Soft Matter Lab

Book “Deep Learning Crash Course” published at No Starch Press

The book Deep Learning Crash Course, authored by Giovanni Volpe, Benjamin Midtvedt, Jesús Pineda, Henrik Klein Moberg, Harshith Bachimanchi, Joana B. Pereira, and Carlo Manzo, has been published online by No Starch Press in July 2024.

Preorder Discount
A preorder discount is available: preorders with coupon code PREORDER will receive 25% off. Link: Preorder @ No Starch Press | Deep Learning Crash Course

Links
@ No Starch Press

Citation
Giovanni Volpe, Benjamin Midtvedt, Jesús Pineda, Henrik Klein Moberg, Harshith Bachimanchi, Joana B. Pereira, and Carlo Manzo. Deep Learning Crash Course. No Starch Press.
ISBN-13: 9781718503922

Nanoalignment by Critical Casimir Torques featured in the Editors’ Highlights of Nature Communications

Artist rendition of a disk-shaped microparticle trapped above a circular uncoated pattern within a thin gold layer coated on a glass surface. *(Image by the Authors of the manuscript.)*

Our article, entitled Nanoalignment by Critical Casimir Torques, has been selected as a featured article by the editor at Nature Communications. This recognition highlights the significance of our research within the field of applied physics and mathematics.

The editors have included our work in their Editors’ Highlights webpage, which showcases the 50 best papers recently published in this area. You can view the feature on the Editors’ Highlights page (https://www.nature.com/ncomms/editorshighlights) as well as on the journal homepage (https://www.nature.com/ncomms/).

Screenshot from the Editors’ Highlights page of Nature Communications, dated 2 July 2024.

Nanoalignment by Critical Casimir Torques published in Nature Communications

Nanoalignment by Critical Casimir Torques
Gan Wang, Piotr Nowakowski, Nima Farahmand Bafi, Benjamin Midtvedt, Falko Schmidt, Agnese Callegari, Ruggero Verre, Mikael Käll, S. Dietrich, Svyatoslav Kondrat, Giovanni Volpe
Nature Communications, 15, 5086 (2024)
DOI: 10.1038/s41467-024-49220-1
arXiv: 2401.06260

The manipulation of microscopic objects requires precise and controllable forces and torques. Recent advances have led to the use of critical Casimir forces as a powerful tool, which can be finely tuned through the temperature of the environment and the chemical properties of the involved objects. For example, these forces have been used to self-organize ensembles of particles and to counteract stiction caused by Casimir-Liftshitz forces. However, until now, the potential of critical Casimir torques has been largely unexplored. Here, we demonstrate that critical Casimir torques can efficiently control the alignment of microscopic objects on nanopatterned substrates. We show experimentally and corroborate with theoretical calculations and Monte Carlo simulations that circular patterns on a substrate can stabilize the position and orientation of microscopic disks. By making the patterns elliptical, such microdisks can be subject to a torque which flips them upright while simultaneously allowing for more accurate control of the microdisk position. More complex patterns can selectively trap 2D-chiral particles and generate particle motion similar to non-equilibrium Brownian ratchets. These findings provide new opportunities for nanotechnological applications requiring precise positioning and orientation of microscopic objects.

Plenary Talk by G. Volpe at ENO-CANCOA, Cartagena, Colombia, 13 June 2024

DeepTrack 2.1 Logo. *(Image from DeepTrack 2.1 Project)*

Deep learning for microscopy
Giovanni Volpe
Encuentro Nacional de Óptica y la Conferencia Andina y del Caribe en Óptica y sus Aplicaciones(ENO-CANCOA)
Cartagena, Colombia, 13 June 2024

Video microscopy has a long history of providing insights and breakthroughs for a broad range of disciplines, from physics to biology. Image analysis to extract quantitative information from video microscopy data has traditionally relied on algorithmic approaches, which are often difficult to implement, time consuming, and computationally expensive. Recently, alternative data-driven approaches using deep learning have greatly improved quantitative digital microscopy, potentially offering automatized, accurate, and fast image analysis. However, the combination of deep learning and video microscopy remains underutilized primarily due to the steep learning curve involved in developing custom deep-learning solutions.

