We are excited to announce our Meetup next Thursday at 7 PM EDT titled “Understanding the progression of Alzheimer’s.” The Meetup will be hosted by me and we will be streaming it live for those are unable to attend in person (although it’s always more fun to attend in person and Eric is bringing authentic New York pizza).

I’ll discuss a disease progression model my colleagues and I built in Stan to probabilistically track the progression of Alzheimer’s Disease using patient biomarkers. I plan to show how Stan and Bayesian Modeling is quite useful on these sorts of models. In particular, whether a given patient’s symptoms worsen or stay the same is usually unknown to us, especially in the face of sparse and/or noisy observations. In light of this, it’s important to consider all the possible disease progression paths a patient may take, not just the single most likely, and Bayesian Modeling and Stan really helps us with this. Once we understand all the possible paths and their probabilities, it becomes easy in the Bayesian framework to judge whether a risky intervention is worthwhile or decide rigorously whether a drug had a positive effect. The latter is something that’s quite practical to clinical trials, an area we are focusing on at Generable.

For those who don’t me, I am 5th year PhD student in the Computational Science and Engineering Group at the University of California, Santa Barbara. My research is partly in numerical analysis for Bayesian inference and partly in applied applied Bayesian modeling for diseases such as coagulopathy and Alzheimer’s. Over the summer, I’ll be modeling and doing numerical work that’s related to my research for Generable, so look out for more blog posts by me on all things Bayesian and how I apply Bayesian inference to medical data in my research.

I’ll try to keep this short since it’s just an announcement for our Meetup, but I should I also say a little bit about Disease Progression Modeling, a hot research area, which I personally only learned about in the last few years. In the words of this paper from Current Pharmacology Reports:

Disease modeling involves the use of mathematical functions to describe quantitatively the time course of disease progression. In order to characterize the natural progression of disease, these models generally incorporate longitudinal data for some biomarker(s) of disease severity or can incorporate more direct measures of disease severity. Disease models are also often linked to pharmacokinetic–pharmacodynamic models so that the influence of drug treatment on disease progression can be quantified and evaluated. Regulatory agencies have embraced disease progression models as powerful tools that can be used to improve drug development productivity.

Disease Progression Models are extremely useful for taking disparate forms of data (e.g. imaging or lab) and quantifying a patient’s disease severity and how they’ll progress. This in turn is useful for judging whether drugs are useful or not. Disease Progression Models are gaining steam in Statistics with techniques such as Joint Modeling, but also in the Machine Learning world. That’s all I’ll say for now, but come to our Meetup next week to see how we’re using these models in Stan to track Alzheimer’s!