Mahshad Lotfinia, M.Sc.

• Since 12/2025, Ph.D. Student in Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
• Since 07/2024, Cloud Network Engineer, GOSP, Generali Operations Service Platform, Aachen, Germany
• 09/2023 to 06/2024, Research Internship, Lab for Artificial Intelligence in Medicine, University Hospital RWTH Aachen, Aachen, Germany
• 08/2022 to 06/2024, Research Employee, Mechanical Engineering, RWTH Aachen University, Aachen, Germany
• 08/2022 to 06/2024, Research Assistant, RWTH Aachen University, Aachen, Germany
• 02/2022 to 07/2022, Working Student, Data Science, WBA Aachener Werkzeugbau Akademie GmbH, Aachen, Germany
• 09/2021 to 01/2022, Working Student, Deep Learning, Virtonomy GmbH, Munich, Germany
• 04/2021 to 07/2022, Visiting Student in Computational Science, University of Regensburg, Regensburg, Germany
• 09/2018 to 07/2020, M.Sc. in Materials Science and Engineering, Sharif University of Technology, Tehran, Iran
• 09/2014 to 07/2018, B.Sc. in Materials Science and Engineering, Bu-Ali Sina University, Hamadan, Iran

2026

• M. Lotfinia, D. Truhn, A. Maier, and S. Tayebi Arasteh. “FRAME: separating sampling variation from performance disparities in medical image analysis.” arXiv preprint, 2026.[URL]
• M. Farajiamiri, J. Sopa, S. Afza, L. Adams, F. Barajas Ordonez, T. Nguyen, M. Lotfinia, S. Wind, K. Bressem, S. Nebelung, D. Truhn, and S. Tayebi Arasteh. “Retrieval-augmented reasoning reshapes collective reliability in text-based radiology question answering.” Patterns, Cell Press, 2026.[URL]
• S. Tayebi Arasteh, S. Ziegelmayer, M. Lotfinia, L. Adams, S. Nebelung, J. Nikolas Kather, and D. Truhn. “Medical foundation models converge less under label supervision.” arXiv preprint, 2026.[URL]
• M. Lotfinia, S. Ziegelmayer, L. Adams, T. Nguyen, D. Truhn, A. Maier, and S. Tayebi Arasteh. “Vision-language models for chest radiography do not always need the image.” arXiv preprint, 2026.[URL]
• S. Wind, T. Nguyen, J. Sopa, M. Lotfinia, S. Bickelhaup, M. Uder, H. Köstler, G. Wellein, S. Nebelung, D. Truhn, A. Maier, and S. Tayebi Arasteh. “Safety and accuracy follow different scaling laws in clinical large language models.” arXiv preprint, 2026.[URL]
• S. Tayebi Arasteh, M. Joodaki, M. Lotfinia, S. Nebelung, and D. Truhn. “Case-grounded evidence verification: a framework for constructing evidence-sensitive supervision.” arXiv preprint, 2026.[URL]
• S. Tayebi Arasteh, M. Farajiamiri, M. Lotfinia, B. Hinrichs-Puladi, J. Bienzeisler, M. Alhaskir, M. Rusu, C. Kuhl, S. Nebelung, and D. Truhn. “The pretraining domain outweighs the training objective in setting the privacy-utility trade-off of differentially private medical image analysis.” arXiv preprint, 2026.[URL]
• M. Mohammadi, M. Vejdanihemmat, M. Lotfinia, M. Rusu, D. Truhn, A. Maier, and S. Tayebi Arasteh. “Differential privacy for medical deep learning: methods, tradeoffs, and deployment implications.” npj Digital Medicine, 2026.[URL]

2025

• S. Wind, J. Sopa, D. Truhn, M. Lotfinia, T. Nguyen, K. Bressem, L. Adams, M. Rusu, H. Köstler, G. Wellein, A. Maier, and S. Tayebi Arasteh. “Multi-step retrieval and reasoning improves radiology question answering with large language models.” npj Digital Medicine, 2025.[URL]
• S. Tayebi Arasteh, M. Lotfinia, P. Andrea Perez-Toro, T. Arias-Vergara, M. Ranji, J. Rafael Orozco-Arroyave, M. Schuster, A. Maier, and S. Hee Yang. “Differential privacy enables fair and accurate AI-based analysis of speech disorders while protecting patient data.” npj Artificial Intelligence, 2025.[URL]
• M. Lotfinia, A. Tayebiarasteh, S. Samiei, M. Joodaki, and S. Tayebi Arasteh. “Boosting multi-demographic federated learning for chest radiograph analysis using general-purpose self-supervised representations.” European Journal of Radiology Artificial Intelligence, 2025.[URL]
• S. Tayebi Arasteh, M. Lotfinia, K. Bressem, R. Siepmann, L. Adams, D. Ferber, C. Kuhl, J. Nikolas Kather, S. Nebelung, and D. Truhn. “RadioRAG: online retrieval-augmented generation for radiology question answering.” Radiology: Artificial Intelligence, 2025.[URL]
• S. Tayebi Arasteh, S. Afza, T. Nguyen, L. Buess, M. Parvin, T. Arias-Vergara, P. Andrea Perez-Toro, H. Ching Hung, M. Lotfinia, T. Gorges, E. Noeth, M. Schuster, S. Hee Yang, and A. Maier. “Perceptual implications of automatic anonymization in pathological speech.” arXiv preprint, 2025.[URL]

2024

• S. Tayebi Arasteh, R. Siepmann, M. Huppertz, M. Lotfinia, B. Puladi, C. Kuhl, D. Truhn, and S. Nebelung. “The treasure trove hidden in plain sight: the utility of GPT-4 in chest radiograph evaluation.” Radiology, 2024.[URL]
• S. Tayebi Arasteh, T. Han, M. Lotfinia, C. Kuhl, J. Nikolas Kather, D. Truhn, and S. Nebelung. “Large language models streamline automated machine learning for clinical studies.” Nature Communications, 2024.[URL]
• S. Tayebi Arasteh, M. Lotfinia, T. Nolte, M. Sähn, P. Isfort, C. Kuhl, S. Nebelung, G. Kaissis, and D. Truhn. “Securing collaborative medical AI by using differential privacy: domain transfer for classification of chest radiographs.” Radiology: Artificial Intelligence, 2024.[URL]

2023

• S. Tayebi Arasteh, J. Romanowicz, D. F. Pace, P. Golland, A. J. Powell, A. K. Maier, D. Truhn, T. Brosch, J. Weese, M. Lotfinia, R. J. van der Geest, and M. H. Moghari. “Automated segmentation of 3D cine cardiovascular magnetic resonance imaging.” Frontiers in Cardiovascular Medicine, 2023.[URL]

2021

• S. Tayebi Arasteh, M. Monajem, V. Christlein, P. Heinrich, A. Nicolaou, H. Naderi Boldaji, M. Lotfinia, and S. Evert. “How will your tweet be received? Predicting the sentiment polarity of tweet replies.” IEEE International Conference on Semantic Computing (ICSC) 2021.[URL]

Editorial boards

• European Radiology Experimental (European Society of Radiology). Editorial Board Member, Section Computed Tomography, since 02/2025

Journal review

Reviewer for Scientific Reports, Communications Medicine, Scientific Data, European Radiology Experimental, Journal of Medical Internet Research, JMIR AI, BMC Medical Informatics and Decision Making, International Journal of Computer Assisted Radiology and Surgery, PeerJ Computer Science, and others.