Author: Mirja Granfors

Poster presentation by M. Granfors at Optics and Photonics in Sweden, Norrköping, 7 October, 2026

GAUDI leverages a hierarchical graph-convolutional variational autoencoder architecture, where an encoder progressively compresses the graph into a low-dimensional latent space, and a decoder reconstructs the graph from the latent embedding. (Image by M. Granfors and J. Pineda.)
Unsupervised Graph Deep Learning Uncovers Protein Organization in Super-Resolution Microscopy
Mirja Granfors, Jesús Pineda, Blanca Zufiria Gerbolés, Joana B. Pereira, Carlo Manzo, and Giovanni Volpe
Date: 7 October 2026
Time: 11:45 – 13:00 and 17:20 – 19:00
Place: Biblioteket, Arbetets museum, Norrköping

Modern optical techniques such as super-resolution microscopy generate rich, spatially resolved data that can be naturally represented as graphs. However, the structural variability of these graphs poses significant challenges for their analysis and classification, particularly when similar underlying processes can produce substantially different graph realizations.

To address this challenge, we introduce Graph Autoencoder Uncovering Descriptive Information (GAUDI), an unsupervised graph deep learning framework designed to capture both local details and global structure. GAUDI employs a hierarchical graph-convolutional variational autoencoder architecture with pooling and upsampling layers connected through skip connections, preserving essential connectivity information throughout the encoding–decoding process.

Despite substantial variability between graph realizations generated from identical or highly similar underlying parameters, GAUDI consistently maps them to nearby regions of a structured, continuous latent space. This enables the identification of invariant process-level features while separating them from stochastic variability.

We demonstrate this capability using super-resolution microscopy, where GAUDI characterizes protein assemblies and organizes structurally variable graphs according to their underlying state. The framework is further demonstrated on small-world network models, collective motion in the Vicsek model, and age-related changes in brain connectivity. These results highlight GAUDI as a broadly applicable approach for extracting meaningful information from complex graph-structured data across microscopy and other fields.

Poster presentation by M. Granfors at Optics and Photonics in Sweden, Norrköping, 7 October, 2026

DeepTrack2 Logo. (Image by J. Pineda.)
DeepTrack2: Physics-Based Microscopy Simulations for Deep Learning
Mirja Granfors, Alex Lech, Jiacheng Huang, Eduard Duta, Guillem Guigo i Corominas, Harshith Bachimanchi, Benjamin Midtvedt, Jesús Pineda, Carlo Manzo, and Giovanni Volpe
Date: 7 October 2026
Time: 11:45 – 13:00 and 17:20 – 19:00
Place: Biblioteket, Arbetets museum, Norrköping

Deep learning has become an increasingly powerful tool for microscopy image analysis, but training robust models often requires large and diverse datasets that are difficult and time-consuming to acquire experimentally. Physics-based simulations provide a way to generate synthetic microscopy data while retaining control over the optical and physical properties of the imaging process.

DeepTrack2 is a flexible and scalable Python library designed to generate physics-based synthetic microscopy datasets for training deep learning models. It supports a wide range of imaging modalities, including brightfield, fluorescence, darkfield, and holography, and allows users to customize optical systems, incorporate optical aberrations and noise, simulate diverse objects and imaging conditions, and apply image augmentations.

Its modular architecture also enables simulations to be executed using a PyTorch backend, providing GPU acceleration and allowing backpropagation through the optical system. This makes it possible to optimize the optical setup and the deep learning model jointly. The DeepTrack2 repository further provides tutorials covering the complete workflow, from physics-based simulation to deep-learning-based analysis using its companion library, deeplay.

Together, these capabilities provide a flexible framework for developing and benchmarking deep learning methods for microscopy while reducing the need for large experimentally acquired training datasets.

Global graph features unveiled by unsupervised deep learning published in Machine Learning: Science and Technology

GAUDI leverages a hierarchical graph-convolutional variational autoencoder architecture, where an encoder progressively compresses the graph into a low-dimensional latent space, and a decoder reconstructs the graph from the latent embedding. (Image by M. Granfors and J. Pineda.)
Global graph features unveiled by unsupervised deep learning
Mirja Granfors, Jesús Pineda, Blanca Zufiria Gerbolés, Joana B. Pereira, Carlo Manzo, Giovanni Volpe
Machine Learning: Science and Technology, 7, 045054 (2026)
arXiv: 2503.05560
doi: 10.1088/2632-2153/ae8d7f

Graphs provide a powerful framework for modeling complex systems, but their structural variability poses significant challenges for analysis and classification. To address these challenges, we introduce Graph Autoencoder Uncovering Descriptive Information (GAUDI), an unsupervised graph deep learning framework designed to capture both local details and global structure. GAUDI employs an hourglass architecture with hierarchical pooling and upsampling layers linked through skip connections, which preserve essential connectivity information throughout the encoding–decoding process. Even though identical or highly similar underlying parameters describing a system’s state can lead to significant variability in graph realizations, GAUDI consistently maps them into nearby regions of a structured and continuous latent space, effectively disentangling invariant process-level features from stochastic noise. We demonstrate GAUDI’s versatility across multiple applications, including small-world networks modeling, characterization of protein assemblies from super-resolution microscopy, analysis of collective motion in the Vicsek model, and identification of age-related changes in brain connectivity. Comparison with related approaches highlights GAUDI’s superior performance in analyzing complex graphs, providing new insights into emergent phenomena across diverse scientific domains.

