Sign in

Bernhard Kainz

@bernhardkainz.bsky.social
189 followers 16 following 17 posts

2xDad, DL practitioner, real-time computing fanatic, Prof. at AIBE, FAU Erlangen, Reader at DoC, Imperial College London, advisor at ThinkSono, former chef

PostsRepliesMedia
Reposted by Bernhard Kainz
Machine Learning for Biomedical Imaging @melbajournal.bsky.social · 03/06/2025
📢 MELBA is growing its editorial team! We’re excited to welcome three new Associate Editors who bring outstanding expertise in machine learning and biomedical imaging. Welcome to: 🔹 @bernhardkainz.bsky.social 🔹 @mariavak.bsky.social 🔹 Lisa Koch We’re thrilled to have you on board! 🚀
052
Bernhard Kainz @bernhardkainz.bsky.social · 10/12/2024
Congratulations to Qiang Ma for passing his PhD viva yesterday evening! And thanks to Bruce Fischl and Abhijeet Ghosh for acting as examiners!
091
Reposted by Bernhard Kainz
Daniel Rueckert @danielrueckert.bsky.social · 05/12/2024
Congratulations to Liu Li for passing her PhD viva today! And thanks to Björn Menze and Björn Schuller for acting as examiners!
0121
Reposted by Bernhard Kainz
Daniel Rueckert @danielrueckert.bsky.social · 06/12/2024
Great BioMedIA Christmas party at Imperial!
0162
Reposted by Bernhard Kainz
Athanasios Vlontzos, PhD @vlontzos.bsky.social · 25/11/2024
Thank you to my co-authors @bernhardkainz.bsky.social and C. Müller
011
Reposted by Bernhard Kainz
Emma Robinson @OHBM 2026 @metricsemma.bsky.social · 22/11/2024
Looking for a London PhD? The London interdisciplinary social science DTP takes student designed PhD projects in a very broad range of topics from health innovation to sustainability and public policy liss-dtp.ac.uk
liss-dtp.ac.uk
Home - LISS DTP
042
Bernhard Kainz @bernhardkainz.bsky.social · 06/10/2023
yes, exactly. Ensembles of methods usually achieve the best performance for challenging problems like this.
010
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
I mean the hidden biases can be learned, not only the known parameters. One can check with a parametrised model if a new input conforms with all the biases in a specific downstream model. This does not necessary need to be ethnicity but could be a combination of latent factors.
210
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
finding a method that allows us to estimate how suitable a model is for a given patient... This solves only one side of the problem but bias is not necessarily 'bad'. After all, 'patient specific' means 100% bias towards one specific patient. Triage and patient stratification will be key in future.
120
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
The 4th International Workshop of Advances in Simplifying Medical UltraSound (ASMUS) - a workshop held in conjunction with #MICCAI23 is only a few days away. Join us on Sunday 8️⃣ Oct from 8️⃣ AM in Meeting Room 8️⃣ Vancouver Convention Center East Building Level 1. miccai-ultrasound.github.io#/asmus23
ASMUS workshop logo
102
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Multiple Instance Learning for Axillary Lymph Node Metastasis (DEMI) x.com/Hanna_Paulal...
000
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Lesion-Level Data Augmentation for Segmentation (DALI) arxiv.org/abs/2308.09026
100
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Probabilistic Self-Supervised Image Anomaly Localisation (UNSURE) x.com/Hanna_Paulal...
000
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Synthetic Data Enrichment for Robust Image Segmentation in Kidney Transplant Pathology (DART) arxiv.org/abs/2304.09534
200
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Neonatal Cortical Surface Reconstruction arxiv.org/abs/2307.11870
100
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Unsupervised anomaly localisation with synthetic tasks arxiv.org/abs/2307.00899
100
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Respiratory Motion Correction and Super-Resolution for 3D Abdominal MRI link.springer.com/chapter/10.1...
100
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Robust Segmentation via Topology Violation Detection link.springer.com/chapter/10.1...
100
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
- Cascaded Diffusion Models to generate echocardiogram videos x.com/reynaud_h/st...
110
Bernhard Kainz @bernhardkainz.bsky.social · 05/10/2023
Getting ready for #MICCAI23! group -> @CityofVancouver ✈️ 📢 This year we present work on 🧵:
some of our MICCAI posters printed
110
Bernhard Kainz @bernhardkainz.bsky.social · 01/10/2023
#HiSciSky! My team and I are immersed in the world of intelligent algorithms for #MedicalImaging. We are on a mission to redefine real-time diagnostics especially for early detection and health screening. We explore normative machine learning to help defining what ought to be physiologically normal.
1122
Bernhard Kainz @bernhardkainz.bsky.social · 01/10/2023
Hello BlueSky. Yet another platform or better?
040