Accurate Segmentation of IONs Using Low-Level Features and Topological Self-Similarity.

MOTIVATION: Intracellular organelle networks (IONs) such as the endoplasmic reticulum (ER) network and the mitochondrial (MITO) network serve crucial physiological functions. The morphology of these networks plays a critical role in mediating their functions. Accurate image segmentation is required for analyzing the morphology and topology of these networks for applications such as molecular mechanism analysis and drug target screening. So far, however, progress has been hindered by their structural complexity and density. RESULTS: In this study, we first establish a rigorous performance baseline for accurate segmentation of these organelle networks from fluorescence microscopy images by optimizing a baseline U-Net model. We then develop the multi-resolution encoder (MRE) and the hierarchical fusion loss (ℓhf) based on two inductive components, namely low-level features and topological self-similarity, to assist the model in better adapting to the task of segmenting IONs. Empowered by MRE and ℓhf, both U-Net and Pyramid Vision Transformer (PVT) outperform competing state-of-the-art models such as U-Net ++, HR-Net, nnU-Net, and TransUNet on custom datasets of the ER network and the MITO network, as well as on public datasets of another biological network, the retinal blood vessel network. In addition, integrating MRE and ℓhf with models such as HR-Net and TransUNet also enhances their segmentation performance. These experimental results confirm the generalization capability and potential of our approach. Furthermore, accurate segmentation of the ER network enables analysis that provides novel insights into its dynamic morphological and topological properties. AVAILABILITY AND IMPLEMENTATION: Code and data are openly accessible at https://github.com/cbmi-group/MRE. SUPPLEMENTARY INFORMATION: Supplementary information is available at Bioinformatics online.

Follistatin is a crucial chemoattractant for mouse decidualized endometrial stromal cell migration by JNK signalling.

Follistatin (FST) and activin A as gonadal proteins exhibit opposite effects on follicle-stimulating hormone (FSH) release from pituitary gland, and activin A-FST system is involved in regulation of decidualization in reproductive biology. However, the roles of FST and activin A in migration of decidualized endometrial stromal cells are not well characterized. In this study, transwell chambers and microfluidic devices were used to assess the effects of FST and activin A on migration of decidualized mouse endometrial stromal cells (d-MESCs). We found that compared with activin A, FST exerted more significant effects on adhesion, wound healing and migration of d-MESCs. Similar results were also seen in the primary cultured decidual stromal cells (DSCs) from uterus of pregnant mouse. Simultaneously, the results revealed that FST increased calcium influx and upregulated the expression levels of the migration-related proteins MMP9 and Ezrin in d-MESCs. In addition, FST increased the level of phosphorylation of JNK in d-MESCs, and JNK inhibitor AS601245 significantly attenuated FST action on inducing migration of d-MESCs. These data suggest that FST, not activin A in activin A-FST system, is a crucial chemoattractant for migration of d-MESCs by JNK signalling to facilitate the successful uterine decidualization and tissue remodelling during pregnancy.

Fast Noninvasive Morphometric Characterization of Free Human Sperms Using Deep Learning.

The selection of high-quality sperms is critical to intracytoplasmic sperm injection, which accounts for 70­80% of in vitro fertilization (IVF) treatments. So far, sperm screening is usually performed manually by clinicians. However, the performance of manual screening is limited in its objectivity, consistency, and efficiency. To overcome these limitations, we have developed a fast and noninvasive three-stage method to characterize morphology of freely swimming human sperms in bright-field microscopy images using deep learning models. Specifically, we use an object detection model to identify sperm heads, a classification model to select in-focus images, and a segmentation model to extract geometry of sperm heads and vacuoles. The models achieve an F1-score of 0.951 in sperm head detection, a z-position estimation error within ±1.5 µm in in-focus image selection, and a Dice score of 0.948 in sperm head segmentation, respectively. Customized lightweight architectures are used for the models to achieve real-time analysis of 200 frames per second. Comprehensive morphological parameters are calculated from sperm head geometry extracted by image segmentation. Overall, our method provides a reliable and efficient tool to assist clinicians in selecting high-quality sperms for successful IVF. It also demonstrates the effectiveness of deep learning in real-time analysis of live bright-field microscopy images.

Safety and efficacy of dexmedetomidine hydrochloride combined with midazolam in fiberoptic bronchoscopy in children: a prospective randomized controlled study. / ç›é ¸å³ç¾Žæ‰˜å’ªå®šè”åˆå’ªè¾¾å”‘ä»‘åœ¨å„¿ç«¥çº¤ç»´æ”¯æ°”ç®¡é•œæ£€æŸ¥ä¸­çš„å®‰å ¨æ€§å’Œæœ‰æ•ˆæ€§­­å‰çž»æ€§éšæœºå¯¹ç §ç ”究.

