Artificial neurons can be combined in a dense neural network (DNN), where the input layer is connected to the output layer via a set of hidden layers. (Image by the Authors.)Deep learning in light–matter interactions
Daniel Midtvedt, Vasilii Mylnikov, Alexander Stilgoe, Mikael Käll, Halina Rubinsztein-Dunlop and Giovanni Volpe
Nanophotonics, 11(14), 3189-3214 (2022)
doi: 10.1515/nanoph-2022-0197
The deep-learning revolution is providing enticing new opportunities to manipulate and harness light at all scales. By building models of light–matter interactions from large experimental or simulated datasets, deep learning has already improved the design of nanophotonic devices and the acquisition and analysis of experimental data, even in situations where the underlying theory is not sufficiently established or too complex to be of practical use. Beyond these early success stories, deep learning also poses several challenges. Most importantly, deep learning works as a black box, making it difficult to understand and interpret its results and reliability, especially when training on incomplete datasets or dealing with data generated by adversarial approaches. Here, after an overview of how deep learning is currently employed in photonics, we discuss the emerging opportunities and challenges, shining light on how deep learning advances photonics.
Emergent Complex Behaviors in Active Matter
Giovanni Volpe
9 June 2022, 14:30 (IST)
Online for MNBF Workshop: International Workshop On Active Systems
IIT Madras, India, 8-9 June 2022
Kymographs of DNA inside Channel II. (Image by the Authors.)Label-free nanofluidic scattering microscopy of size and mass of single diffusing molecules and nanoparticles
Barbora Špačková, Henrik Klein Moberg, Joachim Fritzsche, Johan Tenghamn, Gustaf Sjösten, Hana Šípová-Jungová, David Albinsson, Quentin Lubart, Daniel van Leeuwen, Fredrik Westerlund, Daniel Midtvedt, Elin K. Esbjörner, Mikael Käll, Giovanni Volpe & Christoph Langhammer
Nature Methods 19, 751–758 (2022)
doi: 10.1038/s41592-022-01491-6
Label-free characterization of single biomolecules aims to complement fluorescence microscopy in situations where labeling compromises data interpretation, is technically challenging or even impossible. However, existing methods require the investigated species to bind to a surface to be visible, thereby leaving a large fraction of analytes undetected. Here, we present nanofluidic scattering microscopy (NSM), which overcomes these limitations by enabling label-free, real-time imaging of single biomolecules diffusing inside a nanofluidic channel. NSM facilitates accurate determination of molecular weight from the measured optical contrast and of the hydrodynamic radius from the measured diffusivity, from which information about the conformational state can be inferred. Furthermore, we demonstrate its applicability to the analysis of a complex biofluid, using conditioned cell culture medium containing extracellular vesicles as an example. We foresee the application of NSM to monitor conformational changes, aggregation and interactions of single biomolecules, and to analyze single-cell secretomes.
Harshith Bachimanchi’s half-time seminar. (Photo by Y.-W. Chang.)Harshith Bachimanchi completed the first half of his doctoral studies and defended his half-time on 10th May 2022.
The presentation was held in hybrid format, with part of the audience present in the Nexus room and the rest connected through zoom. The half-time consisted of a presentation of his past and planned projects followed by discussion and questions proposed by his opponent Bernhard Mehlig.
The presentation started with a description of his project about combining holographic microscopy with deep learning to measure the dry mass and three-dimensional swimming patterns of marine microorganisms (Microplankton life histories revealed by holographic microscopy and deep learning). Thereafter, he discussed about some of the new experiments in marine microbial ecology where the technique is currently being used. In the last section, he outlined the proposed continuation of his PhD on studying active matter systems in marine microscopic environments using holographic microscopy and artificial neural networks.
In the event, held on Tuesday, 15 March 2022, 16:00-19:00, the ten teams that had gone through the training at the Startup Camp and developed their company ideas, pitched their companies on stage to a panel of entrepreneur experts, the other nine teams, and all business coaches at Chalmers Ventures. DeepTrack obtained the first place among the ten participants. Congrats!
Here a few pictures from the final pitching event of the Startup Camp.
Henrik. (Picture by Jonas Sandwall, Chalmers Ventures.) DeepTrack team members (left to right) Henrik, Giovanni and Jesus. (Picture by Jonas Sandwall, Chalmers Ventures.) Panelists. (Picture by Jonas Sandwall, Chalmers Ventures.)
An illustration of anomalous diffusion. (Image by Gorka Muñoz-Gil.)The Anomalous Diffusion Challenge: Objective comparison of methods to decode anomalous diffusion
Giovanni Volpe Complex Lagrangian Problems of Particles in Flows
Online, 15 March 2022, 10:15 CET
Deviations from Brownian motion leading to anomalous diffusion are found in transport dynamics from quantum physics to life sciences. The characterization of anomalous diffusion from the measurement of an individual trajectory is a challenging task, which traditionally relies on calculating the trajectory mean squared displacement. However, this approach breaks down for cases of practical interest, e.g., short or noisy trajectories, heterogeneous behaviour, or non-ergodic processes. Recently, several new approaches have been proposed, mostly building on the ongoing machine-learning revolution. To perform an objective comparison of methods, we gathered the community and organized an open competition, the Anomalous Diffusion challenge (AnDi). Participating teams applied their algorithms to a commonly-defined dataset including diverse conditions. Although no single method performed best across all scenarios, machine-learning-based approaches achieved superior performance for all tasks. The discussion of the challenge results provides practical advice for users and a benchmark for developers.
