Deep neural networks are increasingly applied in many branches of applied science such as computer vision and image processing by increasing performances of instruments. Different deep architectures such as convolutional neural networks or Vision Transformers can be used in advanced coherent imaging techniques such as digital holography to extract various metrics such as autofocusing reconstruction distance or 3D position determination in order to target automated microscopy or real-time phase image restitution. Deep neural networks can be trained with both datasets simulated and experimental holograms, by transfer learning. Overall, the application of deep neural networks in digital holographic microscopy and 3D computer micro-vision has the potential to significantly improve the robustness and processing speed of holograms to infer and control a 3D position for applications in micro-robotics.

Modern optical microscopes, from super-resolved fluorescence to label-free mechanisms of contrast, are powerful instruments able to produce images that are rich sources of molecular information towards an unprecedented insight into the morphological and functional properties of biological cells at the nanoscale. Today we are in the realm of multimodal optical microscopy boosted by artificial intelligence that makes intelligent the microscope. Super-resolved fluorescence microscopy, incorporating photochemical parameters from brightness to lifetime, and non-linear approaches, like those associated with multi-photon excitation able to exploit intrinsic fluorescence and SHG/THG, is coupled to label-free polarisation methods like Mueller matrix microscopy, expanding the available data set. Such a data set is the core for developing an artificial microscope aiming to transform a label-free interrogation of the sample into a molecular-rich fluorescence-based image. The intelligent microscope is AI-guided through a computational core based on three modules based on a convolutional neural network (CNN) and a tensor independent component analysis (tICA) un-supervised machine learning within a supervised deep learning strategy having the ambitious target to create a robust virtual environment "to see "what we could not perceive before". An interesting case study is related to understanding the visualisation of chromatin organisation.

Optical approaches have made great strides enabling high-speed, scalable computing necessary for modern deep learning and AI applications. In this study, we introduce a multilayer optoelectronic computing framework that alternates between optical and optoelectronic layers to implement matrix-vector multiplications and rectified linear functions, respectively. The system is designed to be real-time and parallelized, utilizing arrays of light emitters and detectors connected with independent analog electronics. We experimentally demonstrate the operation of our system and compare its performance to a single-layer analog through simulations.

Constraint Bayesian optimization approach is used to optimize the beam pointing and spatial filtering of a laser beam using the capillary transmission and the output beam profile, as the optimization criteria. We have demonstrated that the developed method was able to robustly find the optimal laser parameters and it will be implemented in the SwissFEL UV photocathode laser in the future.

Encoded-diffraction hybrid systems—optical encoding and simple electronic decoding—offers feature distillation via model training. Additionally, the most faithfully reconstructed images are not the ones that are best classified. We parametrize our results with singular value decomposition (SVD) entropy, a proxy for image complexity.

Despite the breadth of scientific literature on micro- and nanoplastics (MNPs), a standardized procedure for detecting MNPs is still lacking so far, leading to incomparable results between published studies. This work innovatively proposed the combination of machine learning with advanced optical photothermal infrared (O-PTIR) spectroscopy to develop an efficient and reliable detection framework for MNPs. Spectra of MPs and non-MPs were first collected and inputted to build a classification model, based on which four important wavenumbers were selected. A simplified support vector machine (SVM) model was subsequently developed using the selected four wavenumbers. Good predictive ability was evidenced by a high accuracy of 0.9133. The developed method can improve speed as well as the reliability of results, having a great potential for routine analysis of MNPs, ultimately leading to the standardization of detection methods.