How doctors are using machine learning to improve health outcomes
Browse Blog Topics

How doctors are using machine learning to improve health outcomes

An ounce of prevention is worth a pound of cure, as the old saying goes. Until recently, that simply meant living a healthy lifestyle, getting regular checkups, and hoping that signs of anything serious were caught early. But today, doctors are using artificial intelligence (AI) and machine learning systems to make preventative care, diagnosis, and treatment more accurate and effective than ever.

“Machine learning involves adaptive learning and as such, can identify patterns over time as new data is aggregated and analyzed,” explains Melissa Manice, co-founder of healthcare startup Cohero Health. “Therefore, machine learning and AI allows doctors to detect abnormal behaviors and predictive insights with the application of clinical thresholds to machine learning algorithms,” she continues.

Cohero, which is one of the startups featured in the End of the Beginning video on digital health, provides connected respiratory devices for asthma patients, which feeds data back into a digital dashboard and machine learning system to help physicians better monitor and prescribe treatment for their patients. It’s just one example of what collaboration between doctors and AI will look like to improve healthcare.

Almost half (48%) of all life sciences organizations are already using machine learning to improve the patient experience, and another 48% are planning to deploy machine learning within the next three years. [Source: Statista]

Big data for physicians
Physician’s lives are consumed with a variety of mundane and repetitive tasks that often decrease the amount of time they’re able to dedicate to actually seeing patients and deploying higher-level expertise. Machine learning and AI are now helping to alleviate that burden, performing tasks that might take doctors days in a matter of minutes.

“Consider a CT scan that can contain hundreds of images,” says Yosi Taguri, founder of, which is part of the Samsung NEXT product team. The company’s deep learning platform is used by healthcare providers to manage and analyze medical images like x-rays and CT scans.

“With the help of machine learning, doctors are able to focus only on the most important images,” says Yosi. “Being notified about a specific image by the software, without having to review them one-by-one, is valuable and time-saving for physicians.”

Melissa adds that machine learning is applicable across the clinical, operational, and financial areas of the healthcare spectrum. This includes reading CT scans and other forms of diagnostic support, assisting in clinical decision making, and even matching patients with the right clinical trials.

“Other areas we hear about are supporting day to day operations,” Melissa says. “Chatbots, for example, can be used in patient scheduling communications. Other outbound notifications can also be generated, such as therapy regimen reminders.”

These are just a few ways in which machine learning is already being used by doctors. Another benefit of using machine learning in the healthcare field is that it can automate the processing of large amounts of patient data, helping physicians to diagnose and treat their patients faster and with greater accuracy.

Making better use of patient data
“Machine learning is currently being used by the healthcare community to analyze large amounts of structured and unstructured data,” Melissa continues, “This can be anything from electronic medical record data to clinical trial data sets.”

The power of such big data capabilities lie in the ability to consider both genetic and environmental medical factors when making a diagnosis. “Precision medicine will be enabled by the integration of genomic data with physical indicators captured from personal data trackers and smart sensors,” Melissa predicts. “This combination is more reflective of the composite health data that doctors would like to see, both at the individual and population-wide levels.

Yosi also thinks doctors are keen to make the most of data from multiple sources. But he says that doctors don’t just want more data about their patients, they want information that’s most useful at present. For example, machine learning software can quickly cross-reference real-time vital signs with past medical records and show doctors only the most relevant treatment history.

“One of the challenges for healthcare providers is having all this continuous data being generated about patients,” Yosi says. “From what I’ve seen, doctors now consider machine learning as a technology that decreases the time it takes to work with data as well as delivering more value from it.”

Doctors will be able to focus more on the patient experience, and less time sifting through patient records. Machine learning will help streamline the entire process, from analyzing vital signs to prescribing the best treatment regimen.

Improving the patient experience
The science fiction version of an AI-enabled hospital of the future usually includes things like roving robot nurses and surgeons performing operations in Virtual Reality (VR). But the reality is, machine learning’s impact on the patient experience will be much more subtle and integrated into new treatment protocols.

“It’s going to be seamless,” says Yosi. “I don’t think people will even know they’re using AI.”

The future of healthcare will be one in which patients interact with AI engines, chatbots, the Smart Home, and even a Smart Bed. Melissa feels that this type of paradigm might actually facilitate deeper bonds between doctors and patients.

“It’s the notion that AI will help improve the human connection and deepen the doctor’s empathy,” she says. “Doctors will be better informed, less stressed or distracted, and more cognitively alert. That’s the game-changer in the digital future of healthcare.”

AI will also assist in the most important part of the patient experience: recovering from treatment or surgery. Machine learning can, for instance, be used to identify and predict potential health problems for individual patients before they occur. This data-driven approach to preventative medicine will help patients make lifestyle changes or start treatment regimens to head off potential health problems at the pass.

“Because AI and machine learning can analyze large amounts of data and isolate abnormal patterns or outliers, it can assist in the identification of at-risk patients,” explains Melissa.

Machine learning is well on its way to mass adoption in the healthcare field, with applications like CT scan analysis already a reality. AI engines will make the most out of the vast array of patient data, and doctors will be able to focus more on what matters: their patients.

Learn more about trends in digital healthcare by watching the End of the Beginning video series.

Related Stories