An algorithm a Day Keeps the Doctor Away: The Rise of AI in Healthcare

The rise of AI and data science is changing the way business is done with crucial decisions now being made based on real data insights rather than guesswork. Marketers are first to understand the impact of these industry developments, with a previously unimaginable wealth of information on their customers – their preferences, habits and user journeys.

One area of business where AI can do so much more than increase profit margins is in healthcare, where inspiring capabilities are now possible, such as identifying disease progression and outbreaks, improving communication between patients and doctors, and, ultimately, saving lives.

Genetic Disorders

FDNA has already built an algorithm which significantly outperforms doctors in identifying genetic disorders. Doctors went head-to-head with AI technology in analysing over 17,000 children with 200+ genetic disorders. The technology was correct 64% of the time, whilst the doctors only achieved a 20% success rate.

Predicting Diseases

IBM have been using a probabilistic AI model to analyse various behaviours, such as shaking hands and mood swings, to try and identify the biomarkers of diseases like Huntington’s, Alzheimer’s, Parkinson’s and others in order to provide predictions for prospective patients. Unlike other AI models, it’s able to analyse multiple aspects of neurodegenerative diseases at the same time.

Evolving Treatment

MIT have been using an AI algorithm to analyse vast swathes of patient data and find the lowest possible doses of life-saving drugs to shrink tumours but minimise side effects. The algorithm has effectively reduced treatments to between a ¼ and a ½ of prior doses.

The pace of change is illustrated by Dr Jack Kriendler, Founder of the Centre for Health and Human Performance, who explains, “I would sooner today trust computer scientists and data scientists to tell me how to treat a really complex system like cancer than my fellow oncologists. I would not have said that two to three years ago.” This goes to show just how far AI has come, and highlights its promising future at the centre of cancer treatment.

Streamlining Patient-Doctor Communication

It’s been proven that allowing patients time and space to talk with medical professionals increases the likelihood that they’ll remember the advice they were given and take their prescriptions on schedule. While doctors are trained to give brilliant medical advice, they also need to be expert communicators.

AI is now used to provide clinicians with in-depth analysis of their communication skills (in work lead by researchers from The Dartmouth Institute for Health Policy and Clinical Practice, Trinity College Dublin and the University of Edinburgh. The technology supports doctors in their day jobs, alerting them when they’re using complicated language. In real time, doctors can be reminded of alternative solutions to the information available, which they may have missed,   ensuring they are always using evidence-based treatments.

Communication analysis could also become a key part of predicting disease. The same research found that AI can identify depression based on a patient’s tonal and stylistic changes when speaking, while heart failure could be predicted by measuring vocal changes in the lungs.