Presentation by H. Zhao at SPIE-ETAI, San Diego, 7 August 2025

BRAPH2 is an open-access software that enables brain network analysis using graph theory. The newly developed BRAPH2 pipeline provides a robust framework for the analysis of individual brain connectomes. (Image by H. Zhao.)
Individual brain connectome analysis for Alzheimer’s disease classification using the BRAPH 2.0 framework
Hang Zhao, Yu-Wei Chang, Joana Pereira, Giovanni Volpe
Date: 7 August 2025
Time: 11:00–11:15 AM
Place: Conv. Ctr. Room 4

Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder that disrupts brain connectivity and cognitive function. Identifying individual-level alterations in brain networks is crucial for early diagnosis and targeted interventions. In this study, we develop an analytical pipeline, packaged as a specialized distribution of BRAPH 2.0, to investigate the individual brain connectome in AD, mild cognitive impairment (MCI), and cognitively normal (CN) individuals. This pipeline integrates graph-theoretical measures to assess functional network properties on Fludeoxyglucose positron emission tomography (FDG-PET), enabling classification of subjects into the three diagnostic groups. Our approach demonstrated the potential of individualized connectome-based analysis for disease stratification. The proposed methodology offers a data-driven framework for identifying key network alterations in AD, supporting the development of more precise diagnostic and therapeutic strategies.

Presentation by D. Pérez Guerrero at SPIE-ETAI, San Diego, 6 August 2025

Automated segnmentation of bacterial structures within a droplet. The image shows a bright-field microscopy view where a large biofilm region (green, outlined in blue) has been segmented from surrounding features. Small aggregates (yellow contours) are also highlighted. This segmentation enables structural differentiation of biofilm components for downstream quantitative analysis. (Image by D. Pérez Guerrero.)
Quantitative analysis of dynamic biofilm structures via time-resolved droplet microfluidics and artificial intelligence
Daniela Pérez Guerrero, Jesús Manuel Antúnez Domínguez, Aurélie Vigne, Daniel Midtvedt, Wylie Ahmed, Lisa Muiznieks, Giovanni Volpe and Caroline Beck Adiels.
SPIE-ETAI, San Diego, CA, USA, 03 – 07 August 2025
Date: 6 August 2025
Time: 2:30 PM – 2:45 PM PDT
Place: Conv. Ctr. Room 4

Biofilms are structured communities of microorganisms that play a crucial role in medicine, biotechnology, and ecology, contributing to microbial adaptation to any environment. Despite their significance, understanding their formation, development, and behavior remains a challenge for the community. We utilize high-throughput droplet microfluidics to enable biofilm growth in miniaturized environments, generating extensive time-lapse bright-field microscopy images. To overcome experimental constraints, including dense structural heterogeneity and skewed illumination, we developed a deep learning-based segmentation approach capable of identifying biofilm structures in complex imaging conditions. Our method operates in an unsupervised manner, reducing the need for ground truth annotations and mitigating the introduced bias of manual segmentation approaches.

Our unsupervised model effectively detects and quantifies biofilm structures, even in late-stage growth, where traditional segmentation techniques fail. The neural network demonstrates robust performance across the development cycle, distinguishing biofilm boundaries and bacteria aggregates separated from the main biofilm structure despite imaging inconsistencies. Additionally, our approach reduces manual intervention, streamlining the analysis of high-throughput biofilm imaging data.

This AI-powered segmentation technique provides a reliable and scalable tool for biofilm analysis, addressing key limitations of conventional methods. By bridging the gap between microbiology research and automated image analysis, our approach facilitates more efficient and reproducible biofilm studies.

Presentation by B. García Rodríguez at SPIE-ETAI, San Diego, 5 August 2025

Biomolecular condensates are analyzed and characterized using a holographic microscope, which utilizes a variational autoencoder (VAE) to quantify size, protein concentration, and the sharpness of the RNA-binding protein interface. (Image by B. Garcia.)
Structure and dynamics of biomolecular condensates revealed by deep learning enhanced interferometric microscopy
Berenice García Rodríguez, Makasewicz Katarzyna, Giovanni Volpe, Paolo Arosio, Daniel Sundås Midtvedt.
Date: 5 August 2025
Time: 5:00 PM – 5:15 PM PDT
Place:
Conv. Ctr. Room 4
Presentation type:
Oral

