ThomasBudd

@FabianIsensee1 Many thanks!

@ThomasBuddML We are currently not planning on preparing a manuscript for this release. It's mostly 1) a switch to nnU-Net, 2) the already released DA5 nnU-Net trainer and 3) some additional augmentations that simulate incomplete scans

HD-BET just became a lot more powerful and robust thanks to nnU-Net (+some data augmentation tweaks) 🥳 Install with `pip install -U hd-bet` or via our GitHub repository: github.com/MIC-DKFZ/hd-bet Happy brain extraction 💀!

@david_rolnick @A_Aspuru_Guzik @sarameghanbeery @BDilkina @priyald17 @MarzyehGhassemi @hannah_kerner @rolf_comma_e @MilindTambe_AI I chose to leave academia because almost all research there is method-driven. Let's be honest, almost all of this work goes to the drawer and does not have an impact for others. In industry we do application-driven ML only, if the model wouldn't solve a problem no on would buy it

What drives innovation in machine learning? In a new paper, we argue that application-driven work is systemically under-valued in the machine learning community, but that it's essential for both innovation and impact. arxiv.org/abs/2403.17381 1/3

@Karl_Lauterbach Bei uns liegt aktuell leider ein wichtiges EU-Projekt auf Eis, da D aufgrund der Haushaltssperre die versprochenen Forschungsgelder noch nicht ausgezahlt hat. Die Taltente sind da, mit Unterstützung kann in D eine blühende Landschaft an KMUs für KI in der Medizin entstehen. (3/3)

@Karl_Lauterbach Mir hat die Sendung gefallen, ich hätte mich beim Thema künstliche Intelligenz darüber gefreut, wenn die medizinische KI-Industrie und Industrieforschungsgelder in D auch erwähnt worden wäre. Ich bin seit 6 Jahren Ingenieur für KI und arbeite in einem KMU zu dem Thema. (1/3)

Hier die Sendung mit Markus #Lanz von gestern. Es ging um Cannabis Legalisierung, Krankenhausreform und künstliche Intelligenz. Kommentare willkommen zdf.de/gesellschaft/m…

@Karl_Lauterbach Diese Branche ist in D nicht sehr stark. Neben dem angesprochenen Datenschutz und der Grundlagenforschung sind insbesondere Forschungsgelder für die Industrie extrem wichtig für uns, damit wir neue und riskante Themen bearbeiten und wachsen können. (2/3)

@RGImageAnalysis @lorena_e_s @Cathal_McCague @imaging_cancer @caromitreka @EvisSala @RamonaWoitek @mireiacrispin It was great to be a part of this project! In my opinion, the difficulty of integration is one of the key reasons why we don't see more AI being adopted in clinical practice.

As important as it is to develop #ArtificialIntelligence tools for cancer research, it is also important to integrate them into the clinical setting. In our latest paper we describe the technical details of our pipeline for #ovariancancer

We use ensembles of highly sensitive to highly precise models and show that their combination can i) be used to learn from unlabeled data, ii) used for active learning procedures and iii) support human-machine interaction.

Looking for uncertainty quantification in medical image segmentation? Check out our new paper published in Computer in Biology and Medicine: sciencedirect.com/science/articl… @lorena_e_s @mireiacrispin @RamonaWoitek @Cathal_McCague James Brenton @EvisSala Leonardo Rundo

@mic_dkfz @annika_reinke1 Looking forward to participate!

The KiTS challenge is back with an expanded training set of 489 cases, brand new test set of 110 cases, and the addition of a new contrast phase! Help us advance the field & improve patient outcomes. Download the data today & compete for the $5,000 prize! kits-challenge.org/kits23

@Cathal_McCague @SyafiqRamlee_ @MarikaReinius @DrIanSelby @DGHulse @PiyatissaPubudu @burav_Rad @mireiacrispin @EvisSala @RamonaWoitek Congrats Cathal! Well deserved!

Delighted to be awarded the editors choice for our article. Thanks and congratulations for your great efforts @SyafiqRamlee_ @MarikaReinius @DrIanSelby @DGHulse @PiyatissaPubudu @burav_Rad @mireiacrispin @EvisSala and @RamonaWoitek!