Are we seeing artificial intelligence (AI) advances in healthcare? The truth is, from the moment AI came on the scene, developers, scientists, and investors had the healthcare industry in mind as the source of greatest impact. Our first introduction to AI software came in the form of IBM Watson, a computer system capable of answering questions posed in natural language, often in a fraction of a second.
For many people, Watson became synonymous with AI; however, it’s important to note that Watson is comprised of several AI applications - Natural Language Processing, Machine Learning (ML), and Knowledge Representation to name a few. Watson was engineered to identify word patterns and predict correct answers for trivia, and in 2011, Watson wowed the tech industry when it went toe-to-toe with Ken Jennings and Brad Rutter (two Jeopardy winning machines in their own right!) outperforming both in an epic game of Jeopardy.
From that moment on it was only a matter of time until Watson’s skills could be applied outside the scope of a gameshow contest into other fields such as finance, law, academia, and most importantly…healthcare. This made the most sense because healthcare is a field rich in data, which is the fuel of modern AI and Machine Learning programs.
On the internet, you’ll often see the abbreviation AI/ML; and that is because AI and ML are related, but different concepts within the same space. AI is the ability of a computer to emulate human cognition, while ML is the mathematical processing of data that helps computers learn specific tasks and reasoning. Nowadays, with the vast amount of data available from electronic health records (EHR) to mobile fitness apps such as FitBit, we can apply AI/ML in several ways to improve health such as:
1. Accurate Diagnosis of Disease
AI/ML methods have been applied to radiographic imaging to better differentiate between COVID-19 pneumonia and other viral pneumonias. As you can imagine, this helps effectively manage patient cases, especially in settings where specialists able to analyze chest radiography are in short supply. There are also general expert systems that can diagnose diseases across all the medical specialties, like Iliad, an internal medicine one of our InnoVet founders helped develop in the 90’s.
2. Patient-Generated Health Data and Health Risk Assessment
We can use AI/ML to leverage a patient’s medical history and continuous data shared from fitness devices (Apple watch or FitBit) to help identify risk factors for different diseases. Your Apple watch, for example, continuously measures your heart rate and uses AI/ML to alert you if your heart rate is too low or if it has spiked (when not being active) outside your normal ranges.
3. Risk of Kidney Disease Complication
ML algorithms (problem-solving processes following a set of detailed instructions) are utilized to identify kidney disease patients at high-risk of progressing to renal failure by collecting demographic and laboratory tests data from their EHR and applying risk-scoring calculations.
4. Early Detection of Brain Disorders
With advances in genomic sequencing, the application of AI/ML is driving the development of new tools for neurological health diagnostics, such as non-invasive detection of early-stage Parkinson’s disease, and diagnosing the presence of concussion and the subsequent risk of prolonged symptoms.
So, will AI have a big role in the future for healthcare IT? We think so. What we’ve seen so far is just the tip of the iceberg for what AI in healthcare is capable of. Recent advances in machine learning techniques, coupled with greatly improved computing power and storage, and the adoption of new healthcare IT standards, AI/ML are yielding a growing number of exciting clinical applications that are easier to integrate into providers’ workflow. Here at InnoVet we are excited to be part of this innovative expansion of healthcare IT.
By: Ivan Castro (Sr. Business Intelligence Lead at InnoVet Health)
InnoVet Health is an IT consultant company specializing in AI and business intelligence, digital services, and health interoperability founded by MIT-alumni & informatics experts. Learn more about us on our website or reach out on LinkedIn.