| 1. |
- Bachimanchi, Harshith, et al.
(författare)
-
Deep-learning-powered data analysis in plankton ecology
- 2024
-
Ingår i: Limnology And Oceanography Letters. - 2378-2242.
-
Forskningsöversikt (refereegranskat)abstract
- 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.
|
|
| 2. |
- Bachimanchi, Harshith, et al.
(författare)
-
Harnessing nonlinear frequency upconversion of Talbot effect with flexible Talbot lengths
- 2024
-
Ingår i: OPTICS EXPRESS. - 1094-4087. ; 32:9
-
Tidskriftsartikel (refereegranskat)abstract
- We report on a simple experimental scheme demonstrating nonlinear frequency upconversion of the Talbot effect with controllable Talbot lengths at high conversion efficiency. Using a microlens array (MLA) as an array illuminator at 1064 nm onto a 1.2 -mm -thick BiBO crystal, we have observed the second harmonic Talbot effect in green at 532 nm with a Talbot length twice that of the pump Talbot length. However, the Talbot length is constant for fixed parameters of the periodic object and the laser wavelength. With the formulation of a suitable theoretical framework, we have implemented a generic experimental scheme based on the Fourier transformation technique to independently control the Talbot lengths of the MLA in both the pump and the second harmonic, overcoming the stringent dependence of MLA parameters on the self-images. Deploying the current technique, we have been able to tune the Talbot lengths from z T = 26 cm to z T = 62.4 cm in the pump and z T = 12.4 cm to z T = 30.8 cm in the second harmonic, respectively. The single pass conversion efficiency of the Talbot images is 2.91% W - 1 , an enhancement of a factor of 10 6 as compared to the previous reports. This generic experimental scheme can be used to generate long-range self-images of periodic structures and also to program desired Talbot planes at required positions at both pump and upconverted frequency to avoid any mechanical constraints of experiments.
|
|
| 3. |
- Bachimanchi, Harshith, et al.
(författare)
-
Microplankton life histories revealed by holographic microscopy and deep learning
- 2022
-
Ingår i: eLife. - 2050-084X. ; 11
-
Tidskriftsartikel (refereegranskat)abstract
- The marine microbial food web plays a central role in the global carbon cycle. However, our mechanistic understanding of the ocean is biased toward its larger constituents, while rates and biomass fluxes in the microbial food web are mainly inferred from indirect measurements and ensemble averages. Yet, resolution at the level of the individual microplankton is required to advance our understanding of the microbial food web. Here, we demonstrate that, by combining holographic microscopy with deep learning, we can follow microplanktons throughout their lifespan, continuously measuring their three-dimensional position and dry mass. The deep-learning algorithms circumvent the computationally intensive processing of holographic data and allow rapid measurements over extended time periods. This permits us to reliably estimate growth rates, both in terms of dry mass increase and cell divisions, as well as to measure trophic interactions between species such as predation events. The individual resolution provides information about selectivity, individual feeding rates, and handling times for individual microplanktons. The method is particularly useful to detail the rates and routes of organic matter transfer in micro-zooplankton, the most important and least known group of primary consumers in the oceans. Studying individual interactions in idealized small systems provides insights that help us understand microbial food webs and ultimately larger-scale processes. We exemplify this by detailed descriptions of micro-zooplankton feeding events, cell divisions, and long-term monitoring of single cells from division to division.
|
|
| 4. |
- Campos, Rebeca Ferrer, et al.
