artificial intelligence woman's eye

Can Artificial Intelligence Predict Patient Outcomes?


Exciting changes are underway for technology in health care. In our post about what to expect in 2017 we touched on many key trends, but one particular emerging technology deserves a closer look: artificial intelligence (AI).

For more on this topic, see New Year, New Health Care: What to Expect in 2017

“While the U.S. health industry lags behind other industries … in deploying emerging technologies such as artificial intelligence, drones, and virtual reality, 2017 is the year to prepare for the eventual arrival of these technologies and their impacts on business models, operations, workforce needs, and cybersecurity risks,” stated consulting firm PWC’s annual trend report.

Here are three ways that artificial intelligence is being used to predict patient outcomes, potentially transforming the way doctors treat patients.

Using artificial intelligence to detect and treat eye disease

In December 2016, Microsoft India, in collaboration with the country’s L V Prasad Eye Institute, launched Microsoft Intelligent Network for Eyecare (MINE), a research group that will leverage artificial intelligence to deliver large-scale eye care.

MINE will work with a global consortium of research and technology institutions, including the U.S.’s University of Miami, Brazil’s Federal University of Sao Paulo, and Australia’s Brien Holden Vision Institute, with the goal of eliminating avoidable blindness and improving worldwide delivery of eye care services.

Microsoft India has launched a global research group that will use artificial intelligence to detect and treat blindness-causing eye diseases.

India is a logical place to launch a project that uses artificial intelligence to detect and treat blindness-causing eye diseases, since the country is home to some 55 million of the world’s 285 million people living with vision impairment, reports mobihealthnews.

“Children will be a big focus for the project, and the consortium will explore how machine learning can be used to study rate of change of myopia and other conditions that impact eyesight in children. The group also aims to develop predictive outcomes of refractive surgery and establish optimal surgery parameters for each individual.”

How artificial intelligence can predict heart disease outcomes

Another example of how artificial intelligence is being used to predict patient outcomes and inform treatment decisions comes from the U.K. Measuring heart function can be a difficult and inaccurate process—a problem when you’re dealing with life-threatening diseases such as pulmonary hypertension, which can lead to heart failure if untreated.

New software developed by scientists at Imperial College London creates virtual 3D versions of patients’ hearts that replicate their individual heartbeats. “Artificial intelligence is able to rapidly learn which features of cardiac function best predict heart failure and death,” explains the Imperial College web site. The results of a study published in the journal Radiology showed that, with the use of MRI imaging and information from blood tests and other medical data, this machine learning program is more accurate and faster at making predictions than current methods.

A British AI program is more accurate and faster at predicting heart failure and death than current methods.

Lead author Dr. Declan O’Regan said, “This is the first time computers have interpreted heart scans to accurately predict how long patients will live. It could transform the way doctors treat heart patients.” The researchers say this is the first study to use artificial intelligence to predict heart disease outcomes.

How artificial intelligence more accurately detects breast cancer risk

Researchers at Houston Methodist have developed artificial intelligence software that “reliably interprets mammograms, assisting doctors with a quick and accurate prediction of breast cancer risk,” reports Science Daily. The AI program “intuitively translates patient charts into diagnostic information at 30 times human speed and with 99 percent accuracy.”

In the U.S., 15 million mammograms are ordered or provided annually, according to the most recent data from Centers for Disease Control and Prevention. Of those, 50 percent yield false positive results, according to the National Cancer Institute, resulting in one in every two healthy women told they have cancer. Biopsy is the recommended next step for these women; an estimated 20 percent of biopsies are unnecessary.

AI has the ability not only to reduce unnecessary biopsies, but to save doctors time. Manual review of 50 charts took two clinicians 50-70 hours. AI reviewed 500 charts in a few hours, saving over 500 physician hours, according to the Science Daily article.

Manual review of 50 breast cancer patient charts took two clinicians 50-70 hours. AI reviewed 500 charts in a few hours, saving over 500 physician hours.

The possibilities for AI in health care are promising, but some challenges still need to be addressed, such as gaining providers’ trust and dealing with policy, regulatory, and commercial obstacles, notes Stanford University’s “One Hundred Year Study on Artificial Intelligence (AI100)”. “The reduction or removal of these obstacles, combined with innovations still on the horizon, have the potential to significantly improve health outcomes and quality of life for millions of people in the coming years.”

For more articles like this, sign up for our newsletter today.


Sign up for the Rendia Insider

Monthly update from experts in the field aimed to improve efficiency in your practice.