So why isn’t Design of Experiments, an often used tool in engineering product development, being used more often in the area of genomics? The following article suggests that “biologists and math usually don’t mix”. In my opinion this is where an opportunity exists to differentiate and define the B.S. Biomedical Engineer. However, there is a catch. Most B.S. Biomedical Engineers it is rare to find more than the most basic statistical course in the Biomedical Engineering curriculum.
Used much more often in industry than it is taught in depth in engineering curricula Design of Experiments was a concept introduced by Fisher in 1935. I was introduced to the subject as a Principal Product Development Engineer at the Hospital Products Division ( now Hospira ) about 17 years ago. I have to admit I felt like I had been severely short changed in my statistics education when I saw its potential.
Basically Design of Experiments is a structured process that applies standard statistics and factorial statistical design to find the optimal combination and magnitude of multiple previously identified factors that impact an important measurable aspect of a product. If any of this is foreign to you or sounds complicated well it probably is. The good news is that there are a number of software packages that are available to help out with the process. Once you learn about its power you get motivated to find ways to use it. It is best used with a good understanding of statistics under your belt or under the watchful eye of a statistician.
The following editorial article from Nature Methods demonstrates how even though the biological community knows that statistical approaches are important to genomics there is still a bottle neck in applying them effectively. It also demonstrates how the biotechnology industry has been effected by the aversion or ignorance of this particular technique. The good news is that things do appear to be changing. It is interesting to note that Charlie Carter, a biochemist, picked up on DoE from his dad a quality control engineer.
In the medical device field I believe it should be the Biomedical Engineer that should be actively leading the charge to apply this technique. Any BME’s working in the genomics field who does not have this technique in their armamentarium should get familiar with it tout suite. Biomedical Engineering Technologists should also be aware of this approach as they will often be tasked with performing the laboratory or field experiments and will be able to perform more effectively if they know what they are doing and why. They will also be more effective at providing feedback and participating actively in the process.
See this article