Anton Widengård joined the Soft Matter Lab on 9 February 2026.
Anton is a master student in Biomedical Engineering at Chalmers University of Technology.
Alongside with his classmate Max Haraldsson, Anton will be working on his Master’s Thesis at Soft Matter Lab, in collaboration with IFLAI, supervised by Jesús Pineda.
Anton’s and Max’s research focuses on evaluating and efficiently adapting pre-trained deep-learning vision models for cell segmentation and tracking.
Max Haraldsson joined the Soft Matter Lab on 9 February 2026.
Max is a master student in Biomedical Engineering at Chalmers University of Technology.
Together with his classmate Anton Widengård, he will be conducting his Master’s thesis at the Soft Matter Lab in collaboration with IFLAI, with Jesús Pineda as supervisor.
Max’s and Anton’s project is about evaluating and efficiently adapting pre-trained deep learning models for cell segmentation and tracking.
The defense took place in FB, Institutionen för fysik, Origovägen 6b, Göteborg, at 09:00.
Title: From Light to Data Using Deep Learning for Quantitative Microscopy
Abstract: Quantitative microscopy aims to measure physical properties of microscopic particles from optical images, but weak and complex signals often make this difficult. This thesis explores how computational methods, especially deep learning guided by physical understanding, can improve particle detection and characterization in microscopy.
The work introduces new approaches for locating and tracking particles, extends these ideas to three-dimensional and label-free imaging, and reviews practical analysis workflows. It further shows how combining complementary imaging techniques can enhance nanoparticle measurements and how deep learning can recover three-dimensional structural information from microscopy images.
Overall, this thesis strengthens the connection between optical measurements and quantitative particle information, expanding the potential of label-free microscopy for biological and nanoscale studies.
Thesis: https://gupea.ub.gu.se/handle/2077/90201
Supervisor: Daniel Midtvedt
Examiner: Raimund Feifel
Opponent: Arrate Munoz Barrutia
Committee: Per Augustsson, Jens Petersen, Rebecka Jörnsten
Alternate board member: Vitali Zhaunerchyk
Giovanni Volpe, Carolina Wählby, Lei Tian, Michael Hecht, Artur Yakimovich, Kristina Monakhova, Laura Waller, Ivo F. Sbalzarini, Christopher A. Metzler, Mingyang Xie, Kevin Zhang, Isaac C.D. Lenton, Halina Rubinsztein-Dunlop, Daniel Brunner, Bijie Bai, Aydogan Ozcan, Daniel Midtvedt, Hao Wang, Nataša Sladoje, Joakim Lindblad, Jason T. Smith, Marien Ochoa, Margarida Barroso, Xavier Intes, Tong Qiu, Li-Yu Yu, Sixian You, Yongtao Liu, Maxim A. Ziatdinov, Sergei V. Kalinin, Arlo Sheridan, Uri Manor, Elias Nehme, Ofri Goldenberg, Yoav Shechtman, Henrik K. Moberg, Christoph Langhammer, Barbora Špačková, Saga Helgadottir, Benjamin Midtvedt, Aykut Argun, Tobias Thalheim, Frank Cichos, Stefano Bo, Lars Hubatsch, Jesus Pineda, Carlo Manzo, Harshith Bachimanchi, Erik Selander, Antoni Homs-Corbera, Martin Fränzl, Kevin de Haan, Yair Rivenson, Zofia Korczak, Caroline Beck Adiels, Mite Mijalkov, Dániel Veréb, Yu-Wei Chang, Joana B. Pereira, Damian Matuszewski, Gustaf Kylberg, Ida-Maria Sintorn, Juan C. Caicedo, Beth A Cimini, Muyinatu A. Lediju Bell, Bruno M. Saraiva, Guillaume Jacquemet, Ricardo Henriques, Wei Ouyang, Trang Le, Estibaliz Gómez-de-Mariscal, Daniel Sage, Arrate Muñoz-Barrutia, Ebba Josefson Lindqvist, Johanna Bergman
Journal of Physics: Photonics 8, 012501 (2026)
arXiv: 2303.03793
doi: 10.1088/2515-7647/ae0fd1
Through digital imaging, microscopy has evolved from primarily being a means for visual observation of life at the micro- and nano-scale, to a quantitative tool with ever-increasing resolution and throughput. Artificial intelligence, deep neural networks, and machine learning (ML) are all niche terms describing computational methods that have gained a pivotal role in microscopy-based research over the past decade. This Roadmap encompasses key aspects of how ML is applied to microscopy image data, with the aim of gaining scientific knowledge by improved image quality, automated detection, segmentation, classification and tracking of objects, and efficient merging of information from multiple imaging modalities. We aim to give the reader an overview of the key developments and an understanding of possibilities and limitations of ML for microscopy. It will be of interest to a wide cross-disciplinary audience in the physical sciences and life sciences.
