7. Natural Language Processing (NLP), Part 1

PETER SZOLOVITS: OK. So today and next Tuesday, we're talking

about the role of natural language processing in machine learning in health care.

And this is going to be a heterogeneous kind of presentation.

Mainly today, I'm going to talk about stuff that happened or that takes advantage of methods

that are not based on neural network representations.

And on Tuesday, I'm going to speak mostly about stuff that does depend on neural network representations,

but I'm not sure where the boundary is going to fall. I've also invited Dr. Katherine Liao

over there, who will join me in a question and answer session and interview like we did a couple of weeks

ago with David. Kat is a rheumatologist in the Partners HealthCare system.

And you'll actually be hearing about some of the work that we've done together in the past

before we go to the interview. So roughly, the outline of these two lectures

is that I want to talk a little bit about why we care about clinical text.

And then I'm going to talk about some conceptually very

appealing, but practically not very feasible methods that involve analyzing these narrative

texts as linguistic entities, as linguistic objects

in the way that a linguist might approach them. And then we're going to talk about what is very often done,

which is a kind of term spotting approach that says, well, we may not be able to understand exactly everything

that goes on in the narratives, but we can identify certain words and certain phrases that

are very highly indicative that the patient has a certain disease, a certain symptom,

that some particular thing was done to them. And so this is a lot of the bread and butter

of how clinical research is done nowadays.

And then I'll go on to some other techniques. So here's an example.

This is a discharge summary from MIMIC.

When you played with MIMIC, you notice that it's de-identified. And so names and things are replaced

with square brackets, star, star, star kinds of things. And here I have replaced-- we replaced those

with synthetic names. So Mr. Blind isn't really Mr. Blind,

and November 15 probably really isn't November 15, et cetera.

But I wanted something that read like real text. So if you look at something like this, you see that Mr.

Blind is a 79-year-old white white male-- so somebody repeated a word--

with a history of diabetes mellitus and inferior MI, who underwent open repair of his increased diverticulum

on November 13 at some-- again, that's not the name of the actual place--

medical center. And then he developed hematemesis, so he was spitting up blood, and was intubated

for respiratory distress. So he wasn't breathing well. So these are all really important things about what

happened to Mr. Blind. And so we'd like to be able to take advantage of this.

And in fact, to give you a slightly more quantitative version of this, Kat and I worked on a project

back around 2010 where we were looking at trying to understand what are the genetic correlates

of rheumatoid arthritis. And so we went to the research patient data repository

of Mass General and the Brigham Partners HealthCare,

and we said, OK, who are the patients who have been billed for a rheumatoid arthritis visit?