A recent article in Wired Magazine discusses Etiometry and the importance of developing impactful analytics to improve the quality of care in the ICU. While the article focused on artificial intelligence (AI) and whether it would in some ways eventually automate physician responsibilities, it reached the following logical conclusion:

“If we are able to develop systems that enhance their [physicians] capabilities and allow them to provide their patients better care, we will be in a win-win situation for healthcare professionals, patients and taxpayers.”

The article is provocative, and as Etiometry’s CTO it made me think quite a bit about the role of AI in the hospital ICU. Although the common perception of AI is the Terminator-like scenario where we worry about machines taking over the world, in more practical terms it refers to adding increased intelligence into smart platforms to augment, accelerate and enhance decision-making. If you accept that AI strives to automatically aggregate, correlate, and contextualize data from multiple sources to help humans make better decisions faster then you can see the potential value of AI in the ICU.

In the ICU, it all circles back to the data utilization problem. Right now, the ICU is not necessarily a “Big Data” environment since the data generated by a single patient pales in comparison to the data generated by video or image processing systems. However, currently the sole data fusion and processing mechanism in the ICU is the clinicians’ brain, which primes this complex and data-intensive environment for data overload.

The clinicians have to act quickly and effectively, and have to swiftly synthesize data from multiple devices and sources to make informed and often life-critical decisions. AI will play a role in alleviating mechanistic demands on clinicians by transforming the generated data into actionable information. This will allow clinicians to view and more quickly process patient information, thus facilitating more informed and hence more effective decision making.

For example, our Etiometry Platform is designed to raise the clinical team’s situational awareness by elucidating the rapidly evolving patient trajectory. Specifically, the platform intuitively visualizes salient physiologic trends among multiple data streams while enhancing their interpretation by near real-time analysis of actionable patient risks.

AI implementations like this will alleviate the data overload that ICU healthcare providers face every day so clinicians can focus more on the patient interaction, allowing them to be healers first and data processing units second. ICU clinicians help patients through a difficult time, and AI technology will increasingly support ICU care by providing the data collection and analysis that can free up clinician time for focusing on the big picture, ensuring that patient preferences are addressed, and helping patients and their families through challenging times.

Share this post!

Give Us Feedback