How the AI Fridays program helped SOS Care Now’s founder advance his organization
When Dr. John Moustoukas, a 2006 Vanderbilt School of Engineering alum, launched SOS Care Now roughly six months ago, a Chicago-area house-call practice that sends emergency-trained physicians to patients’ homes for acute and urgent care, growth quickly outpaced his ability to manage it by hand. Coordinating schedules across a growing team, with patient preferences and travel time layered on top, had become a problem of its own. Vanderbilt’s Data Science Institute (DSI), part of the College of Connected Computing, helped him build a tool that solved it.
“It’s a full spectrum of time management, including provider availability, patient preference, and also incorporating logistics times and travel times into that,” Moustoukas said. “Trying to incorporate as much technology into that efficiency was a challenge.”
Moustoukas connected with the DSI’s AI Fridays program, open weekly office hours where faculty, students, staff and Vanderbilt affiliates can bring a data or AI problem and work through it with the DSI team. There he began working with Umang Chaudhry, a senior data scientist at the DSI, with support from a student in the M.S. in Data Science program to help build the tool. Over about 30 days, following an initial round of discovery calls to map out his scheduling workflow, the team built a piece of software that syncs with Google Calendar, calculates travel time between each provider’s location and a patient’s home, and automatically sends calendar invites once a visit is booked. Data privacy came up early and often — because the tool handles patient information, Chaudhry said the team spent time upfront making sure the design would hold up under HIPAA.
“In the healthcare space, there are always concerns about data privacy,” Chaudhry said. “That was one of the first things we discussed at length, making sure that whatever we’re doing preserves HIPAA laws and all the kind of things that we need to be cognizant of when working with patient data.”
Moustoukas is still in the early stages of putting the tool to use across his team, but he said the payoff is already changing how he practices medicine. Less time on scheduling and administrative work means more time with patients, and more capacity to take on new ones.
“These sorts of solutions allow me to spend more time at the bedside,” Moustoukas said. “It’s been great, and I think as the team starts to rely on it, it’s going to really help us.”
Chaudhry said the DSI’s goal in every engagement, whether with a faculty researcher, a student or an alumnus like Moustoukas, is to leave people equipped to keep building on their own, not just to hand over a finished product. He noted that projects like this one, built for an alumnus running his own company, are still relatively rare for the DSI, and that expanding that kind of engagement is a priority going forward. Chaudhry also pointed to companies such as Nissan, Bridgestone, and Asurion, which have worked with DSI students on class-based projects and gone on to hire several of them.
Moustoukas said the real opportunity for other alumni is simply knowing that a resource like this exists. “I think if more alumni knew that a program like this was out there, they would be more inclined to ask for help,” he said. AI Fridays runs every Friday from 10 a.m. to noon and is open to anyone in the Vanderbilt community with a data or AI problem to work through, no matter how early-stage the idea may be.