The defense will take place in SB-H7 lecture hall, SB-Building, Institutionen för fysik, Johanneberg Campus, Göteborg, at 13:00.
Title: A Unified Software-Generating Framework for Biological Data Analysis
Abstract: Biological data analysis relies heavily on software, but as projects grow it becomes hard to keep code, interfaces, and tests aligned, and to reuse methods without rewriting them. This thesis presents Genesis, which generates runnable modules, GUIs, and unit tests from a single human-readable .gen.m description of each analysis component. By maintaining a central library of these descriptions, analyses can be recombined for new questions while staying consistent. Four studies across neuroimaging, light-sheet microscopy, and plant Raman spectroscopy show the framework is reusable and extensible across domains.
Thesis: http://hdl.handle.net/2077/90289
Supervisor: Giovanni Volpe (Main), Caroline Beck Adiels
Examiner: Raimund Feifel
Opponent: Arvind Kumar
Committee: Wojciech Chachólski, Rita Almeida, Paolo Vinai
Alternate board member: Mohsen Mirkhalaf
Eduard Andrei Duta Costache started his PhD at the Physics Department of the University of Gothenburg on the 19th of January 2026.
Eduard has a double Master’s degree in Artificial Intelligence from the University of Alicante (Spain) and in Machine Learning & Data Mining from Jean Monnet University (France).
During the course of his PhD, as part of the GREENS MSCA Doctoral Network, he will focus on developing AI frameworks to model and optimize the lifecycle of micro-robotic platforms.
Steven brings years of expertise from the laboratory of Professor Carlos Bustamante, where they pioneered the use and development of optical tweezers. As the main developer of the widely used ‘minitweezers’ instrument, which today is used by dozens of research groups, he has helped shape the field of single-molecule biophysics on a global scale. We look forward to his visit, during which he will work on refining cutting-edge single-molecule measurement techniques.
The nailing took place in Universitetsbyggnaden i Vasaparken, Universitetsplatsen 1, Göteborg, at 13:30.
In Swedish academia, “nailing” (spikning) is the formal public announcement and publication of a doctoral thesis. It happens weeks before the defence so that the public has time to read the thesis in advance and prepare questions for the defence. In addition to the physical nailing, the thesis is also published electronically (e-spikning) via GUPEA.
Yu-Wei will defend his thesis on 23 January at 13:00 in SB-H7 lecture hall, SB-Building, Institutionen för fysik, Johanneberg Campus, Göteborg.
Thesis (GUPEA handle): http://hdl.handle.net/2077/90289
Fredrik will defend his thesis on 29 January at 09:00 in FB-salen, Institutionen för fysik, Origovägen 6, Göteborg.
Jesus Pineda defended his PhD thesis on November 11th, 2025. Congrats!
The defense took place in SB-H7 lecture hall, Institutionen för fysik, Johanneberg Campus, Göteborg, at 9:00.
Title: Inductive Biases for Efficient Deep Learning in Microscopy
Abstract: Deep learning has become an indispensable tool for the analysis of microscopy data, yet its integration into routine research remains uneven. Several factors contribute to this gap, including the limited availability of well-annotated datasets and the high computational demands of modern architectures. Microscopy introduces further challenges, as it spans diverse modalities and scales, from proteins to tissues, producing heterogeneous data that defy standardization. Generating reliable annotations also requires expertise and time, while unequal access to high-performance computing further widens the divide between well-resourced institutions and smaller laboratories.
This dissertation argues that the prevailing paradigm of scaling models with ever-larger datasets and computational resources yields diminishing returns for microscopy. Instead, it explores the role of inductive biases as a foundation for building models that are more data-efficient, computationally accessible, and scientifically meaningful. Inductive biases are structural assumptions embedded in model design that guide learning toward patterns aligned with the underlying problem. The first part of this work examines their central role in the advancement of modern deep learning and the diverse ways they shape model behavior.
This potential is demonstrated through three case studies. First, MAGIK employs graph neural networks to analyze biological dynamics in time-lapse microscopy, uncovering local and global properties with high precision, even when trained on limited data. Next, MIRO leverages recurrent graph neural networks to process single-molecule localization datasets, improving the efficiency and reliability of clustering for variable biological structures and scales while retaining strong generalization with minimal supervision. Finally, GAUDI introduces a representation learning framework for characterizing biological systems, providing a physically meaningful representation space for interpretable and transferable analysis.
The findings presented here demonstrate that the integration of inductive biases provides a cohesive strategy to extend the reach of deep learning in the life sciences, enhancing accessibility and ensuring scientific utility under resource constraints.
Thesis: https://gupea.ub.gu.se/items/672c7946-51d6-4773-ad8c-35a3eed41499
Supervisor: Giovanni Volpe
Co-Supervisor: Carlo Manzo
Examiner: Raimund Feifel
Opponent: Anna Kreshuk
Committee: Juliette Griffié, Daniel sage, Daniel Persson
Alternate board member: Jonas Enger
