correction: A previous version of this story stated one in 10 Americans who suffer from rare diseases were children. One in 10 Americans suffer from rare diseases.