Johnson Thomas, MD, FACE

Endocrinologist. Editor in Chief - Endocrine AI. Section chair. Passionate about thyroid nodules, cancer & AI. Creator of Endo Tools iphone app, AIBx.

[Endocrine Pathology @EndoPath]

Can AI identify PitNET lineage from routine H&E slides alone? An attention-guided neural network achieved high scores and uncovered lineage-specific morphologic patterns, highlighting the potential of AI to support pituitary tumor classification.

https://t.co/9C4K3bqQeA

[Daniel van Strien @vanstriendaniel]

Hugging Face is the home for AI & ML across every domain, including biomedical!

The @NIH just added the @huggingface Hub to its official list of Generalist Repositories for data sharing.

NIH-funded? You can point to the Hub in your data sharing plan 🤗

[Yakup Ergün @dr_yakupergun]

In a presentation I watched at #ASCO26, it was striking that the speaker had created ALL of their slides using AI.

As an audience member, I have to say that the slides were so dense that I had difficulty deciding where to look. As a result, I did not get the expected benefit from the presentation. The aim of a scientific presentation is not to put as much information as possible on the screen; it is to select, simplify, and provide the audience with a clear intellectual flow.

AI-assisted preparation can, of course, be used. However, this support should not replace the speaker’s academic filter and responsibility for narration.

Dear speakers, do not abandon your classical presentation discipline. The AI that prepares your slides today may give the presentation on your behalf tomorrow. And you will no longer be needed!!

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

@ParthJadhav8 @bevel_health No. 😞

[Vineeta Agarwala @vintweeta]

The details of this @CMSGov "Medicare GLP-1 Bridge" (which runs entirely *outside* the part D payment flow) are actually kind of interesting

Believe this is the first time CMS has set up this kind of price and access bridge to drive rapid access to a specific prescription med class? If there are prior examples would love to learn...

Starting July 1, 2026 for 18 months, CMS will enable access to Foundayo, Wegovy, and Zepbound at a $50 co-pay for patients who meet clinical criteria, while manufacturers will receive $245/mo net pricing (this is regardless of the exact product - so clearly Lilly/Novo will prefer to sell orals)

CMS is setting up its own, separate central processing for prior auth, claims adjudication, and pharmacy payment for these meds‼️- presumably to guarantee that patients won't fall into long and unnecessary prior auth / UM traps with their plans

https://t.co/cNCIvT8PxP

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

vMLX running JANGQ-AI/Qwen3.6-35B-A3B-JANGTQ4 on M5 Max 128 GB. One shot Tetris game. #MLX #AppleSilicon #m5 #localllm #OnDevice #localAI #localInference #AI
5386 tokens, 1206.8 t/s1, 6113.7 pp/s8927, prompt (8828 paged+ssm cached) 0.55s TTFT, 70.4s total

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

AI Tutorial Series Part 2: Key metrics for evaluating AI models #EndoTwitter #MedTwitter #AI @TheAACE @AACEMENA @TheEndoSociety

https://t.co/7AcCAmeB4Q

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

@Samir_Qamar

[Qiao Jin, MD @DrQiaoJin]

🔥 Excited to share our latest work - Med-V1: Small Language Models for Zero-shot and Scalable Biomedical Evidence Attribution

Recent reports in @Nature and @TheLancet on fabricated citations have drawn substantial attention, but even real citations may fail to support the statements attached to them.

This makes evidence attribution — verifying if the citations really support the claims — essential for auditing both human- and AI-generated texts. As AI generates billions of medical references every day, we need a scalable model for this task.

In this work:
- We generated MedFact-Synth, a high-quality dataset of 1.5M synthetic claim-article pairs.
- Using MedFact-Synth, we trained and open-sourced Med-V1, a family of 3B-parameter LLMs.
- Med-V1 surpasses its backbone models by 27-71%, matching the performance of GPT-5.
- Med-V1 can be used to identify high-stakes misattributions and detect LLM hallucinations.

🔗 Paper: https://t.co/R5KmjDbjdJ
🔗 Model: https://t.co/aBu14MDMhI

🙌 Kudos to all our great collaborators: Yin Fang, Lauren He, Yifan Yang, Guangzhi Xiong, Zhizheng Wang, Nicholas Wan, Joey Chan, Donald Comeau, Robert Leaman, Charalampos Floudas, Aidong Zhang, Michael F. Chiang, Yifan Peng & Zhiyong Lu

#MedicalAI #HealthAI #LLMs #Hallucination #EvidenceBasedMedicine #ChatGPT

[Elliot Tapper @ebtapper]

Went to a lecture long ago

Speaker did a GWAS for diabetes

Discovered that iron metabolism and heme genes caused diabetes!

