Diagnosing Problems During Medical Device Development

The word diagnosis is made of up of two Greek roots meaning “to know throughout” or to “understand completely”.  We usually think of diagnosis in the context of illness, but it applies to problem solving within medical device development as well.  

Almost every medical device that we work on at Simbex has some digital component, requiring us to continually update our diagnostic techniques in order to keep pace with innovation. Even the mechanical and electrical portions of medical devices have become more complex and components have been miniaturized, making it challenging to isolate portions of the system to find the origin of an issue rather than simply addressing surface-level symptoms.  This concept of isolating portions of a system to identify and fix problems has been used for hundreds of years in the mechanical engineering field. In the past, engineers would dismantle a machine to identify and fix a problem. There is still a place for screwdrivers and socket wrenches, but today they may need to be paired with advanced software tools to simulate and analyze the behavior of individual components in real time. 

One of the key aspects of the diagnostic process is thorough error logging. Reviewing a timestamped list of error occurrences is an excellent first step in an analysis to isolate the source of a problem. If there is a software component to the device, error logging can be automated and stored in the database. If not, a test engineer can simply document their observations. Common errors occur in conjunction with pairing wireless devices, other wireless communication issues, permission levels at login, device states, device resets, time syncing, on device component to component communication (accelerometer to microprocessor for example). Each of these possible failure modes can have a distinct log message created to verify expected behavior or log abnormal behavior. Error logs are shared with members of our interdisciplinary development teams and stored in an accessible location, allowing our engineers to collaborate and quickly diagnose and fix problems. We can also review the error logs after a software update to verify that deployment was smooth.  

We are not pessimists, rather realists, which is why we make it a point to include debugging features in any of our product designs. During the Failure Modes and Effects Analysis (FMEA) process we identify technical risks and mitigation techniques. If we are alert to the possibility of a certain type of risk and we will have to check for it frequently, we try to make this process as painless as possible. For example, many of our wearable devices implement Inertial Measurement Units (IMU) that provide raw acceleration and gyro data to an on-board algorithm. The algorithm then does some processing and outputs a user-facing metric. For the algorithm to be working correctly, we need to be sure that the IMU is configured correctly and is sending data to the algorithm. An easy way to confirm this is to add a Bluetooth characteristic to view the raw IMU data. By doing this we can see that at least the IMU to microprocessor to Bluetooth pipeline is working correctly. 

At Simbex, we are committed to building reliable and safe medical devices that improve the lives of patients and healthcare professionals. Our approach to diagnosing and fixing performance issues throughout development is key to achieving this goal. By isolating portions of the system, using advanced techniques to analyze components, and building in thorough error logging and debugging features, we are able to identify and fix problems quickly and effectively.

About the Authors:

Jesse Kuhn is the Systems Team Lead at Simbex. Jesse is a hands-on, highly self-motivated Mechanical Engineer with an emphasis on electromechanical product development from concept through volume production. Jesse has 15 years of professional experience including systems design for an ultra-pure distilled water device, complex rehabilitation powered wheelchair systems, and high-volume industrial grade LED lighting systems. Jesse has a BS in Mechanical Engineering from Northeastern University. 

Jay Avis is a Systems Engineer at Simbex.  He has a mechanical engineering degree from Northeastern University, with 14 years of professional expertise. His early career focused on consumer product development with lots of ideation and concept iteration using CNC machining, rapid prototyping, and manual manufacturing technologies.  He has extensive 3D modeling experience in Solidworks, often with a focus on design for injection molding.  His efforts at Simbex have mostly focused on developing and supporting football helmet technologies, and has led scrum events for a team of engineers.  He has served as a liason to domestic and overseas vendors and contract manufacturers, including off-site assembly and testing support.  Jay also owned a small design consulting business and developed production processes and physical products for his own entrepreneurial ventures.