📣 Comparison of performance of backbones (AlexNet to Found. Model) and classification paradigms (kNN to end2end) on 12 medical datasets.

-> comprehensive baseline on MedMNIST+

-> informs baseline choise and the influence of input resolution

Paper: arxiv.org/abs/2404.15786

New work to be presented at midl.conference. Anna Wundram and Paul Fischer showed that considering the full probability distribution of uncertain pipeline steps can substantially improve downstream performance.

📜 Paper: openreview.net/forum?id=kcbAZ…
🐍 Code: github.com/annawundram/gl…

Yesterday, I started as Assistant Professor for data science in diabetes care at the University of Bern. Thank you for the warm welcome, I am very much looking forward to working with my new colleagues at Insel Gruppe Universität Bern and DCB!

It is our pleasure to announce that Adrian Dalca
is now an Executive Editor. Prof. Dalca has been an integral member of the MELBA team since our inception and has served as an associate editor and launching editor. We are grateful for the experience, expertise, and knowledge he

We are excited to present two papers at #NeurIPS2023 ! 'Expert load matters: operating networks at high accuracy and low manual effort' by Sara Sangalli , Ertunc Erdil & Ender Konukoglu, and 'Canonical normalizing flows for manifold learning' by Kyriakos Flouris & Ender Konukoglu.

We are now only missing 7 responses. If you haven't already, please take 5-10 minutes to do our survey!

In our latest In Conversation, we chat to Dr Emma Robinson about the importance of supportive #mentorship in #academia , particularly for those from disadvantaged backgrounds.
Emma R - @metricsemma.bsky.social kcl.ac.uk/news/in-conver…

Don’t miss the chance to apply to the MICCAI Student Board!! Happy to share my experience of being on the board with anyone who is interested 🤓

🧠What is the functional role of cortico-thalamic feedback in visual processing?
✨Using data efficient variant of generalized linear models (Spline-GLM) we model activity in dLGN

biorxiv.org/content/10.110…
w/ Ann Kotkat Yannik Bauer Ziwei Huang Philipp Berens Laura Busse

1/7

Really nice exercise on explainable AI. Looking forward to the results.

Take a few minutes to participate in our survey on saliency maps for detecting spurious features in chest X-ray!

#interpretableML #XAI

Excellent keynote!

DART taking place in Room 18, with wonderful attendance. Join us for the Keynote talks by Tal Arbel (McGill), Alison Q O'Neil (Canon Medical) and oral presentations of our accepted papers.

If you are at #MICCAI2023 today and interested in interpretable ML for medicine, please visit our poster on 'Right for the Wrong Reason: Can Interpretable ML Techniques Detect Spurious Correlations?' at T-03-101 from 9:30-11:00.
MICCAI MICCAI Students

🗓️ The schedule for the DART Workshop at MICCAI 2023 is online!

Have a look at the great line up we have prepared with keynotes and orals, as well as a poster session!

More details: sites.google.com/view/dart2023/…

Happy to have been among the honorable mentions for the best lighting talk at last week's grad school boot camp Intelligent Systems ! 🌟

It was fun to present my PhD project in < 1 min! ⏰

(1/3) Can interpretable ML methods identify shortcut learning? Looking forward to presenting our latest work on 'right for the wrong reason' in the coming #MICCAI2023 . Joint work with Lisa Koch, Christian Baumgartner

Pre-print: arxiv.org/abs/2307.12344
code: github.com/ss-sun/right-f…

The decisions for DART workshop @MICCAI 2023 are out. Congratulations to the authors of accepted papers, we look forward to seeing your work at the DART workshop at #MICCAI2023 !

Massive thanks to all the reviewers and emergency reviewers for their hard work and commitment!