Biomolecular condensates are biological structures that form through weak, multivalent interactions primarily between low complexity domains of intrinsically disordered proteins, existing in cells as submicrometer structures. Their functions rely sensitively on physical properties such as size, internal protein concentration, and interfacial tension. However, direct measurements of these properties in submicrometer condensates remain scarce. In this work, we employ deep learning enhanced interferometric imaging to quantify size, protein concentration, and the sharpness of the interface of submicrometer condensates formed by the low complexity domain of the RNA-binding protein DDX4-LCD. We find that, within the two-phase region, DDX4-LCD forms spherical condensates with an internal protein concentration that can be slightly modulated by adding salt to the solution. Furthermore, we find that multiple populations of protein clusters coexist in the sample, some persisting even outside the two-phase region of the phase diagram, separable by their interface properties and dynamics. This hints at a more complex phase diagram of DDX4-LCD condensation than previously anticipated.

 

Presentation by N. C. Palmero Cruz at SPIE-ETAI, San Diego, 5 August 2025

Light sheet fluorescence image of a zebrafish larva showing neuronal structures in the brain (green) and gut (purple). Schematic network representations illustrate putative neuronal connectivity within each region, with nodes representing neuronal cell bodies and edges indicating potential functional or structural links. The eye is indicated for anatomical reference. (Image by N. C. Palmero Cruz.)
Exploring gut-brain connectivity using zebrafish and graph theory
Norma Caridad Palmero Cruz, Sarah B. Flensburg, Hang Zhao, Antonio Ciarlo, Caroline Beck Adiels, Gilles Vanwalleghem, Giovanni Volpe
Date: 5 August 2025
Time: 3:15 – 3:30 PM
Place: Conv. Ctr. Room 4

The gut-brain axis constitutes a fundamental communication network linking the microbiome, enteric nervous system, and central nervous system, and is increasingly recognized for its role in mental health disorders. Although substantial progress has been made in characterizing this axis, the translation of research findings into clinical applications remains limited. In the present study, zebrafish are employed as a model organism due to their genetic and physiological similarities to humans, allowing the investigation of gut-brain interactions under controlled conditions. An experimental platform has been developed that integrates microfluidic, optical tweezers, and light sheet microscopy to introduce defined bacterial strains into the gut and to manipulate the local environment. This setup enables real-time, in vivo recording of neuronal activity across different stages of infection. By applying graph theoretical analysis to high-resolution imaging data, the study aims to characterize the neural connectivity of the gut-brain axis, potentially informing new strategies for understanding and treating mental and neurological disorders.

 

Presentation by A. Lech at SPIE-ETAI, San Diego, 5 August 2025

DeepTrack2 Logo. (Image by J. Pineda)
Deeplay: Enhancing PyTorch with Customizable and Reusable Neural Networks
Alex Lech, Mirja Granfors, Benjamin Midtvedt, Jesús Pineda, Harshith Bachimanchi, Carlo Manzo, Giovanni Volpe
Date: 5 August 2025
Time: 12:00 – 12:15 PM
Place: Conv. Ctr. Room 4

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 lets users efficiently build and refine complex neural network architectures by seamlessly integrating reusable components based on PyTorch as well as adding a plethora of functionalities to alter and customize existing models without introducing boilerplate code. Deeplay is accompanied by a dedicated GitHub page, featuring extensive documentation, examples, and an active community for support and collaboration: https://github.com/DeepTrackAI/deeplay.

Invited presentation by A. Ciarlo at SPIE-MNPM, San Diego, 5 August 2025

Graphical representation of colloidal interaction measurements using the automated miniTweezer. (Image by A. Ciarlo.)
miniTweezer: an autonomous high-throughput optical tweezers platform for force spectroscopy
Antonio Ciarlo, Martin Selin, Giuseppe Pesce, Carlos Bustamante, and Giovanni Volpe
Date: 5 August 2025
Time: 9:45 AM – 10:15 AM
Place: Conv. Ctr. Room 4

Optical tweezers are essential for single-molecule studies, providing direct access to the forces underlying biological processes such as protein folding, DNA transcription, and replication. However, manual experiments are labor-intensive, costly, and slow, especially when large data sets are required. Here we present the miniTweezer, a fully autonomous force spectroscopy platform that integrates optical tweezers with real-time image analysis and adaptive control. Once configured, it operates independently to perform high-throughput trapping, molecular attachment, and force measurements. Its robust design allows for extended unattended operation, significantly increasing the efficiency of data acquisition. We demonstrate its capabilities through DNA pulling experiments and highlight its broader applicability to microparticle interactions, colloidal assembly, and soft matter mechanics. By automating force spectroscopy, the miniTweezer enables large-scale, high-precision studies in biophysics, materials science, and nanotechnology.

