Thomas Jochmann


Ph.D. student




Thomas has attained his graduate degree in physics from the Friedrich Schiller University Jena. Currently, he is a PhD student in biomedical engineering and is actively exploring the application of machine learning to the analysis of biomedical data. He is also dedicated to the development of new machine learning methods for extracting biological underpinnings of the MRI signal. His particular area of interest lies in MRI phase contrast mechanisms, and he has successfully developed the DEEPOLE QUASAR method, which quantifies non-dipolar frequency shift contributions in the brain.

In his free time, Thomas enjoys engaging in activities that challenge his technical abilities; he loves cooking, exploring the latest technology, tinkering with electronics, and repairing things. He is also a devoted father to his three children, and takes great pleasure in spending time with them. He is passionate about his hobbies and takes pride in mastering them.

Novel MRI Technique Reveals Subtypes of Paramagnetic Rim Lesions and Predicts 5-year Rim Disappearance


J.A. Reeves*, T. Jochmann*, M. Mohebbi, D. Jakimovski, S. Hametner, F. Salman, N. Bergsland, B. Weinstock-Guttman, M.G. Dwyer, R. Zivadinov, F Schweser


Americas Committee for Treatment and Research in Multiple Sclerosis Forum , San Diego, CA, 2023


Self-calibrating χ-separation using single-subject (N=1) physics-constrained deep learning.


Benslimane I, Grabner G, Hametner S, Jochmann T, Zivadinov R, Schweser F


15th Annual Buffalo Neuroscience Research Day, Buffalo, NY, 2022


Beyond qMRI: Biological tissue properties forom single-subject unsupervised deep learning with theoretical signal constraints


Benslimane I, Grabner G, Hametner S, Jochmann T, Zivadinov R, Schweser F


Proc Intl Soc Mag Reson Med, 2022, p. 370


Improved self-calibrated signal model for χ-separation using single-subject (N=1) physics-constrained deep learning with histological validation.


Benslimane I, Grabner G, Hametner S, Jochmann T, Zivadinov R, Schweser F


Artificial Intelligence in Clinical Imaging (AICI) Forum, Graz, Austria, 2022


Self-calibration, histological validation, and an improved signal model for χ-separation using single-subject (N=1) physics-constrained deep learning.


Benslimane I, Grabner G, Hametner S, Jochmann T, Zivadinov R, Schweser F


Joint Workshop on MR phase, magnetic susceptibility and electrical properties mapping, Lucca, Italy, 2022


Nonsusceptibility frequency shifts in the human brain and their impact on quantitative suscep-tibility mapping


Jochmann T, Jakimovski D, Hametner S, Zivadinov R, Haueisen J, Schweser F


Workshop on MR phase, magnetic susceptibility and electrical properties mapping, Lucca, Italy, 2022


Embedding medium alters local phase contrast in postmortem MRI of the human brain


Schweser F, Hametner S, Jochmann T, Riedl C, Bachrata B, Laux M, Grabner G


Joint Workshop on MR phase, magnetic susceptibility and electrical properties mapping, Lucca, Italy, 2022


Unsupervised physics-informed deep learning (N=1) for solving inverse qMRI problems–Relaxometry and field mapping from multi-echo data.


Benslimane I, Jochmann T, Zivadinov R, Schweser F


Proc Intl Soc Mag Reson Med, 2021, p. 330


Quantitative mapping of susceptibility and non-susceptibility frequency with DEEPOLE QUASAR


Jochmann T, Jakimovski D, Küchler N, Zivadinov R, Haueisen J, Schweser F


Proc Intl Soc Mag Reson Med, 2021, p. 789


Quantitative mapping of susceptibility and non-susceptibility frequency from gradient-echo phase images


Jochmann T, Jakimovski D, Küchler N, Zivadinov R, Haueisen J, Schweser F


12th Ultrahigh Field Magnetic Resonance Symposium , 2021, p. 13


Mapping Magnetic Susceptibility and Non-Susceptibility Sources from MRI Frequency Shift with Physics-informed Deep Learning.


Jochmann T, Schweser F, Küchler N, Jakimovski D, Zivadinov R, Haueisen J


16th International Workshop on Optimization and Inverse Problems in Electromagnetism, 2021


U3-Net for Deep Vector QSM – Solving the Susceptibility Tensor Phase Model in Single-Orientation MRI


Baader EF, Jochmann T, Haueisen J, Zivadinov R, Schweser F


Proc Intl Soc Mag Reson Med, 2020, p. 3202


How to train a Deep Convolutional Neural Network for Quantitative Susceptibility Mapping (QSM)


Jochmann T, Haueisen J, Schweser F


Proc Intl Soc Mag Reson Med, 2020, p. 3195


SOJU-Net—Denoising MR phase images with physics-informed deep learning using artificial Rician noise augmentation


Jochmann T, Kuechler N, Haueisen J, Schweser F


Proc Intl Soc Mag Reson Med, 2020, p. 3196


A Fourier transformation based convolutional neural network layer for physics-informed deep learning of magnetic dipole inversion


Küchler N, Haueisen J, Schweser F, Jochmann T


German Society for Biomedical Engineering , 2020, p1570638759


Physics-aware augmentation, artificial noise, and synthetic samples to train a convolutional neural network for QSM


Jochmann T, Haueisen J, Schweser F


5th International Workshop on MRI Phase Contrast and QSM, Seoul, Korea, 2019


U2-Net for DEEPOLE QUASAR–A Physics-Informed Deep Convolutional Neural Network that Disentangles MRI Phase Contrast Mechanisms.


Jochmann T, Haueisen J, Zivadinov R, Schweser F


Proc Intl Soc Mag Reson Med, Montreal, QC, Canada, 2019, p. 320


DEEPOLE QUASAR–A deep learning-based approach for quantitative susceptibility and non-susceptibility phase mapping.


Jochmann T, Haueisen J, Zivadinov R, Schweser F


5th International Workshop on MRI Phase Contrast and QSM, Seoul, Korea, 2019


Overcoming the Rician Noise Bias of T2* Relaxometry with an Artificial Neural Network (ANN).


Schweser F, Jochmann T, Zivadinov R


Proc Intl Soc Mag Reson Med, Proc Intl Soc Mag Reson Med, Montreal, QC, Canada, 2019, p. 4854