Gavin Seegoolam

AI, Deep Learning, Medical Imaging, Tech and Innovation

[Julian Suk @sukjulian]

One month left 'til @iclr_conf, about time to launch our ✨@GRaM_org_ Competititon✨ The theme is geometry x AI4science with a dataset kindly provided by @BeyondMathLtd .

Deadline: April 22, 2026 (AoE)

🔗 https://t.co/S5KaCpGtcU

[BeyondMath @BeyondMathLtd]

A great reception for @BeyondMathLtd Co-Founder @alan_patterson at #CTW23 today talking about how we’re applying AI to the physical world.

#AI #technology #CFD #Simulation #Physics

[Alexander Terenin @avt_im]

I'm going to try an experiment, and open-source my PhD dissertation writing by making the thesis document visible to the public while I write it.

Check it out and follow along!

https://t.co/tqFRseamCp

[Zeju Li @li_zeju]

This is the extension of our previous MICCAI paper, we study with more datasets (kidney tumor, brain lesion, small organ) and more models (our beloved DeepMedic and nnUnet), showing new insights and solutions for the class imbalance problems.

Code: https://t.co/YcHxnhwa2y

[HCI Research @HCI_Research]

Ken Perlin:​ Chalktalk in Augmented Reality​
Full video here: https://t.co/RcjSqCJ1vl
Chalktalk is open source: https://t.co/7opXgGlHxX

[Sotirios (Sotos) Tsaftaris @STsaftaris]

Do you want to learn about disentangled representations and applications in medical imaging? Join us for the DREAM @MICCAI2020 : a comprehensive tutorial on the matter at #miccai2020 in Lima. (Carefully crafted by yours truly and @alisonQoneil). Site: https://t.co/W7Bwc2bpS6

[Sotirios (Sotos) Tsaftaris @STsaftaris]

Please circulate widely. PostDoc position: Can we measure/check if machine learning models leak data? An exciting opportunity to join iCAIRD project (https://t.co/JUkaw42Gzk) to bring AI within the NHS. Deadline: 1 April 2020 @iCAIRD_Scot @CanonMedicalEDI https://t.co/LhzSQAmpIk

[Athanasios (Thanos) Vlontzos, PhD @vlontzos]

@alkalait @geoffreyhinton To add to that, from experience we’ve seen ML solutions that fail to generalize to other hospital sites due to slightly different settings of the machines, even though the task is the same. At the moment I’m going to go with the human, but I’m working on making the ML better.

[Giacomo Tarroni @g_tarroni]

Are you using #UKBiobank for your research? Are you interested in #qualitycontrol metrics for #cardiac #MRI images? If so, have a look at our new paper (https://t.co/ljJPdujHLI) published by Nature's @SciReports!

[Chen Qin @Chen_Qin615]

Check our great and awesome work! I am delighted to be in the team holykspace. Thanks for all the efforts!

[Michal Sofka @MichalSofka]

Congratulations to the winners of the #fastMRI knee reconstruction challenge! The disagreement of the Structural Similarity Index and radiologist ranking suggests we need to improve measures of image quality.

[Qi Dou @QiDou_]

lots of enjoyment and water

[CDT in Smart Medical Imaging at King's & Imperial @ImagingCDT]

We are now accepting applications from UK, EU/EEA/Swiss and international students for September 2020 entry. Details of exciting new projects can be found at https://t.co/0zjXXioY8K where you can also apply. Applications close on Monday 20 January 2020.

[JM Rothberg @JMRothberg]

World will never be the same. First #MRI scan on a boat. ⁦@HyperfineR⁩ & ⁦@ButterflyNetInc⁩ are on a mission to bring advanced medical imaging to everyone. Simple. Intuitive. Low cost. The machines learn! And connect to doctors! #unbox & #scan ⁦@gatesfoundation⁩

[Julia Schnabel @ja_schnabel]

@MiccaiStudents Join this vibrant community and contribute to some great activities at the next @MICCAI2020 !

[Julia Schnabel @ja_schnabel]

Thought I might just leave this here...

Register to vote - https://t.co/gc0SwIxrrZ https://t.co/D8OBzlKfFZ

[Alexander Terenin @avt_im]

I've been interested in applying differential geometry to study machine learning problems with physical interpretations for some time now. These notes are the cleanest presentation I've seen yet of the geometric and physical aspects of this setup. https://t.co/h5EXeAEOHW

[Qi Dou @QiDou_]

Our #NeurIPS2019 paper presents a new model-agnostic learning scheme for domain generalization! Simulating domain shift at training with regularizing the semantic space helps promote generic features!

Lastest work with @dccastr0 @KostasKamnitsas @GlockerBen in @BioMedIAICL

[Swoop® Portable MR Imaging® System–Hyperfine, Inc @Hyperfine]

Meanwhile, back at the lab...
@J_Schlemper, a member of #teamholykspace, has joined @HyperfineR Deep Learning to revolutionize medical imaging with the world’s first POC MRI. Congrats the team for smashing the top of #fastmri leaderboard! @khammernik @Chen_qin615 @Jinming19741073

[Daniel Coelho de Castro @dccastr0]

Excited to share that Morpho-MNIST will appear in @JmlrOrg! 🎉
This extended version discusses additional disentanglement metrics, predictive model eval (e.g. stratification), & morphometry on other datasets.
https://t.co/gVtVL5zLWn
https://t.co/WoBU3dg77i

Summary thread 👇