Presentation by A. Ciarlo at SPIE-OTOM, San Diego, 4 August 2025

Experimental trajectory (blue) of a particle trapped in air when the laser rotates at 1 Hz. The orange line represents the experimental laser trajectory. (Image by A. Ciarlo.)
Probing fluid dynamics inertial effects of particles using optical tweezers
Antonio Ciarlo, Giuseppe Pesce, Bernhard Mehlig, Antonio Sasso, and Giovanni Volpe
Date: 4 August 2025
Time: 11:45 AM – 12:00 PM
Place: Conv. Ctr. Room 3

Many natural phenomena involve dense particles suspended in a moving fluid, such as water droplets in clouds or dust grains in circumstellar disks. Studying these systems at the single particle level is challenging and requires precise control of flow and particle motion. Optical tweezers provide a powerful method for studying inertial effects in such environments. Here, we trap micrometer-sized particles in air and induce controlled dynamics by moving the trapping laser. We show that inertia becomes significant when the trap motion frequency is less than the harmonic trapping frequency, while at much higher motion frequencies, inertia has no effect. These results demonstrate the potential of trapping particles in air for studying inertial phenomena in fluids.

Soft Matter Lab members present at SPIE Optics+Photonics conference in San Diego, 3-7 August 2025

The Soft Matter Lab participates to the SPIE Optics+Photonics conference in San Diego, CA, USA, 3-7 August 2025, with the presentations listed below.

Giovanni Volpe, who serves as Symposium Chair for the SPIE Optics+Photonics Congress in 2025, is a coauthor of the following invited presentations:

Giovanni Volpe will also be the reference presenter of the following Poster contributions:

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.

Invited Presentation by B. Zufiria-Gerbolés at SPIE-ETAI, San Diego, 7 August 2025

Memory capacity in aging. A Brain reservoir computing architecture with uniform random signals applied to all nodes. (Image from the article.)
Computational memory capacity predicts aging and cognitive decline
Blanca Zufiria-Gerbolés, Mite Mijalkov, Ludvig Storm, Dániel Veréb, Zhilei Xu, Anna Canal-Garcia, Jiawei Sun, Yu-Wei Chang, Hang Zhao, Emiliano Gómez-Ruiz, Massimiliano Passaretti, Sara Garcia-Ptacek, Miia Kivipelto, Per Svenningsson, Henrik Zetterberg, Heidi Jacobs, Kathy Lüdge, Daniel Brunner, Bernhard Mehlig, Giovanni Volpe, Joana B. Pereira
SPIE-ETAI, San Diego, CA, USA, 3 – 7 August 2025
Date: 7 August 2025
Time: 11:15 AM – 11:45 AM PDT
Place: Conv. Ctr. Room 4

Using reservoir computing and diffusion-weighted imaging, we explored changes in brain connectivity patterns and their impact on cognition during aging. We found that whole-brain networks perform optimally at low densities, with performance decreasing as network density increases, particularly in regions with weaker connections. This decline was strongly associated with age and cognitive performance. Our results suggest that a core network of anatomical hubs is essential for optimal brain function, while peripheral connections are more vulnerable to aging. This study highlights the potential of reservoir computing for understanding age-related cognitive decline.

Reference
Mite Mijalkov, Ludvig Storm, Blanca Zufiria-Gerbolés, Dániel Veréb, Zhilei Xu, Anna Canal-Garcia, Jiawei Sun, Yu-Wei Chang, Hang Zhao, Emiliano Gómez-Ruiz, Massimiliano Passaretti, Sara Garcia-Ptacek, Miia Kivipelto, Per Svenningsson, Henrik Zetterberg, Heidi Jacobs, Kathy Lüdge, Daniel Brunner, Bernhard Mehlig, Giovanni Volpe, Joana B. Pereira, Computational memory capacity predicts aging and cognitive decline
Nature Communications 16, 2748 (2025)