I read Vijay Pande’s piece on the A16Z blog from start to finish multiple times. It’s worth checking out if you haven’t seen it already.
I have zero biochemistry background, so take most of my commentary with a reasonable grain of salt. I run a site that helps chronically ill patients aggregate their medical records and find relevant clinical trials. We also make a product that helps patients and their families translate difficult to read research papers into understandable content.
My Initial Thoughts.
I perceive that the speed of the tools matters less than we think it does. This is purely speculative and an opinion.
A pharma expert recently pushed the question to me:
How is it possible that the technologies that most people think are important for drug discovery have become hundreds, thousands, or billions of times cheaper, while the cost of R&D, per drug discovered, increased roughly 100 fold between 1950 and 2010?
Meta-research suggests we need more predictive validity/predictive modeling than we do faster tools. In simple words, “we need more maths.”
One of the parts of the interview that struck me was this question about software connecting the dots.
a16z: How can you make the claim that software connects the dots? Because when I think of bio, I think of tissue and flesh; I don’t think of computation and algorithms. How do those two actually come together?
Vijay: Let’s take machine learning. You can now do so much with image recognition there. And a big part of medicine involves images. Sure, when you go to your doctor, a bit of listening happens, but most of it is really about analyzing your x-rays (radiology), examining your skin (dermatology), or looking at your eyes (ophthalmology).Of course, these doctors aren’t just using their eyes; they’re applying and honing decades of medical training to do the pattern recognition, which in many cases is very subtle and requires significant expertise. There’s going to be many examples like this where computation can do something beyond what a human being can. It’s not limited to just vision. Think of all the inputs that humans take in with their senses; each of those are amenable to machine learning and deep learning: Listening with a stethoscope. Smelling something. And so on.In many cases, algorithms can do better than humans. Just as computer vision has had a huge impact in non-medical areas, it’s now getting to the point where it can set a new gold standard. If the gold standard in radiology is to predict what radiologists would do, computers can go beyond that. In radiation oncology for example the gold standard would be to predict the biopsy results … without having to actually put the patient through one.
I think Pande’s comments are very in line with reality and we’re already starting to see such technology in place, at companies like Semantic.md.
Given that cost of MRIs in the USA is aberrationally high, will we still need predictive algorithms of biopsies then?
The efficacy of biopsies as a precaution still seems to be up in the air for colorectal cancer/UC….
Curious to see more maths done in this entire arena. In any case, Pande is worth following on twitter.