Next year, Siri will be a decade old and Alexa will soon be six. These and other digital assistants like Google Home and Cortana use artificial intelligence (AI) to recognize and respond to voice commands. They serve millions of users daily by giving directions, setting alarms, answering trivia questions and even telling jokes (try asking Siri what zero divided by zero is). In 2019, 111.8 million people in the U.S. – over 33% of the population – are expected to use an AI voice assistant at least monthly. That’s no surprise considering that Siri alone is active on 500 million devices, responding to 10 billion requests each month. AI is clearly making an impact in consumer goods. And analysts project that it will become a nearly $200 billion industry by 2025, making major strides in the areas of manufacturing and customer service. What we’re watching, however, is how AI is disrupting healthcare. While we haven’t yet seen AI transform healthcare to the point where there are more robots than humans in hospital hallways delivering care, we are seeing costs getting cut as AI augments human activity to make our industry more efficient and knowledgeable. One AI tool making its mark is Machine Learning (ML). The primary objective of ML is for computing technology to mimic human operations. In healthcare, ML adds exponential benefits to administration, acting as the router between systems and data by automating repetitive high traffic tasks. Here are some examples: Christus Health System, headquartered in Texas, automated 80% of pre-registration tasks and increased productivity by 60% An AI system from Google-owned DeepMind can identify 50 sight-threatening eye diseases with up to 94.5% accuracy, and is a lower cost alternative than a traditional MRI The development of smartphone apps allows for at-home monitoring of health conditions, communicating with doctors and scheduling appointments There are numerous time and cost efficiencies to be gained through these and other AI-driven tools continually being researched and launched; however, advancement depends largely on one thing: data. In a recent report, Microsoft’s European office suggests the following gaps need to be worked out to support the widespread use of AI within healthcare: Organizational and technical barriers to data sharing and data use Insufficient public trust and lack of a regulatory framework that promotes more access to and use of patient data for research purposes, while addressing privacy and security concerns A lack of clear rules, or even a tentative discussion framework, governing the ethical and social implications of patient data, AI and its growing use in the field of healthcare One of the biggest organizational challenges mentioned is that health data is often stored across numerous silos, making access to a complete view of an individual patient very cumbersome. This, along with security, is a driver in the decommissioning of legacy systems. Decommissioning applications allows data to get extracted, migrated and consolidated to a single archive in health systems nationwide. AI can be leveraged in this ETL process as well as in tools that can utilize that common historical data store for research or analytics. The report goes on to make several technological recommendations, including the need to: Advance a common framework for documenting and explaining key characteristics of datasets Invest in technical solutions, including through research funding, to enable secure machine learning with multiple data sources/systems At Harmony Healthcare IT, where we develop and deploy HealthData Archiver™, our data experts are utilizing machine learning and predictive analysis to help us ingest, index, and properly identify patient data. We are constantly exploring how AI can help to automate and standardize our ETL (extract, transform, load) process to increase the speed, quality, predictability and punctuality of our health data management work. The future of AI in Healthcare – especially as data volumes skyrocket – is promising. We are all anxious to see how AI can help reduce the quarter of a million of American deaths each year due to preventable medical errors. But as those more complex care delivery algorithms get written, we will continue to see advancements on the administrative and data management side of healthcare. Want to learn more? Contact us.