Through digital imaging, microscopy has evolved from primarily being a means for visual observation of life at the micro- and nano-scale, to a quantitative tool with ever-increasing resolution and throughput. Artificial intelligence, deep neural networks, and machine learning are all niche terms describing computational methods that have gained a pivotal role in microscopy-based research over the past decade. This Roadmap is written collectively by prominent researchers and encompasses selected aspects of how machine learning is applied to microscopy image data, with the aim of gaining scientific knowledge by improved image quality, automated detection, segmentation, classification and tracking of objects, and efficient merging of information from multiple imaging modalities. We aim to give the reader an overview of the key developments and an understanding of possibilities and limitations of machine learning for microscopy. It will be of interest to a wide cross-disciplinary audience in the physical sciences and life sciences.
Over the past decade, optical tweezers (OT) have been increasingly used in neuroscience for studies of molecules and neuronal dynamics, as well as for the study of model organisms as a whole. Compared to other areas of biology, it has taken much longer for OT to become an established tool in neuroscience. This is, in part, due to the complexity of the brain and the inherent difficulties in trapping individual molecules or manipulating cells located deep within biological tissue. Recent advances in OT, as well as parallel developments in imaging and adaptive optics, have significantly extended the capabilities of OT. In this review, we describe how OT became an established tool in neuroscience and we elaborate on possible future directions for the field. Rather than covering all applications of OT to neurons or related proteins and molecules, we focus our discussions on studies that provide crucial information to neuroscience, such as neuron dynamics, growth, and communication, as these studies have revealed meaningful information and provide direction for the field into the future.
We present an advanced optical-trapping method that is capable of trapping arbitrary shapes of transparent and absorbing particles in air. Two parabolic reflectors were used to reflect the inner and outer parts of a single hollow laser beam, respectively, to form two counter-propagating conical beams and bring them into a focal point for trapping. This novel design demonstrated high trapping efficiency and strong trapping robustness with a simple optical configuration. Instead of using expensive microscope objectives, the parabolic reflectors can not only achieved large numerical aperture (N.A.) focusing, but were also able to focus the beam far away from optical surfaces to minimize optics contamination. This design also offered a large free space for flexible integration with other measuring techniques, such as optical-trapping Raman spectroscopy, for on-line single particle characterization.
