High-throughput sperm assay using label-free microscopy: morphometric comparison between different sperm structures of boar and stallion spermatozoa

Label-free, high-throughput holographic imaging to evaluate mammalian gametes and embryos†

Assisted reproduction is one of the significant tools to treat human infertility. Morphological assessment is the primary method to determine sperm and embryo viability during in vitro fertilization cycles. It has the advantage of being a quick, convenient, and inexpensive means of assessment. However, visual observation is of limited predictive value for early embryo morphology. It has led many to search for other imaging tools to assess the reproductive potential of a given embryo. The limitations of visual assessment apply to both humans and animals. One recent innovation in assisted reproduction technology imaging is interferometric phase microscopy, also known as holographic microscopy. Interferometric phase microscopy/quantitative phase imaging is the next likely progression of analytical microscopes for the assisted reproduction laboratory. The interferometric phase microscopy system analyzes waves produced by the light as it passes through the specimen observed. The microscope collects the light waves produced and uses the algorithm to create a hologram of the specimen. Recently, interferometric phase microscopy has been combined with quantitative phase imaging, which joins phase contrast microscopy with holographic microscopy. These microscopes collect light waves produced and use the algorithm to create a hologram of the specimen. Unlike other systems, interferometric phase microscopy can provide a quantitative digital image, and it can make 2D and 3D images of the samples. This review summarizes some newer and more promising quantitative phase imaging microscopy systems for evaluating gametes and embryos. Studies clearly show that quantitative phase imaging is superior to bright field microscopy-based evaluation methods when evaluating sperm and oocytes prior to IVF and embryos prior to transfer. However, further assessment of these systems for efficacy, reproducibility, cost-effectiveness, and embryo/gamete safety must take place before they are widely adopted.

An Exploration of Current and Perspective Semen Analysis and Sperm Selection for Livestock Artificial Insemination

Artificial insemination of livestock has been a staple technology for producers worldwide for over sixty years. This reproductive technology has allowed for the rapid improvement of livestock genetics, most notably in dairy cattle and pigs. This field has experienced continuous improvements over the last six decades. Though much work has been carried out to improve the efficiency of AI, there are still many areas which continue to experience improvement, including semen analysis procedures, sperm selection techniques, sperm sexing technologies, and semen storage methods. Additionally, the use of AI continues to grow in beef cattle, horses, and small ruminants as the technology continues to become more efficient and yield higher pregnancy rates. In this review, AI trends in the various livestock species as well as cutting edge improvements in the aforementioned areas will be discussed at length. Future work will continue to refine the protocols which are used for AI and continue to increase pregnancy rates within all livestock species.

Multiscale Assay of Unlabeled Neurite Dynamics Using Phase Imaging with Computational Specificity

Primary neuronal cultures have been widely used to study neuronal morphology, neurophysiology, neurodegenerative processes, and molecular mechanism of synaptic plasticity underlying learning and memory. However, the unique behavioral properties of neurons make them challenging to study, with phenotypic differences expressed as subtle changes in neuronal arborization rather than easy-to-assay features such as cell count. The need to analyze morphology, growth, and intracellular transport has motivated the development of increasingly sophisticated microscopes and image analysis techniques. Due to its high-contrast, high-specificity output, many assays rely on confocal fluorescence microscopy, genetic methods, or antibody staining techniques. These approaches often limit the ability to measure quantitatively dynamic activity such as intracellular transport and growth. In this work, we describe a method for label-free live-cell cell imaging with antibody staining specificity by estimating the associated fluorescence signals via quantitative phase imaging and deep convolutional neural networks. This computationally inferred fluorescence image is then used to generate a semantic segmentation map, annotating subcellular compartments of live unlabeled neural cultures. These synthetic fluorescence maps were further applied to study the time-lapse development of hippocampal neurons, highlighting the relationships between the cellular dry mass production and the dynamic transport activity within the nucleus and neurites. Our implementation provides a high-throughput strategy to analyze neural network arborization dynamically, with high specificity and without the typical phototoxicity and photobleaching limitations associated with fluorescent markers.

Effect of deep transfer and multi-task learning on sperm abnormality detection

Analyzing the abnormality of morphological characteristics of male human sperm has been studied for a long time mainly because it has many implications on the male infertility problem, which accounts for approximately half of the infertility problems in the world. Yet, detecting such abnormalities by embryologists has several downsides. To clarify, analyzing sperms through visual inspection of an expert embryologist is a highly subjective and biased process. Furthermore, it takes much time for a specialist to make a diagnosis. Hence, in this paper, we proposed two deep learning algorithms that are able to automate this process. The first algorithm uses a network-based deep transfer learning approach, while the second technique, named Deep Multi-task Transfer Learning (DMTL), employs a novel combination of network-based deep transfer learning and multi-task learning to classify sperm's head, vacuole, and acrosome as either normal or abnormal. This DMTL technique is capable of classifying all the aforementioned parts of the sperm in a single prediction. Moreover, this is the first time that the concept of multi-task learning has been introduced to the field of Sperm Morphology Analysis (SMA). To benchmark our algorithms, we employed a freely-available SMA dataset named MHSMA. During our experiments, our algorithms reached the state-of-the-art results on the accuracy, precision, and f_0.5, as well as other important metrics, such as the Matthews Correlation Coefficient on one, two, or all three labels. Notably, our algorithms increased the accuracy of the head, acrosome, and vacuole by 6.66%, 3.00%, and 1.33%, and reached the accuracy of 84.00%, 80.66%, and 94.00% on these labels, respectively. Consequently, our algorithms can be used in health institutions, such as fertility clinics, with further recommendations to practically improve the performance of our algorithms.