PhD student @ TU Ilmenau 🇩🇪 / Visiting researcher @ U Buffalo 🇺🇸.
Biomedical data science in quantitative MRI and M/EEG.
Physicist by heart, and training.
Join us at the EMTP2026 Workshop on September 20–23, 2026, in Annapolis, Maryland, USA!
Abstract submission is now open!
Regular abstract deadline: June 15, 2026
More information:
https://t.co/ucjmUlvhp6
#MRI#QSM#EPT#EMTP#Neuroimaging#QuantitativeMRI
We hope this contribution leads to more reliable QSM for studying neurodegeneration, demyelination, and iron homeostasis.
Shoutout to the @ISMRM community for the reaffirming feedback and the awards that kept us going the extra miles from prototype to clinical data robustness.
Now published in Magnetic Resonance in Medicine:
Quantitative susceptibility mapping in magnetically inhomogeneous tissues
https://t.co/UJpHcvebHR
@schweserf@Jens_Haueisen@TU_Ilmenau
Conventional QSM struggles in white matter and CSF due to oversimplified assumptions.
1/3
We introduce DEEPOLE, a deep learning method that integrates nondipolar Larmor frequency shifts into QSM.
DEEPOLE produces more anatomically consistent and quantitatively accurate susceptibility maps, especially in regions affected by microstructure effects. 2/3
Implications: neural networks for detecting pathology from raw EEG are easily sex-biased. Recommendations: thoroughly remove artifacts, balance your dataset, and test for biases when using deep learning for EEG! Shout-out to @Jens_Haueisen@mshamalai@dr_sheraz and @mne_news 3/3
Paper just out at HBM: https://t.co/lUBqA1Hi8N. In short: deep learning for EEG analysis is intriguing as it can utilize patterns that us humans are not even aware of. It can uncover new features and evaluate much more complex patterns in these multichannel zig-zag curves. … 1/3
However, deep learning is shameless in exploiting illegitimate features. When we replicated sex detection from EEG (by @mjamvanputten et al., https://t.co/hH7MB83674), our CNN first focused on sex differences in heart artifacts, rather than actual brain activity! 2/3
Thanks to the reviewer for recognizing our effort for #openscience: "the authors are to be commended on the thoroughness and documented reproducibility of the described science." 74p. SI, code+doc+data on Zenodo. Such kind words encourage to do that again! https://t.co/5y7iETwXyE
Immensely happy and grateful for being awarded an ISMRM Research Exchange Grant of US$5,000. What a huge help for the upcoming year in the US with my family of five! Heartfelt thanks to the @ISMRM, and my home and host mentors, @Jens_Haueisen and @schweserf.
Renewed call for abstracts for Biomag2022 (28/8-1/9) in Birmingham! Deadline 18/3/2022. The conference will be hybrid (in-person+virtual) but we hope for strong in-person attendance. Submit abstract now - decide later! Help us spread the word https://t.co/WpNsz9VJxJ
I wonder if the prediction of the limb position in pigs with a Neuralink implant only works because of muscle artifacts, i.e. electrical potentials from the limbs' muscles propagating to the head. How has that been filtered? #askneuralink
For #bestvirtualISMRMever the Quantitative MR Study Group held a Public Engagement Competition. The winners are @AnitaKarsa, @MarcoAZampini, and @thomas_jochman! Join us on Aug 18 at 13 UTC for the SG mtg to see their work (pre-reg required).
@ISMRM#ISMRM20
@ISMRM, what is the recommended size for digital poster PDFs at the 2020 virtual conference? The guidelines mention a 16:9 ratio, but what is a good width and font size for your conference system? Will users open the PDF in their native PDF viewer or see it embedded on the web?
Best #ClinicalPractice 👨⚕️🧠 - Should #Nusinersen be used to treat adults with #SpinalMuscularAtrophy ? This preliminary study concludes the treatment is feasible and patients seek continuation
https://t.co/1uCmjjINhH