[R]eliability

A Random Walk In the Medical Device Space

Learning from Failure - Nitinol Fracture Mechanics in R

Despite our best efforts, nitinol implants fracture and fail. Sometimes we want them to fail (on the bench, to learn). Other times they fail unexpectedly and we need find out why. When the failure is a fractured nitinol structural element, high magnification imaging of the fracture surface via optical microscopy and SEM is essential. A trained engineer can quickly identify the nature of the fracture (fatigue or overload) and the presence of obvious special causes like witness marks or foreign material transfer become apparent.

A Real World Use Case for a Bayesian Reliability Model - How to Incorporate FEA into Risk Estimates

Frequentist statistical methods, despite their flaws, are generally serviceable for a large suite of practical problems faced by engineers during product development of medical devices. But even in domains where simple models usually do the trick, there remain instances where a Bayesian approach is the best (and perhaps only logical) way to tackle a problem. In the rest of this post, I will lay out a technical use-case and associated modeling workflow that is based on a real business problem encountered in the medical device industry.

Durability Testing of Stents Using Sensitivity-Based Methods in R

The current industry protocol for durability testing of vascular stents and frames involves testing many implants simultaneously at a range of different stimulus magnitudes (typically strain or stress). The test levels are spread out like a grid across the stimulus range of interest. Each implant is tested to failure or run-out at its specified level and a model is fit to the data using methods similar to those described in Meeker and Escobar.

Trying to Trick Linear Regression - Estimating Coefficients for Variables in R

In this post we will try to trick linear regression into thinking that a redundant variable is statistically significant. By redundant, I mean a candidate predictor variable that in reality is just noise (no effect on the outcome) but that we might include in an experiment because we don’t know if it is important or not. The trick is that we can set up the data generating process such that a redundant variable is highly correlated with the response.

Could AutoML win in the 'Sliced' Data Science Competition? The answer may shock you!

In this post I’ll be taking a break from my normal explorations in the medical device domain to talk about Sliced. Sliced is a 2-hour data science competition streamed on Twitch and hosted by Meg Risdal and Nick Wan. Four competitors tackle a prediction problem in real time using whatever coding language or tools they prefer, grabbing bonus points along the way for Data Visualization and/or stumbling onto Golden Features (hint: always calculate the air density when training on weather data).

Exploring Frequentist and Bayesian Tolerance Intervals in R

Tolerance intervals are used to specify coverage of a population of data. In the frequentist framework, the width of the interval is dependent on the desired coverage proportion and the specified confidence level. They are widely used in the medical device industry because they can be compared directly vs. product specifications, allowing the engineer to make a judgment about what percentage of the parts would meet the spec taking into account sampling uncertainty.

Tutorial - Design Study in Solidworks with Data Analysis in R

I decided to do something a little bit different with this post and show how R can be used in tandem with a traditional engineering CAD program. Together they comprise a streamlined and repeatable workflow that I’ve tried to leverage on my job when it makes sense to do so. Solidworks is a 3d CAD program that is used commonly in industry. One powerful feature that I don’t see used that much is the Design Study module.

Boundary Conditions and Anatomy - Correlated Data and Kernel Density Estimation in R

Measurements taken from patient anatomy are often correlated. For example, larger blood vessels might tend to have less curvature. Additionally, data are rarely Gaussian, favoring skewed shapes with some very large values and a lower bound of zero. These properties can make simulation and inference hard. In this post I will walk through a workflow for an engineering problem that might be presented in my industry.

Simulation With a Random Effects Model - Gage R&R as a Case Study

I've heard it said that common statistical tests are just linear models.1 It turns out that Gage R&R, a commonly used measurement system analysis (MSA), is no different. In this post I'll attempt to provide some background on Gage R&R, describe the underlying model, and then walk through a method for simulation that can be useful for things like power analysis or visualization of uncertainty. What is Gage R&R?

Bayesian Stress-Strength Analysis for Product Design (in R and brms)

Whether you are building bridges, baseball bats, or medical devices, one of the most basic rules of engineering is that the thing you build must be strong enough to survive its service environment. Although a simple concept in principle, variation in use conditions, material properties, and geometric tolerances all introduce uncertainty that can doom a product. Stress-Strength analysis attempts to formalize a more rigorous approach to evaluating overlap between the stress and strength distributions.