Developing A Novel Synchronization Method

Problem Explored:

Pressures across each pump had to be synchronized prior to a treatment to ensure steady flow to the patient. Our prior pressure synchronization method and pump setup revealed some key problems:

• Highly inconsistent results across the 3 operational prototypes

• Too much reliance upon a single pump to set a balance point in the dialyzer 

• High amplitude signals, causing potentially inaccurate software calculations 

• Pressure drifts, resulting in inaccurate pressure reference values 

Action:

To address these issues, I helped to design and test a new pump setup and synchronization method. The goals were: 

• To generate more consistent results across the existing prototypes, such that internal variability between machines would not result in out-of-spec treatments. 

• To create a more consistent, low-amplitude pressure signal within the span of 2-3 minutes that could be used before an official treatment to determine pressure references accurately

I conducted iterative manual testing (wherein automated calculations and adjustments usually handled by software were instead manually conducted by an operator such as myself) to continuously update and redesign the fundamental procedure behind the synchronization method. In collaboration with upper management, we explored and tested a wide variety of methods.

Result:

Through an iterative process, our team settled upon a new synchronization method that was generating more consistent results. The manual procedure was automated and testing shifted to focus on optimizing how it was implemented in our software. The resulting synchronization procedure produced treatment errors within spec and with a decreased sensitivity to external noise. For example, the treatment errors were no longer greatly affected by sudden increases or decreases in pressure of 25 mmHg to degrees outside of company specifications.