Links:
Complex Lagrangian Problems of Particles in Flows program
After a brief overview of artificial intelligence, machine learning and deep learning, I will present a series of recent works in which we have employed deep learning for applications in microscopy, optical tweezers, and active matter. In particular, I will explain how we employed deep learning to enhance digital video microscopy [1,2], to perform virtual staining of [3], to estimate the properties of anomalous diffusion [4,5,6], to characterize microscopic force fields [7], to improve the calculation of optical forces [8], and to characterize nanoparticles [9]. Finally, I will provide an outlook on the future for the application of deep learning in these fields.
References
[1] S. Helgadottir, A. Argun, and G. Volpe. Digital video microscopy enhanced by deep learning. Optica 6, 506 (2019).
[2] B. Midtvedt, S. Helgadottir, A. Argun, J. Pineda, D. Midtvedt, and G. Volpe. Quantitative digital microscopy with deep learning. Appl. Phys. Rev. 8, 011310 (2021).
[3] S. Helgadottir, B. Midtvedt, J. Pineda, et al. Extracting quantitative biological information from bright-field cell images using deep learning. Biophys. Rev. 2, 031401 (2021).
[4] S. Bo, F. Schmidt, R. Eichhorn, and G. Volpe. Measurement of anomalous diffusion using recurrent neural networks. Phys. Rev. E 100, 010102 (2019).
[5] A. Argun, G. Volpe, and S. Bo. Classification, inference and segmentation of anomalous diffusion with recurrent neural networks. J. Phys. A: Math. Theor. 54, 294003 (2021).
[6] G. Muñoz-Gil, G. Volpe, M. A. Garcia-March, et al. Objective comparison of methods to decode anomalous diffusion. Nat. Commun. 12, 6253 (2021).
[7] A. Argun, T. Thalheim, S. Bo, F. Cichos, and G. Volpe. Enhanced force-field calibration via machine learning. Appl. Phys. Rev. 7, 041404 (2020).
[8] I.C.D. Lenton, G. Volpe, A.B. Stilgoe, T.A. Nieminen, and H. Rubinsztein-Dunlop. Machine learning reveals complex behaviours in optically trapped particles. Mach. Learn.: Sci. Technol. 1, 045009 (2020).
[9] B. Midtvedt, E. Olsén, F. Eklund, F. Höök, C.B. Adiels, G. Volpe, and D. Midtvedt. Fast and accurate nanoparticle characterization using deep-learning-enhanced off-axis holography. ACS Nano 15, 2240 (2021).
Link:Physics Days 2022 – Future Leaders
The Physics Days 2022 is organized by the Finnish Physical Society and the Department of Applied Physics at Aalto University.
Sketch of the experimental setup for the measurement of nonadditivity of critical Casimir forces. (Image by S. Paladugu.)Experimental Study of Critical Fluctuations and Critical Casimir Forces Giovanni Volpe
729. WE-Heraeus Stiftung Seminar on Fluctuation-induced Forces
14 February 2022, 16:35 CET
Critical Casimir forces (CCF) are a powerful tool to control the self-assembly and complex behavior of microscopic and nanoscopic colloids. While CCF were theoretically predicted in 1978 [1], their first direct experimental evidence was provided only in 2008, using total internal reflection microscopy (TIRM) [2]. Since then, these forces have been investigated under various conditions, for example, by varying the properties of the involved surfaces or with moving boundaries. In addition, a number of studies of the phase behavior of colloidal dispersions in a critical mixture indicate critical Casimir forces as candidates for tuning the self-assembly of nanostructures and quantum dots, while analogous fluctuation-induced effects have been investigated, for example, at the percolation transition of a chemical sol, in the presence of temperature gradients, and even in granular fluids and active matter. In this presentation, I’ll give an overview of this field with a focus on recent results on the measurement of many-body forces in critical Casimir forces [3], the realization of micro- and nanoscopic engines powered by critical fluctuations [4, 5], and the creation of light-controllable colloidal molecules [6] and active droploids [7].
References
[1] ME Fisher and PG de Gennes. Phenomena at the walls in a critical binary mixture. C. R. Acad. Sci. Paris B 287, 207 (1978).
[2] C Hertlein, L Helden, A Gambassi, S Dietrich and C Bechinger. Direct measurement of critical Casimir forces. Nature 451, 172 (2008).
[3] S Paladugu, A Callegari, Y Tuna, L Barth, S Dietrich, A Gambassi and G Volpe. Nonadditivity of critical Casimir forces. Nat. Commun. 7, 11403 (2016).
[4] F Schmidt, A Magazzù, A Callegari, L Biancofiore, F Cichos and G Volpe. Microscopic engine powered by critical demixing. Phys. Rev. Lett. 120, 068004 (2018).
[5] F Schmidt, H Šípová-Jungová, M Käll, A Würger and G Volpe. Non-equilibrium properties of an active nanoparticle in a harmonic potential. Nat. Commun. 12, 1902 (2021).
[6] F Schmidt, B Liebchen, H Löwen and G Volpe. Light-controlled assembly of active colloidal molecules. J. Chem. Phys. 150, 094905 (2019).
[7] J Grauer, F Schmidt, J Pineda, B Midtvedt, H Löwen, G Volpe and B Liebchen. Active droploids. Nat. Commun. 12, 6005 (2021).