OBJECTIVES: To study the safety and efficacy of dexmedetomidine hydrochloride combined with midazolam in fiberoptic bronchoscopy in children. METHODS: A total of 118 children who planned to undergo fiberoptic bronchoscopy from September 2018 to February 2021 were enrolled. They were divided into a control group (n=60) and an observation group (n=58) using a random number table. The observation group received intravenous pumping of dexmedetomidine hydrochloride (2 µg/mL) at 1 µg/kg and then intravenous injection of midazolam at 0.05 mg/kg, followed by dexmedetomidine hydrochloride pumped intravenously at 0.5-0.7 µg/(kg·h) 10 minutes later to maintain anesthesia. The control group was given intravenous pumping of propofol at 2 mg/kg and then intravenous injection of midazolam at 0.05 mg/kg, followed by propofol pumped intravenously at 4-6 mg/(kg·h) 10 minutes later to maintain anesthesia. Fiberoptic bronchoscopy was performed after the children were unconscious. Heart rate (HR), respiratory rate, blood oxygen saturation, and mean arterial pressure (MAP) were recorded before inserting the bronchoscope (T0), at the time of inserting the bronchoscope (T1), when the bronchoscope reached the glottis (T2), when the bronchoscope reached the carina (T3), and when the bronchoscope entered the bronchus (T4). The intraoperative peak airway pressure (Ppeak), examination time, degree of sedation, extent of amnesia, incidence of adverse reactions, postoperative awakening time, and postoperative agitation score were also recorded. RESULTS: Compared with the control group, the observation group had significantly decreased MAP at T1 to T4 and HR at T1 to T3 (P<0.05). Compared with that at T0, MAP was significantly increased at T1 to T4 in the control group and at T3 in the observation group (P<0.05). HR was significantly higher at T1 to T3 than at T0 (P<0.05). Compared with the control group, the observation group showed significantly lower intraoperative Ppeak value, incidence of intraoperative adverse reactions, and postoperative agitation score, significantly shorter examination time, and better effects of amnesia and anesthesia (P<0.05). There was no significant difference in the degree of intraoperative sedation and postoperative awakening time between the two groups (P>0.05). CONCLUSIONS: Dexmedetomidine hydrochloride combined with midazolam is a safe and effective way to administer general anesthesia for fiberoptic bronchoscopy in children, which can ensure stable vital signs during examination, reduce intraoperative adverse reactions and postoperative agitation, shorten examination time, and increase amnesic effect.

DeepETPicker: Fast and accurate 3D particle picking for cryo-electron tomography using weakly supervised deep learning.

To solve three-dimensional structures of biological macromolecules in situ, large numbers of particles often need to be picked from cryo-electron tomograms. However, adoption of automated particle-picking methods remains limited because of their technical limitations. To overcome the limitations, we develop DeepETPicker, a deep learning model for fast and accurate picking of particles from cryo-electron tomograms. Training of DeepETPicker requires only weak supervision with low numbers of simplified labels, reducing the burden of manual annotation. The simplified labels combined with the customized and lightweight model architecture of DeepETPicker and accelerated pooling enable substantial performance improvement. When tested on simulated and real tomograms, DeepETPicker outperforms the competing state-of-the-art methods by achieving the highest overall accuracy and speed, which translate into higher authenticity and coordinates accuracy of picked particles and higher resolutions of final reconstruction maps. DeepETPicker is provided in open source with a user-friendly interface to support cryo-electron tomography in situ.

GeneCompass: deciphering universal gene regulatory mechanisms with a knowledge-informed cross-species foundation model.

Deciphering universal gene regulatory mechanisms in diverse organisms holds great potential for advancing our knowledge of fundamental life processes and facilitating clinical applications. However, the traditional research paradigm primarily focuses on individual model organisms and does not integrate various cell types across species. Recent breakthroughs in single-cell sequencing and deep learning techniques present an unprecedented opportunity to address this challenge. In this study, we built an extensive dataset of over 120 million human and mouse single-cell transcriptomes. After data preprocessing, we obtained 101,768,420 single-cell transcriptomes and developed a knowledge-informed cross-species foundation model, named GeneCompass. During pre-training, GeneCompass effectively integrated four types of prior biological knowledge to enhance our understanding of gene regulatory mechanisms in a self-supervised manner. By fine-tuning for multiple downstream tasks, GeneCompass outperformed state-of-the-art models in diverse applications for a single species and unlocked new realms of cross-species biological investigations. We also employed GeneCompass to search for key factors associated with cell fate transition and showed that the predicted candidate genes could successfully induce the differentiation of human embryonic stem cells into the gonadal fate. Overall, GeneCompass demonstrates the advantages of using artificial intelligence technology to decipher universal gene regulatory mechanisms and shows tremendous potential for accelerating the discovery of critical cell fate regulators and candidate drug targets.