Row concave Shape Decorative svg added to bottom

Editor’s Message

interior-hero

Watching playoff hockey, I am always amazed by the scope of the statistics flooding the screen in real time—fastest players to get to 100 playoff points, trend of distance from the goal when taking a shot, chance of victory after being 2 goals down versus 3 goals down. It seems there is no question or scenario that can’t be sliced and diced down to a point percentage or ranking.

But what if the question is the frequency of need for milrinone at 2 weeks after the Norwood operation even at a single center? Or the proportion of bivalirudin checks that need to be repeated due to concern for contamination? There is generally painful data extraction and cleaning and analysis, done usually without compensation, in the interest of science. What if informatics and artificial intelligence (AI) could be utilized to get us this wealth of information in real time?

I hope you will all enjoy this newsletter on the theme of emerging roles and scope of AI in the cardiac intensive care unit (ICU). The cardiac ICU is a fast-paced and dynamic environment, with teams that are used to adapting to change. It is also the place where the wealth of data available is the most abundant, and the need for high-fidelity tracking and predictions is the most critical. Hence it is perhaps the best suited to adapt AI into our heart centers and lead the way in helping everyone work smart, in addition to working hard.

Analyzing vectors of the QRS axis to predict adverse events, modeling hemodynamic parameters to assess potential need for mechanical circulatory support, generating sedation wean protocols that reduce the length of narcotic exposure as well as the number of days lost to failed weans and steps back, helping with the ‘billing’ portion of the notes so that clinicians can focus on the actual scientific and clinical communication aspects—the scope is unlimited. As we engage and learn more, we will find what data is best suited for which analyses, which monitoring is or is not additive to the picture, which data might still need a human to cross-check, and some sure-to-be exciting reveals of clinical questions we have all wanted to study but never had the time or information technology support to explore. As our invited authors also discuss, while AI is not the ultimate solution, it has immense potential for the betterment of our patients, and the times of AI ahead of us are inspiring indeed!

Happy reading, and see you all in San Diego!

Puri_Kriti

Kriti Puri, MBBS

Assistant Professor of Pediatrics
Pediatric Critical Care Medicine and Cardiology
Texas Children’s Hospital, Baylor College of Medicine, Houston, TX