Daniel Coelho de Castro

Researcher in #ML for #healthcare @MSFTResearch | Honorary Research Fellow & PhD alumnus @ImperialCollege @BioMedIAICL. He/him

[Daniel Coelho de Castro @dccastr0]

1/3 How can causal reasoning help us better understand pitfalls in medical image analysis?

Proud to share our latest @NatureComms Perspective, with Ian Walker and @GlockerBen:
https://t.co/1P0hFxRzUz

@BioMedIAICL @ICComputing @imperialcollege

[Microsoft @Microsoft]

From generating novel materials with MatterGen to helping to improve diagnostic accuracy with RAD-DINO, Microsoft’s latest AI models are set to transform the fields of science & healthcare. https://t.co/bwuJ0Y78D1

[Satya Nadella @satyanadella]

The foundation models we'll build with @MayoClinic promise to help transform how radiologists do their work, using multimodal AI to help them analyze X-rays faster and more accurately.

[_hylandSL - not here @_hylandSL]

Our MAIRA-2 model for (grounded) radiology report generation from CXRs is out now! https://t.co/1vHcB6vXFt

[_hylandSL - not here @_hylandSL]

if you find yourself wanting to understand how and why models work, in a way that could be useful for biomedical discovery, come to Cambridge (UK) for a postdoc at MSR: https://t.co/30ZdOVSi4A

[Fernando Pérez-García @fepegar_]

THIS X-RAY DOESN'T EXIST

@eccvconf #ECCV2024

🧵1/9

[Microsoft Research @MSFTResearch]

RadEdit stress-tests biomedical vision models by simulating dataset shifts through precise image editing. It uses diffusion models to create realistic, synthetic datasets, helping to identify model weaknesses and evaluate robustness: https://t.co/s4g0zLP6q6

[_hylandSL - not here @_hylandSL]

my team's radiology report generation metric is now open source:

microsoft/RadFact: A metric suite leveraging the logical inference capabilities of LLMs, for radiology report generation both with and without grounding (https://t.co/HfvVKhXuEp)#MoreMetrics
https://t.co/O8BUiFKuGN

[Daniel Coelho de Castro @dccastr0]

@SnchzPedro_ Parabéns, Dr Sanchez! It was wonderful to work with you, and I wish you success for your new endeavours!

[Daniel Coelho de Castro @dccastr0]

@GlockerBen @BioMedIAICL @ICComputing @imperialcollege @KheironMedical @HeartFlow Congratulations on the well-deserved honours, @GlockerBen!

[Fernando Pérez-García @fepegar_]

We've released the Rᴀᴅ-DINO model weights!!

Benchmark it, encode some datasets, show us some UMAPs, plug it into your classifiers, LLMs, MLMs, SLMs...

We're excited to discover what the community will create on top of Rᴀᴅ-DINO.

🤗https://t.co/adKMxlMf3o

@MSFTResearch

[_hylandSL - not here @_hylandSL]

We are hiring a senior researcher in ML for healthcare at MSR Cambridge (UK)! The position is in my team, so if you get it you will work with me (is this a pro or a con? do not answer). Focus is multimodal (~vision-language) models for radiology! Link: https://t.co/iq39xsVOSF

[Daniel Coelho de Castro @dccastr0]

@SnchzPedro_ Thanks for the shout-out, Pedro! It was a great pleasure to host you for an internship as part of this journey. Good luck for your thesis!

[Charles Jones @c_jones_ai]

It's great to see our work on causality in fair machine learning showcased in the @NaturePortfolio collection on AI and robotics!

https://t.co/qWR47aAez3

[Daniel Coelho de Castro @dccastr0]

🔬 Optimising pathology workflows for early detection of oesophageal cancer, using large-scale ML models for whole-slide image analysis. Many interesting clinical, modelling, and engineering challenges, working in close collaboration with stakeholders. Proud to see it published!

[Microsoft Research @MSFTResearch]

Microsoft Research worked with Cyted & Cancer Research UK to build AI models that improve early detection of esophageal cancer—the 6th most common cause of cancer deaths. The models reduce pathologists’ workload by up to 63% and could save countless lives: https://t.co/dBYimBmvEh

[Matthew Lungren MD MPH @mattlungrenMD]

Vision-Language Models will revolutionize radiology and enhance patient care.

How do we ensure they achieve their goals in practice?

Latest research from @MSFTResearch led by @runmiridliy explores clinician+AI interaction for VLMs in healthcare!

https://t.co/knrzwbkiIp

[Daniel Coelho de Castro @dccastr0]

@amt_shrma Thanks for the great thread! You may also be interested in our #ICLR2024 paper, where we demonstrate that combining (1) LLM causal hypothesis generation & critique with (2) grounding on evidence from real data greatly improves causal discovery.
https://t.co/5VpJfUUMfM

[Daniel Coelho de Castro @dccastr0]

@amt_shrma Thanks for the great thread! You may also be interested in our #ICLR2024 paper, where we demonstrate that combining (1) LLM causal hypothesis generation & critique with (2) grounding on evidence from real data greatly improves causal discovery.
https://t.co/5VpJfUUMfM