(författare)
-
Bubble-propelled micromotors for ammonia generation
- 2023
-
Ingår i: NANOSCALE. - 2040-3364 .- 2040-3372. ; 15:38, s. 15785-15793
-
Tidskriftsartikel (refereegranskat)abstract
- Micromotors have emerged as promising tools for environmental remediation, thanks to their ability to autonomously navigate and perform specific tasks at the microscale. In this study, we present the development of MnO2 tubular micromotors modified with laccase for enhanced oxidation of organic pollutants by providing an additional oxidative catalytic pathway for pollutant removal. These modified micromotors exhibit efficient ammonia generation through the catalytic decomposition of urea, suggesting their potential application in the field of green energy generation. Compared to bare micromotors, the MnO2 micromotors modified with laccase exhibit a 20% increase in rhodamine B degradation. Moreover, the generation of ammonia increased from 2 to 31 ppm in only 15 min, evidencing their high catalytic activity. To enable precise tracking of the micromotors and measurement of their speed, a deep-learning-based tracking system was developed. Overall, this work expands the potential applicability of bio-catalytic tubular micromotors in the energy field. Here, we introduce self-propelled biocatalytic micromotors for simultaneous organic pollutant removal and green energy generation. The study demonstrates remarkable results, showcasing the potential to generate ammonia from wastewater in short time.
|
|
| 5. |
- Midtvedt, Benjamin, et al.
(författare)
-
Single-shot self-supervised object detection in microscopy
- 2022
-
Ingår i: Nature Communications. - : Springer Science and Business Media LLC. - 2041-1723 .- 2041-1723. ; 13:1
-
Tidskriftsartikel (refereegranskat)abstract
- Object detection is a fundamental task in digital microscopy, where machine learning has made great strides in overcoming the limitations of classical approaches. The training of state-of-the-art machine-learning methods almost universally relies on vast amounts of labeled experimental data or the ability to numerically simulate realistic datasets. However, experimental data are often challenging to label and cannot be easily reproduced numerically. Here, we propose a deep-learning method, named LodeSTAR (Localization and detection from Symmetries, Translations And Rotations), that learns to detect microscopic objects with sub-pixel accuracy from a single unlabeled experimental image by exploiting the inherent roto-translational symmetries of this task. We demonstrate that LodeSTAR outperforms traditional methods in terms of accuracy, also when analyzing challenging experimental data containing densely packed cells or noisy backgrounds. Furthermore, by exploiting additional symmetries we show that LodeSTAR can measure other properties, e.g., vertical position and polarizability in holographic microscopy.
|
|
| 6. |
- Pineda, Jesus, et al.
(författare)
-
Geometric deep learning reveals the spatiotemporal features of microscopic motion
- 2023
-
Ingår i: Nature Machine Intelligence. - : Springer Science and Business Media LLC. - 2522-5839. ; 5, s. 71-82
-
Tidskriftsartikel (refereegranskat)abstract
- Despite recent improvements in microscopy acquisition methods, extracting quantitative information from biological experiments in crowded conditions is a challenging task. Pineda and colleagues propose a geometric deep-learning-based framework for automated trajectory linking and dynamical property estimation that is able to effectively deal with complex biological scenarios. The characterization of dynamical processes in living systems provides important clues for their mechanistic interpretation and link to biological functions. Owing to recent advances in microscopy techniques, it is now possible to routinely record the motion of cells, organelles and individual molecules at multiple spatiotemporal scales in physiological conditions. However, the automated analysis of dynamics occurring in crowded and complex environments still lags behind the acquisition of microscopic image sequences. Here we present a framework based on geometric deep learning that achieves the accurate estimation of dynamical properties in various biologically relevant scenarios. This deep-learning approach relies on a graph neural network enhanced by attention-based components. By processing object features with geometric priors, the network is capable of performing multiple tasks, from linking coordinates into trajectories to inferring local and global dynamic properties. We demonstrate the flexibility and reliability of this approach by applying it to real and simulated data corresponding to a broad range of biological experiments.
|
|
| 7. |
- Volpe, Giovanni, et al.
(författare)
-
Roadmap on Deep Learning for Microscopy
- 2023
-
Annan publikation (övrigt vetenskapligt/konstnärligt)abstract
- A multi-authored roadmap paper on deep learning in microscopy. Our contribution a section on deep learning in microscopy of plankton
|
|
