.png)
For over a century, this renowned nonprofit multispecialty academic healthcare institution has integrated clinical and hospital care with research and education to provide outstanding patient outcomes at over 300 locations globally.
The healthcare provider logs over 12 million outpatient encounters and hundreds of thousands of inpatient visits annually. It employs 77,000 physicians, nurses, advanced practice providers, researchers, and support staff worldwide.
According to the most recent research for the U.S., surgical items are unintentionally left inside of patients’ bodies in approximately 1500 cases annually. Although a fraction of the tens of millions of surgeries performed, the consequences can be considerable.
Determined to eliminate the post-surgical risks of retained surgical items (RSIs) at its institution, the healthcare provider envisioned developing an AI-enabled real-time tracking and warning system using image processing and advanced data analytics to detect missing surgical items.
Preliminary investigations established that the best technology foundation for desired outcome was MATLAB by MathWorks. However, the healthcare provider lacked this highly specialized technology expertise for designing the tracking system.
Seeking to innovate rapidly, the provider desired a faster way to obtain specialized talent than the traditional months-long methods for contracting with a consulting firm or hiring internally.
Upon exploring available talent acquisition strategies, the provider learned about Graphite’s proven on-demand model and contacted the company. An experienced Account Executive (AE) was assigned to the provider and he explained Graphite’s streamlined process. This included identifying and presenting 2-4 curated independent experts, typically within 24-48 hours.
After gathering requirements, the provider met with the AE to discuss the scope of work and budget as well as the necessary technology and healthcare skills. In addition to MATLAB expertise, this included an extensive background in designing and developing a system for tracking objects in images and videos using Python, PyTorch, and TensorFlow.
With these insights, the AE turned to Graphite’s AI-powered platform to access a pool of over 9,000 independent experts and obtain a handful of qualified candidates. The AE further curated the list and, within hours of the scoping meeting, presented the top four candidates offering the appropriate MATLAB skills, fees to match the budget, and availability.
Using the Graphite in-app messaging tools, the provider vetted each candidate and quickly selected the expert with the best combination of skills and cultural fit with the internal team.
To kick off the project, the provider and the independent MATLAB expert reviewed the project objectives, scope of work, timeline, and other deliverables. Next, the MATLAB expert gathered information from the provider’s internal teams to understand desired tracking capabilities and other outcomes.
A two-phase project, the first phase was developing a prototype that would serve as a real-world test of the new tracking system. The second phase was refining the system based on findings during the pilot.
In less than three weeks, the MATLAB expert delivered a prototype and the pilot began. When the pilot ended, the MATLAB expert collected objective data and subjective feedback. He quickly optimized the tracking system, enabling the healthcare provider to use in-house resources for rolling out the system enterprise wide.
Along the way, the Graphite platform supplied the provider with detailed project tracking and an integrated interface for invoicing and payments. These features ensured that the healthcare provider could keep administrative overhead low.
When the project was completed, the independent expert seamlessly rolled-off. For the provider, it was a smooth and professional experience from start to finish.
Overall, the healthcare provider realized multiple benefits from engaging with Graphite, including:
Explore how this national healthcare provider leveraged an independent consultant to develop an image-processing deep-learning model to detect potential sources of surgeon error.
Eager to onboard a team that could achieve results within six months, the client and their dedicated Graphite account manager partnered to determinze the specifications for each of the four roles. They were:
.png)