To overcome this issue, we have introduced a software, currently at version DeepTrack 2.1, to design, train and validate deep-learning solutions for digital microscopy. We use it to exemplify how deep learning can be employed for a broad range of applications, from particle localization, tracking and characterization to cell counting and classification. Thanks to its user-friendly graphical interface, DeepTrack 2.1 can be easily customized for user-specific applications, and, thanks to its open-source object-oriented programming, it can be easily expanded to add features and functionalities, potentially introducing deep-learning-enhanced video microscopy to a far wider audience.

Jason Lewis joins the Soft Matter Lab

Jason Lewis started to work as a researcher at the Physics Department of the University of Gothenburg on 1st June 2024.

Jason received his Ph.D. degree in Complexity Science from the University of Warwick, UK, with a thesis titled “Topological models of swarming”, which studied the dynamics of bird flocks, specifically under topological constraints, via theory and numerical simulation.

After his PhD, he undertook a postdoc in the group of Joakim Stenhammar at Lund University, Sweden, where he investigated chemotaxis and the collective behaviour of microswimmers, known as active turbulence, in addition to other projects at the interface of machine learning and active matter.

His research focuses on the theory and simulation of active matter systems at all scales, specifically on modelling the structure and dynamics of self-organising groups of motile robots.

Harshith Bachimanchi won best early-career researcher presentation award at AIM 2024, La Ràpita, Spain

Committee and winners for the IOP award at AIM24. From left to right: Susan Cox, Wylie Ahmed, Celia Rowland (IOP), Harshith Bachimanchi, Blanca Zufiria Gerboles, Mirja Granfors, Carlotta Viana, Gajendra Pratap Singh, Giorgio Volpe. *(Photo by G. Volpe)*

Harshith Bachimanchi won the best early career researcher presentation award at AIM 2024 meeting (Artificial Intelligence for iMaging 2024) held in La Ràpita, Spain, from 26 May – 1 June 2024.

The award, consisting of a certificate, and a cash prize of 500 €, is sponsored by Journal of Physics: Photonics (JPhys Photonics) from IOP Publishing.

Harshith received the prize for his presentation on “Bringing microplankton to focus: Holography and deep learning”, where he demonstrated that the combination of holographic microscopy and deep learning can be used to follow the marine microorganisms throughout their lifespan, continuously measuring their three-dimensional positions and dry mass. The deep-learning algorithms circumvent the computationally intensive processing of holographic data and allow rapid measurements over extended periods of time. He exemplified this by showing detailed descriptions of micro-zooplankton feeding events, cell divisions, and long-term monitoring of single cells from division to division.

The article related to this presentation can be found at the following link: Microplankton life histories revealed by holographic microscopy and deep learning.

Award Certificate. *(Image by H. Bachimanchi)*

Harshith Bachimanchi receives the prize. *(Photo by A. Callegari)*

Mirja Granfors won best early-career researcher presentation award at AIM 2024, La Ràpita, Spain

Mirja Granfors won the best early career researcher presentation award at AIM 2024 meeting (Artificial Intelligence for iMaging 2024) held in La Ràpita, Spain, from 26 May – 1 June 2024.

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 “Global graph features unveiled by unsupervised geometric deep learning”. In her presentation, she introduced a novel graph autoencoder designed to capture complex relationships modelled by graphs. She demonstrated the performance of the network across a spectrum of datasets, including the classification of protein assembly structures from single-molecule localization microscopy data, as well as the analysis of collective behavior and correlations between brain connections and age.

Award Certificate. *(Image by M. Granfors)*

Mirja presents at AIM24 Conference. *(Photo by N. C. Palmero Crúz)*

Seminar by A. Rohrbach on 15 May 2024

Correlated photons in superresolution imaging and correlated motions in biophysical interaction
Alexander Rohrbach
15 May 2024
12:30
Nexus

Abstract
Our research concentrates on light scattering at small biological structures enabling image formation and particle tracking in biophysics.
Coherent light, i.e. correlated photons enable higher scattering cross-sections than for instance incoherent fluorescence light. Thereby laser light enables to acquire images with millisecond integration times and small motion blur of dynamic particles, such as viruses in the cell periphery. The inherent speckle formation in coherent imaging is avoided by a novel technique called Rotating Coherent Scattering (ROCS) microscopy, which is the only technique that can image diffusing viruses and thereby allows to investigate their binding behavior to the cell periphery.
In the second part of my talk I discuss correlated particle motions, i.e. timescale dependent memory effects in viscoelastic media such as the cell periphery. Using a frequency decomposition of the tracked particle motions, apparently invisible binding of particles to the cell can be made visible.

