Chris Le

👁 resident at Yale| MD/MSE at UMD and JHU| Let's talk eyeballs, machine learning, and med ed. He/him

[UIowa Eye @uiowaeye]

In June 2023, the regional Medicare Administrative Contractors (MACs) almost uniformly introduced draft policy language indicating cessation of coverage for many MIGS procedures, including goniotomy and cyclophotocoagulation.

[erkin ötleş @eotles]

lots of talk about Vision OS, but what about Vision OD?

[Kevin Fischer @kevinafischer]

I don't talk much about this - I obtained one of the first FDA approvals in ML + radiology and it informs much of how I think about AI systems and their impact on the world. If you're a pure technologist, you should read the following:

There's so much to unpack for both why Geoff was wrong, and why his future predictions should not be taken seriously either.

Geoff made a classic error that technologists often make, which is to observe a particular behavior (identifying some subset of radiology scans correctly) against some task (identifying hemorrhage on CT head scans correctly), and then to extrapolate based on that task alone.

The reality is that reducing any job, especially a wildly complex job that requires a decade of training, to a handful of tasks is quite absurd.

Here's a bunch of stuff you wouldn't know about radiologists unless you built an AI company WITH them instead of opining about their job disappearing from an ivory tower.

(1) Radiologists are NOT performing 2d pattern recognition - they have a 3d world model of the brain and its physical dynamics in their head. The motion and behavior of their brain to various traumas informs their prediction of hemorrhage determination.

(2) Radiologists have a whole host of grounded models to make determinations, and actually, one of the most important first order determination they make is whether there is anything notably wrong with a brain structure that "feels" off. As a result, classifiers aren’t actually performing the same task even as radiologists.

(3) Radiologists, because they have a grounded brain model, only need to see a single example of a rare and obscure condition to both remember it and identify it in the future. This long tail of rare conditions to avoid missing is a large part of their training, and no one has any clue how to make a model that acts similar in this way.

(4) There’s so many ways to make Radiologist lives easier instead of just replacing them, it doesn’t even make sense to try. I interviewed and hired 25 radiologists, whose primary and chief complaint was that they had to reboot their computers several times a day.

(5) A large part of the radiologist job is communicating their findings with physicians, so if you are thinking about automating them away you also need to understand the complex interactions between them and different clinics, which often are unique.

(6) Every hospital is a snowflake, data is held under lock and key, so your algorithm might not work in a bunch of hospitals. Worse, the imagenet datasets have such wildly different feature sets they don’t do much for pretraining for you.

(7) Have you ever tried to make anything in healthcare? The entire system is optimized to avoid introducing any harm to patients - explaining the ramifications of that would take an entire book, but suffice to say even if you had an algorithm that could automate away radiologists I don’t even know if you could create a viable adoption strategy in the US regulatory environment.

(8) The reality is that for every application, the amount of specific and UNKNOWABLE domain knowledge is immense.

LONG STORY SHORT: thinkers have a pattern where they are so divorced from implementation details that applications seem trivial, when in reality, the small details are exactly where value accrues.

Should you be worried about GPT5 being used to automate vulnerability detection on websites before they’re patched? Maybe.

Should you be worried GPT5 is going to interact with SOCIAL systems and destroy our society single-handedly? No absolutely not.

[David Ouyang, MD @David_Ouyang]

We are excited to announce that our blinded randomized clinical trial of AI LVEF (EchoNet-Dynamic) has been published in @nature! @bryandhe @james_y_zou @StanfordDBDS @CedarsSinaiMed

https://t.co/n7ntvPELUm

1/n

[Chris Le @ChristopherTNLe]

@ericsongg @Ophthalmologymg Smeagol is going to have such a busy practice

[Weill Cornell Medicine's Radiology Residents @WCMRadRes]

Thrilled to welcome our newest matched class of #match2022 #radiology #radres!!!

[Chris Le @ChristopherTNLe]

@TrishaMiglani @UMmedschool @GoldFdtn Congrats 👏

[Maarten van Smeden @MaartenvSmeden]

NEW PREPRINT
The increasingly popular class imbalance approaches (such as SMOTE) for risk prediction modeling: they are likely to do more harm than good
https://t.co/CJiH4bhloL

[Chris Le @ChristopherTNLe]

@EmXuMed @OhioStateEye YESSSSSS CONGRATS!

[Chris Le @ChristopherTNLe]

@lyna_azzouz Congrats!

[Chris Le @ChristopherTNLe]

As always - immensely grateful to co-authors for their collaboration on this study and to @ojsaeedi for his mentorship and getting me interested in the world of ocular blood flow in glaucoma

[Chris Le @ChristopherTNLe]

Excited and humbled to share our recently published work in @AAOjournal #Glaucoma looking at the intra/intersession repeatability and differences in real-time dynamic blood flow velocity index measures in glaucoma, suspects, and controls https://t.co/rmmWAj7cMc

[Chris Le @ChristopherTNLe]

I remember as an M1 thinking how neat it was to see erythrocyte velocity profiles roughly align with systole-diastole. Now, we not only can reliably measure high-res flow waveforms, but have identified waveform characteristics that could help with glaucoma diagnosis in the future

[Chris Le @ChristopherTNLe]

@ojsaeedi @NatEyeInstitute @NIH @UMmedschool @UMDEye @US_FDA Congratulations!

[Chris Le @ChristopherTNLe]

Fantastic and approachable episode for anyone interested in learning about pediatric ophthalmology from an incredible educator at @UMDEye

[The Cribsiders - Pediatric Podcast @TheCribsiders]

Is the pediatric eye exam a bit fuzzy to you? Join us as we talk through all things “kids eyes” with peds ophthalmologist Dr. Roni Levin! She teaches us about the red reflex, the 4 aspects of an 👁 exam, and what findings should worry us.

#MedEd #peds
https://t.co/f2xuyW4Xbg

[Chris Le @ChristopherTNLe]

Me: wow look at this cool ophtho thing
@ElizabethYLiu: wow look at this cool rads thing

🤝:

[Ophthalmology @AAOjournal]

Work from @ojsaeedi shows that a newly FDA cleared retinal laser speckle contrast imaging device may hold promise as a new tool in the diagnosis of glaucoma. @ChristopherTNLe #bloodflow #glaucoma #biomarker #ophthalmology https://t.co/b6PFxuIgBS

[Pranav Rajpurkar @pranavrajpurkar]

Why is medical data sharing so incredibly hard?

Our latest episode of @AIHealthPodcast with Prof. Zak Kohane (@zakkohane) goes into the incentives, standards, and breakthroughs that are core to medical data sharing!

I'm excited to share this with you!
https://t.co/KiwApGJpid

[Chris Le @ChristopherTNLe]

Huge thanks to the mentorship of @ojsaeedi, co-authors and collaborators at the @FDA, and a talented and bright co-first author Ricardo Villanueva, now M1 at USF!

[Chris Le @ChristopherTNLe]

Cell – Vessel Mismatch in Glaucoma: Correlation of Ganglion Cell Layer Soma and Capillary Densities | IOVS | ARVO Journals https://t.co/QFiL3Y0PoS

Hot of the press, here's a 🧵 on our recent paper in IOVS (1/a lot I'm sorry)

[Chris Le @ChristopherTNLe]

This study was limited by its cross sectional nature and sample size, so no definite answers on timing/mechanism yet but definitely cool to see how the capability to visualize single ganglion cells in small fields of view using AO-OCT may add to the narrative