Wikipedia defines artificial intelligence in healthcare as the use of complex algorithms and software to emulate human cognition in the analysis, interpretation and comprehension of complicated medical and healthcare data. This “emulation” is done in less time and at a fraction of the cost. Artificial intelligence in healthcare was valued at about $600 million in 2014 and is projected to reach $150 billion by 2026. Reinventing and reinvigorating healthcare through the use of artificial intelligence is happening predominantly through assisting in better diagnosis, better processes, drug development and robot-assisted surgery.
Better Diagnosis
In 2015 misdiagnosing illness and medical error accounted for 10% of all U.S. deaths. AI has been shown to be able to predict and diagnose disease at a faster rate than most medical professionals. With the help of medical imaging training data AI can model critical diseases like cancers and heart attacks. AI-enabled radiology assistants receive imaging scans and automatically analyze them for various clinical findings. These analyses are then passed onto the radiologists who include this information when making their diagnosis. Algorithms are outperforming radiologists at spotting malignant tumors and guiding researchers in how to construct cohorts for clinical trials.
Natural Language Processing, an AI technique, can analyze unstructured clinical notes on patients, prepare reports, transcribe patient interactions and conduct conversational AI along with classifying and integrating clinical documentation and published research.
Better Processes
A study in 2016 of 35,000 physician’s reviews showed that 96% of patient complaints were about lack of customer service, confusion over paperwork and negative front desk experience. There are AI platforms automating repetitive tasks to free up administrators to work on higher level ones. There is AI-based software that solves operational problems for emergency rooms and those related to patient safety. The platform can prioritize patients by illness or injury, track waiting times and chart the fastest ambulance routes. Mining health data, hospitals can predict and detect sepsis which in turn reduces death rates…the ultimate patient service. Canada’s public health system uses an AI app that scans a patient’s survey answers and immediately connects them via video with the right healthcare provider. AI provides the methodology to sift through clinical and molecular data to personalize healthcare treatments. Analyzing data from genetic sequencing to image recognition gives physicians better insights into treatments and cures.
Drug Development
It costs $2.6 billion to put a new drug through clinical trials and only 10% of those drugs are successfully brought to market. The use of artificial intelligence in drug development can improve the efficiency, accuracy and knowledge for biopharma companies. XtalPi’s crystal structure prediction technology predicts complex molecular systems for drug design in days rather than weeks or months. AI can help researchers find the right candidates for developmental drugs that statistically raise the chances of successfully passing clinical trials while also decreasing time and cost to market. Targeting treatment to get the right treatment to the right patients at the right time by using artificial intelligence to produce a better target selection usually provides previously undiscovered insights through deep learning.
Robotic Surgery
Surgical robots, initially approved in the US in 2000, provide surgeons with superpowers, improving their ability to see, create precise and minimally invasive incisions and stitch wounds. This leads to fewer surgery-related complications, less pain and a quicker recovery time. Intuitive’s da Vinci platform was the first robotic surgery assistant approved by the FDA. During each surgery it is constantly recording information and analytics to train surgeons and improve future surgeries. It has assisted in over five million operations. Endoscopies have been improved by micro-instrumentation, design, data science and AI. Vicarious Surgical uses virtual reality with AI-enabled robots so surgeons can virtually shrink and explore the inside of a patient’s body in much more detail. The Accuray CyberKnife robot uses 6D motion-sensing technology to aggressively track and attack cancerous tumors while saving healthy tissue.
In a report on artificial intelligence in healthcare for the Future Healthcare Journal of the Royal College of Physicians, Thomas Davenport and Ravi Kolakata say, “It also seems increasingly clear that AI systems will not replace human clinicians on a large scale, but rather will augment their efforts to care for patients. Over time, human clinicians may move toward tasks and job designs that draw on uniquely human skills like empathy, persuasion and big-picture integration. Perhaps the only healthcare providers who will lose their jobs over time maybe those who refuse to work alongside artificial intelligence.”
