Label-free concurrent 5-modal microscopy (Co5M) resolves unknown spatio-temporal processes in wound healing

Abstract

The non-invasive investigation of multiple biological processes remains a methodological challenge as it requires capturing different contrast mechanisms, usually not available with any single modality. Intravital microscopy has played a key role in dynamically studying biological morphology and function, but it is generally limited to resolving a small number of contrasts, typically generated by the use of transgenic labels, disturbing the biological system. We introduce concurrent 5-modal microscopy (Co5M), illustrating a new concept for label-free in vivo observations by simultaneously capturing optoacoustic, two-photon excitation fluorescence, second and third harmonic generation, and brightfield contrast. We apply Co5M to non-invasively visualize multiple wound healing biomarkers and quantitatively monitor a number of processes and features, including longitudinal changes in wound shape, microvascular and collagen density, vessel size and fractality, and the plasticity of sebaceous glands. Analysis of these parameters offers unique insights into the interplay of wound closure, vasodilation, angiogenesis, skin contracture, and epithelial reformation in space and time, inaccessible by other methods. Co5M challenges the conventional concept of biological observation by yielding multiple simultaneous parameters of pathophysiological processes in a label-free mode.

High-speed optical resolution photoacoustic microscopy with MEMS scanner using a novel and simple distortion correction method

“…S1 ). This scanning sequence was constructed based on the theory used in previous research on high-speed PAM 35 , 43 , 48 . As an overview of the scanning sequence, the PA data acquisition in the X-direction was performed during the half cycle of the MEMS scan, controlled by the driving frequency.…”

Spiking neural networks for predictive and explainable modelling of multimodal streaming data with a case study on financial time series and online news

“…Integrated modelling of multimodal streaming data is an open problem [1][2][3][4][5] . Methods for integrating multimodal data into one machine learning model have been developed across domain areas, such as fault detection and diagnostics in the context of sparse multimodal data and expert knowledge assistance [6][7][8] . In 9 , application of integrated approach is demonstrated for hydro generators.…”