Poster by M. Granfors at the Protein Folding in Real Time conference, Stockholm, 11 March 2026

Fluorescence microscopy image of yeast cells, with Hsp104-GFP marking protein aggregates, making them visible as bright spots. (Image by J. Masaryk.)
Machine learning based tracking of protein aggregates in yeast
Mirja Granfors, Jakub Masaryk, Carlo Manzo, Markus Tamas, Giovanni Volpe
Date: 11th March 2026
Time: 18:00 – 20:00
Place: Aula Medica, Karolinska Institute, Solna
Conference Protein Folding in Real Time, 11-13 March 2026, Stockholm, Sweden

Arsenic is a toxic metal linked to serious diseases like cancer and neurodegeneration. One proposed mechanism of toxicity is that arsenic causes proteins to misfold and aggregate inside cells, but the dynamics and regulation of this process remain poorly understood. Using fluorescence microscopy data from living yeast cells, we are developing a machine learning approach to automatically detect, track, and analyze protein aggregate movement over time.

Presentation by M. Granfors at BNMI 2025, Gothenburg, 20 August 2025

DeepTrack2 Logo. (Image by J. Pineda)
DeepTrack2: physics-based microscopy simulations for deep learning
Mirja Granfors, Alex Lech, Benjamin Midtvedt, Jesús Pineda, Harshith Bachimanchi, Carlo Manzo, and Giovanni Volpe
BNMI 2025, 19-22 August 2025, Gothenburg, Sweden
Date: 20 August 2025
Time: 15:00 – 15:15
Place:  Wallenberg Conference Centre

DeepTrack2 is a flexible and scalable Python library designed to generate physics-based synthetic microscopy datasets for training deep learning models. It supports a wide range of imaging modalities, including brightfield, fluorescence, darkfield, and holography, enabling the creation of synthetic samples that accurately replicate real experimental conditions. Its modular architecture empowers users to customize optical systems, incorporate optical aberrations and noise, simulate diverse objects across various imaging scenarios, and apply image augmentations. DeepTrack2 is accompanied by a dedicated GitHub page, providing extensive documentation, examples, and an active community for support and collaboration: https://github.com/DeepTrackAI/DeepTrack2.

Mirja Granfors received the Best Early-Career Researcher Presentation Award at ETAI 2025, San Diego

(Photo by M. Granfors.)

Mirja Granfors received the Best Early Career Researcher Presentation Award at Emerging Topics in Artificial Intelligence (ETAI) 2025 held in San Diego, from 3 to 7 August 2025.

The award, which includes a certificate, a cash prize of $300, and a T-shirt, is presented by the organizers of the conference in collaboration with SPIE Optics + Photonics.

Mirja was awarded the prize for her presentation titled “DeepTrack2: physics-based microscopy simulations for deep learning”. Below is the full abstract of her presentation:

DeepTrack2 is a flexible and scalable Python library designed to generate physics-based synthetic microscopy datasets for training deep learning models. It supports a wide range of imaging modalities, including brightfield, fluorescence, darkfield, and holography, enabling the creation of synthetic samples that accurately replicate real experimental conditions. Its modular architecture empowers users to customize optical systems, incorporate optical aberrations and noise, simulate diverse objects across various imaging scenarios, and apply image augmentations. DeepTrack2 is accompanied by a dedicated GitHub page, providing extensive documentation, examples, and an active community for support and collaboration: https://github.com/DeepTrackAI/DeepTrack2.