Short CV
I studied physics at the university of Erlangen-Nürnberg (Germany), where I did my diploma in 1994 at the institute of optics. During my PhD in physics in Heidelberg I investigated different kinds of light scattering at the University, as well as evanescent wave microscopy at the Max-Planck-Institute for medical research. In both cases I worked on applications in cell biology. After my PhD in 1998, I continued my research as a Post-Doc at the European Molecular Biology Laboratory (EMBL) in Heidelberg. I intensified my studies on microscopy, light scattering and optical forces. In 2001 I became project leader of the photonic force microscopy group at EMBL, where I concentrated on the further technical development of this scanning probe microscopy and on applications in biophysics and soft matter physics. In 2005 I was awarded with the habilitation in physics at the university of Heidelberg. Since January 2006 I have been a full professor for Bio- and Nano-Photonics at IMTEK, Faculty of engineering and since 2007 also a member of the physics faculty, University of Freiburg.
I love mathematical models and I hate when the performance of scientists is squeezed into metric numbers.

Deep-learning-powered data analysis in plankton ecology published in Limnology and Oceanography Letters

Segmentation of two plankton species using deep learning (N. scintillans in blue, D. tertiolecta in green). *(Image by H. Bachimanchi.)*

Deep-learning-powered data analysis in plankton ecology
Harshith Bachimanchi, Matthew I. M. Pinder, Chloé Robert, Pierre De Wit, Jonathan Havenhand, Alexandra Kinnby, Daniel Midtvedt, Erik Selander, Giovanni Volpe
Limnology and Oceanography Letters (2024)
doi: 10.1002/lol2.10392
arXiv: 2309.08500

The implementation of deep learning algorithms has brought new perspectives to plankton ecology. Emerging as an alternative approach to established methods, deep learning offers objective schemes to investigate plankton organisms in diverse environments. We provide an overview of deep-learning-based methods including detection and classification of phytoplankton and zooplankton images, foraging and swimming behavior analysis, and finally ecological modeling. Deep learning has the potential to speed up the analysis and reduce the human experimental bias, thus enabling data acquisition at relevant temporal and spatial scales with improved reproducibility. We also discuss shortcomings and show how deep learning architectures have evolved to mitigate imprecise readouts. Finally, we suggest opportunities where deep learning is particularly likely to catalyze plankton research. The examples are accompanied by detailed tutorials and code samples that allow readers to apply the methods described in this review to their own data.

Emiliano Gómez will defend his PhD thesis on 22 May 2024

Emiliano Gómez will defend his PhD thesis on the 22th of May at 10:30. The defense will take place in KA (Chemistry Department, Johanneberg Campus)

Title: Development and Application of a software to analyse networks with multilayer graph theory and deep learning

Abstract:
Network theory gives us the tools necessary to produce a model of our brain, how the brain is wired will give us a new level of insight into its functionality. The brain network, the connectome, is formed by structural links such as synapses or fiber pathways in the brain. This connectome might also be interpreted from a statistical relationship between the flow of information, or activation correlation between brain regions. Mapping these networks can be achieved by using neuroimaging, which allows obtaining information on the brain in vivo. Different neuroimaging modalities will capture different properties of the brain. Statistical analysis is necessary for extracting meaningful insights regarding the network patterns obtained from neuroimages. For this, huge data banks are a byproduct of the need for enough data to be able to tackle medical and biological questions.

In this work, we present a software “Brain Analysis using Graph Theory 2” (BRAPH 2) (Paper I), which addresses the need for a toolbox designed for both complex graph theory and deep learning analyses of different imaging modalities. With BRAPH 2, we offer the neuroimaging community a tool that is open-source, flexible, and intuitive. BRAPH 2, at its core, comes with multi-graph capabilities. For Paper II, we employed the power of multiplex and multigraph capabilities of BRAPH 2 to analyze sex differences in brain connectivity for an aging healthy population. Finally, for Paper III, BRAPH 2 has been adapted to two new graph measures (global memory capacity, and nodal memory capacity), which obtain a prediction of memory capacity using Reservoir Computing and relate this new measure to biological and cognitive characteristics of the cohort.

Supervisor: Giovanni Volpe
Examiner: Raimund Feifel
Opponent: Saikat Chatterjee, KTH, Stockholm
Committee: Marija Cvijovic, Alireza Salami, Wojciech Chachólski
Alternate board member: Mats Granath