The debate of risk vs. reward of AI and ML has been raging for years. For the most part, though, business leaders are embracing the potential of AI to drive innovation and efficiencies in many industries.
One of the most exciting areas for AI and ML innovation is in healthcare, particularly in the diagnosis and treatment of rare diseases.
Artificial Intelligence and Machine Learning: A Primer
Artificial intelligence (AI) and machine learning (ML) have evolved dramatically since their inception in the 1950s, but broadly speaking, AI is the theory and development of computer systems that can store and synthesize vast amounts of data to perform tasks normally requiring human intelligence (for example, facial recognition, speech recognition, decision-making, etc.). ML is a subset of AI where computers can be programmed to learn from experience and modify processing based on new information or data.
With the increasing avalanche of data available, industries such as financial services and manufacturing are realizing the advantages of AI and ML to automate tasks as well as deliver insights and models. Healthcare, and the rare disease space in particular, is in a unique position to take advantage of the advancements in AI and ML to drive improvements in complex patient diagnostics and care.
“If we can use AI to predict a problem before we waste resources and time on both sides of the patient-provider relationship, we’re going to create better experiences for everyone.”
Katherine Andriole, Ph.D., Director of Research Strategy and Operations at the MGH & BWH Center for Clinical Data Science (CCDS).
Diagnosing Rare Disease: The Untapped Potential of AI and ML
There are over 7,000 identified rare diseases worldwide. It is unrealistic to expect a primary care physician to have a depth of knowledge or experience in the complex symptoms, diagnostic criteria, and therapies of this plethora of rare diseases. Even specialists will not have in-depth knowledge of every rare disease. This is where the benefits of AI and ML can address the challenges in diagnosing and treating rare diseases.
Today’s computing power allows for huge amounts of data from multiple sources to be easily stored and rapidly analysed. This unstructured data can then be synthesized and structured in meaningful ways such as classifying patients based on their level of risk of certain diseases and/or predicting progression based on data trends. The combination of these multiple benefits yields a high potential for improving the speed and accuracy of the differential diagnosis and treatment of complex rare disease patients.
While several companies are developing platforms to leverage the power of AI and ML to identify genetic variants at the roots of rare diseases, the real potential and power for AI and ML have been in the advancement of clinical decision support.
The use of advanced technologies including leveraging the vast amount of unstructured data within electronic medical records is proving to accelerate high-risk rare disease patient identification and care optimization.
Identifying the undiagnosed
Regardless of years of practice and experience, expecting a primary care physician to accurately identify and diagnose a rare disease patient on the first visit is unrealistic. At Khure Health, our mission is to help solve that problem by providing primary care physicians with a platform that can help them identify at-risk patients faster and accelerate the process of getting them on the right care pathway.
Khure Health’s mission: Help physicians improve the lives of rare disease patients and their families.
We are leveraging the power of AI and ML to aggregate the vast amounts of clinical diagnostic information on rare disease states and combine that with the physician’s own case data stored in their electronic medical record (EMR). Through an intuitive clinical dashboard, an individual patient’s electronic medical record can be instantly screened with the click-of-a-button, against a rare disease algorithm to determine if the patient may be at risk of the particular disease. Physicians can also use the dashboard to scan their complex patients against an entire panel of the top rare diseases, making it possible to proactively identify high-risk patients, accelerate their time to diagnosis, and initiate the best next steps, whether that includes specific lab work or making a referral to an appropriate specialist.
While there is still much work to do, Khure Health is on the path to leveraging artificial intelligence and machine learning to enable physicians to accelerate the diagnosis of complex rare disease patients, end their suffering, and improve their quality of life.
Interested in learning more about how Khure Health helps primary care physicians identify rare disease patients?