TINY p values
CAUSAL effects, because genes

Then, from the audience, a question

“How was diabetes defined?”

Answer: hgb A1c

Stunned silence

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

https://t.co/7AcCAmeB4Q

[Eric Topol @EricTopol]

A once a night pill for severe obstructive sleep apnea effective in a Phase 3 randomized trial
https://t.co/zq2Ew65WHK
https://t.co/ovQEh7taeN

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

@OpenlifesciAI @HongKongPolyU & https://t.co/5IBjPLdbO8

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

@OpenlifesciAI @HongKongPolyU Similar to https://t.co/MMfcuD3zLH

[Cristina Tejera @cristinatejerap]

¿Cuándo elegir semaglutida y cuándo tirzepatida?
La EASO acaba de actualizar su algoritmo con evidencia hasta noviembre 2025 📊 @EASOobesity

Spoiler: depende del objetivo clínico. No es solo pérdida de peso.

🔗 @NatureMedicine

https://t.co/RKkccNGYTq

#ECO2026

[The Lancet @TheLancet]

Presented today at #ECO2026 ⤵️

The SURMOUNT-MAINTAIN trial found that continuing tirzepatide at the maximum tolerated dose maintained bodyweight reduction and health-related benefits in adults with obesity.

Explore the research: https://t.co/RjLxWa5qjV

[Vincent Rajkumar @VincentRK]

Here is the inside story on how @UpToDate does what it does.

From my experience as an author/section editor for UpToDate for 20 years, here is how it works. It’s amazing!

Every day a team of people @UpToDate scour the top medical journals (and major meeting abstracts) for articles of importance. In my case they pertain to myeloma and related disorders. Any article that is felt to be important is sent to a expert MD medical editor at UpToDate who reviews the paper. If the information is felt to be important, the relevant chapter is updated with this content, summarized in a few sentences in Word track and then sent to the author of the chapter who is a top expert in the field.

As an author I have to review the update in detail and then make a decision whether to include the new information, and whether the way it has been incorporated is correct or needs to be edited.

Once I make a decision, it is then reviewed by an expert section editor. For 15 years for me, the section editor was Dr. Robert Kyle! (Currently I’m the section editor and we have about 6 experts serve as authors on over 40 chapters pertaining to myeloma and related disorders). The section editor independently decides whether the author made the right call and either approves or we have a back and forth till we agree.

The updated chapter is then reviewed again by the expert MD editor at UpToDate and then posted.

For important phase III trials or FDA approvals we usually make the update within a week. Sometimes by the next day! With so many important papers, every week there are 2-3 updates pertaining to my chapters that go through the above process! It’s hard work but it keeps me ridiculously current. And more importantly it keeps the content accurate and current. This is not random addition of new information. It’s vetted information that’s added.

Every UpToDate chapter is also peer reviewed annually. Each chapter also undergoes an annual author and editor review to make sure that the overall content and flow is good and to delete content that may be no longer important or accurate. The extraordinary process and rigor involved is why it’s such an invaluable irreplaceable resource. It is AMAZING and hard to replicate.

UpToDate is easy to navigate. But the newly launched UpToDate Expert AI takes it to the next level! It makes the entire content available to query just like you query a top ranked LLM. But it’s not searching the random internet or a stack of published articles. It’s primarily trained and derives vetted and adjudicated UpToDate content. It will not randomly hallucinate. It will not list 10 options for treatment newly diagnosed myeloma but will give you what experts currently feel is the best option based on the latest evidence. All answers are referenced and pointing to specific content sections in UpToDate (which also has links to source material). So you can immediately get all the background information if you need more detail.

Working with UpToDate has helped my career immensely. For the last 20 years it has been impossible for me to not know about every important study related to myeloma and related disorders as they get published! For 20 years, I have received a precise summary of every important article published in my field. More importantly I am forced to read them, review the source material, and make a judgment call. As a result I remember this stuff. A gift indeed!

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

@Prince_Canuma Which one will you suggest for TTS, STT and voice cloning locally on M5 MacBook Pro 128 gb

[Johnson Thomas, MD, FACE @JohnsonThomasMD]

@_philschmid @googlehealth @bevel_health