Hi Explorer! A new webinar series about analytical techniques for the pharmaceutical industry is starting soon. Plus, a couple of videos about deep learning.
It might be hard to understand what this quote means without the context, but I laughed out loud when I heard it in a presentation. Some are born curious and must know why when seeing a peculiar manifestation of math or physics. I'm one of them.
This webinar series covers some of the key analytical techniques, such as X-ray diffraction (XRD), X-ray fluorescence (XRF), thermal analysis, Raman scattering, and CT, and their applications for pharmaceutical process control and research. Each episode covers different stages, including discovery, preclinical development, preformulation, formulation development, manufacturing, and QC.
XRD, XRF, and CT are all X-ray analyses. While XRD and XRF excel at identifying phases, polymorphs, or elements, CT provides a 3D view of internal features, such as non-uniform mixing or packing, aggregates, cracks, etc. These techniques play different roles at different stages of the pharmaceutical development process.
XRD and XRF have been the bread and butter of pharmaceutical analysis for decades, but the application of CT is a relatively new development. Thanks to the advancement of AI-driven segmentation, CT is now uncovering defects and structures that were nearly impossible to analyze a decade ago, providing valuable insights to optimize the manufacturing process.
Join us to learn best practices and the latest advancements in pharmaceutical materials analysis. In episode 4, Angela will discuss how CT is applied to manufacturing and QC.
Many people probably use deep learning to segment CT images today. But do you understand how it works? You don't need to know how these neural networks are trained and applied to segment complex and challenging CT images, but if you are curious, you might find these videos interesting.
This video also includes a brief introduction to Python and a hands-on workshop on open tools like Colab and ZeroCostDL4Mic, but the first thirty minutes cover how deep learning learns and works for image segmentation. I like this video because, unlike the general explanation videos, it uses scientific image segmentation as a specific application area.
I have referenced these videos many times, but if you have not watched them, I highly recommend the first four videos of the Neural Networks playlist by 3Blue1Brown. These are arguably the best explanations of how deep learning works.
The man behind this channel, Grant Sanderson, does an excellent job visually explaining how we can train neural networks and slowly introduce the mathematics behind it. I had to watch them a couple of times to understand it, but it was a deeply satisfying experience.
Real Scientists, Not Actors
A collection of priceless and embarrassing moments curated by Carlos Astudillo.
Answer: Grant Sanderson
A math educator and YouTuber.
I'm not going to spoil it in case you haven't watched it yet, but he shouts, "If you have a soul, you have to know why!" in his TEDx Talk when talking about a bizarre and intriguing math phenomenon.
That's a wrap. Please let us know how we can help you learn more about X-ray CT. We love to hear from you!
