Artificial Intelligence
The hand of humanoid robot AILA (artificial intelligence lightweight android) operates a switchboard during a demonstration by the German research centre for artificial intelligence at the CeBit computer fair in Hanover March, 5, 2013. Reuters/Fabrizio Bensch

While artificial intelligence (AI) is currently unpopular because of opinions voiced by thought leaders such as Stephen Hawking who warned that development of full AI could spell the end of human race as robots kill humans, many health professionals think otherwise.

In an article at the World Economic Forum, Eugene Borukhovich, senior vice president and global vertical leader for healthcare of SoftServe, listed five areas in the healthcare sector that would benefit from AI.

  • A predictor of drug resistance – Scientists are working on ways to use AI to predict responses from two chemotherapy medication to treat breast cancer patients. This is because of the different responses to people to the same cancer drug. A team put up by Peter Rogan, professor at the University of Western Ontario, published a study in the journal Molecular Oncology that found it was possible to predict which female breast cancer patients would experience improvements when they used the drug Paclitaxel.
  • To support adherence to medication – There is a startup in the US, AiCure, which uses AI via the patient’s mobile phone to confirm if the person took his or her prescription medicine. The smartphone’s camera is used to capture and analyse evidence that the drug was taken. The real-time data is then used to centralise immediate intervention and longtitudinal tracking of adherence patterns. The study says that when the medication is not followed, it is associated with poor therapeutic outcomes, worsening of the ailment and contributes to higher health care costs.
  • Smarter drug development – IBM Watson, by understanding and extracting vital information from scientific medical literature, attempts to visualise the relationship between drugs and other potential ailments. In 2014, IBM announced that Johnson & Johnson and rival Sanofi would collaborate with the IBM Watson’s Discovery Advisor team to teach the supercomputer to read and understand scientific studies with clinical trial outcomes. The result would be used to help develop and evaluate drugs and other treatments. The implication of the IBM statement is that it could help physicians match a medicine with the right set of patients to maximise effectiveness and minimise side effects.
  • For Alzheimer’s patients – The University of Washington and its Department of Computer Science had a project that explored the use of AI to support and enhance the independence and quality of life of patients with Alzheimer’s. Cognition systems developed by AI would replace some of the memory and problem-solving abilities lost by the patient due to the debilitating ailment.
  • Wearable health – According to Zulfi Alam, general manager for personal devices of Microsoft, the smart upcoming algorithms of Internet of Things applications will know enough about the user and his or her biometrics in a steady state to recognise patterns and opportunities to improve the owner’s health and fitness.

While the five applications of AI could add value to healthcare, particularly areas that could go beyond human abilities, since AI is still “in an infant stage of development,” it would not replace physicians, says Daniela Hernandez of Kaiser Health News.

In an example of how AI is now being used, Wired cites the case of Long Island dermatologist Kavita Mariwalla who logs in to Modernizing Medicine, a repository of medical information and insights online. She had a patient with a rare skin condition that causes big, watery blisters, but the medication that doctors prescribe for the disease, an autoimmune disorder, was not available.

Within seconds, Mariwalla got the name of another drug which worked in comparable cases. The Web site gathers its data from about 3,700 providers and over 14 million patient visits and treatment data provided to patients with similar profiles.

Contact the writer at or tell us what you think below