Presentation by M. Granfors at SPIE-ETAI, San Diego, 7 August 2025

GAUDI leverages a hierarchical graph-convolutional variational autoencoder architecture, where an encoder progressively compresses the graph into a low-dimensional latent space, and a decoder reconstructs the graph from the latent embedding. (Image by M. Granfors and J. Pineda.)
Global graph features unveiled by unsupervised geometric deep learning
Mirja Granfors, Jesús Pineda, Blanca Zufiria Gerbolés, Joana Pereira, Carlo Manzo, and Giovanni Volpe
Date: 7 August 2025
Time: 2:45 PM – 3:00 PM
Place: Conv. Ctr. Room 4

Graphs are used to model complex relationships, such as interactions between particles or connections between brain regions. The structural complexity and variability of graphs pose challenges to their efficient analysis and classification. Here, we propose GAUDI (Graph Autoencoder Uncovering Descriptive Information), a graph autoencoder that addresses these challenges. GAUDI is trained in an unsupervised manner to capture the most critical parameters of graphs in the latent space, thereby enabling the extraction of essential parameters characterizing the graphs. We demonstrate the performance of GAUDI across diverse graph data originating from complex systems, including the estimation of the parameters of Watts-Strogatz graphs, the classification of protein assembly structures from single-molecule localization microscopy data, the analysis of collective behaviors, and correlations between brain connections and age. This approach offers a robust framework for efficiently analyzing and interpreting complex graph data, facilitating the extraction of meaningful patterns and insights across a wide range of applications.

Presentation by M. Granfors at SPIE-ETAI, San Diego, 5 August 2025

DeepTrack2 Logo. (Image by J. Pineda)
DeepTrack2: physics-based microscopy simulations for deep learning
Mirja Granfors, Alex Lech, Benjamin Midtvedt, Jesús Pineda, Harshith Bachimanchi, Carlo Manzo, and Giovanni Volpe
Date: 5 August 2025
Time: 2:45 PM – 3:00 PM
Place: Conv. Ctr. Room 4

DeepTrack2 is a flexible and scalable Python library designed to generate physics-based synthetic microscopy datasets for training deep learning models. It supports a wide range of imaging modalities, including brightfield, fluorescence, darkfield, and holography, enabling the creation of synthetic samples that accurately replicate real experimental conditions. Its modular architecture empowers users to customize optical systems, incorporate optical aberrations and noise, simulate diverse objects across various imaging scenarios, and apply image augmentations. DeepTrack2 is accompanied by a dedicated GitHub page, providing extensive documentation, examples, and an active community for support and collaboration: https://github.com/DeepTrackAI/DeepTrack2.

Mirja Granfors won best early-career researcher presentation award at AnDi+ 2025, Gothenburg

Mirja Granfors receives the award. From left to right: Giorgio Volpe, Mirja Granfors, Wojciech Chachólski, Arrate Muñoz-Barrutia, Gorka Muñoz-Gil, Carlo Manzo. (Photo by A. Callegari.)

Mirja Granfors won the best early career researcher presentation award at AnDi+ 2025 workshop (AI for Bioimaging Beyond Trajectory Analysis) held in Gothenburg, from 2 June – 5 June 2025.

The award, consisting of a certificate and a cash prize of 250€, is sponsored by Nanophotonics.

Mirja was awarded the prize for her presentation titled “DeepTrack2: Physics-based Microscopy Simulations for Deep Learning & Deeplay: Enhancing PyTorch with Customizable and Reusable Neural Networks”. In her presentation, she presented the Python libraries DeepTrack2 and Deeplay, both developed by the Soft Matter Lab to support AI-driven microscopy.

DeepTrack2 is a flexible and scalable Python library designed to generate physics-based synthetic microscopy datasets for training deep learning models. It supports a wide range of imaging modalities, including brightfield, fluorescence, darkfield, and holography, enabling the creation of synthetic samples that accurately replicate real experimental conditions. Its modular architecture empowers users to customize optical systems, incorporate optical aberrations and noise, simulate diverse objects across various imaging scenarios, and apply image augmentations.

Deeplay is a flexible Python library for deep learning that simplifies the definition and optimization of neural networks. It provides an intuitive framework that makes it easy to define and train models. With its modular design, Deeplay enables users to efficiently build and refine complex neural network architectures by seamlessly integrating reusable components.

Presentation by M. Granfors at EUROMECH Colloquium 656 in Gothenburg, 22 May 2025

Mirja Granfors presenting at the EUROMECH Colloquium. (Photo by A. Lech.)
DeepTrack2: Physics-based Microscopy Simulations for Deep Learning
Mirja Granfors

Date: 22 May 2025
Time: 15:15
Place: Veras Gräsmatta, Gothenburg
Part of the EUROMECH Colloquium 656 Data-Driven Mechanics and Physics of Materials

DeepTrack2 is a flexible and scalable Python library designed to generate physics-based synthetic microscopy datasets for training deep learning models. It supports a wide range of imaging modalities, including brightfield, fluorescence, darkfield, and holography, enabling the creation of synthetic samples that accurately replicate real experimental conditions. Its modular architecture empowers users to customize optical systems, incorporate optical aberrations and noise, simulate diverse objects across various imaging scenarios, and apply image augmentations. DeepTrack2 is accompanied by a dedicated GitHub page, providing extensive documentation, examples, and an active community for support and collaboration: https://github.com/DeepTrackAI/DeepTrack2.