Computational and Structural Biotechnology Journal

In the medical field, bone abnormality detection is a very important issue. Bone abnormalities include various diseases such as fractures, osteoporosis, bone tumors, and joint diseases. If these diseases are not diagnosed and treated in a timely manner, they can seriously affect the health and quality of life of patients. Artificial intelligence has made remarkable advances in Cluster analysis of medical big data, effectively mining its hidden associations to provide effective information for clinical diagnosis and medical research. However, the effectiveness of deep learning in domains with limited or no labeled data is often limited. To address this issue, we propose a novel and reliable two-stage unsupervised deep clustering framework for skeletal anomaly detection. This framework combines neural network parameters with feature clustering for collaborative learning to detect anomalies. We trained eight separate models, one for classification and seven for anomaly detection, using the MURA dataset, the largest publicly available skeletal imaging dataset. In the first stage, our approach achieved an average sensitivity and specificity of 99.76% and 99.53%, respectively. The second stage performed optimally with an average sensitivity and specificity of 83.28% and 97.56%, respectively. Our method can be easily implemented as software modules and used as a visualization tool for skeletal physicians, making it a promising approach for future development.

The establishment of a complex multi-scale model of biological tissue is of great significance for the study of related diseases, and the integration of relevant quantitative data is the premise to achieve this goal. Whereas, the systematic collation of data sets related to placental tissue is relatively lacking. In this study, 18 published transcriptomes (a total of 425 samples) datasets of human pregnancy-related tissues (including chorionic villus and decidua, term placenta, endometrium, in vitro cell lines, etc.) from public databases were collected and analyzed. We compared the most widely used dimensionality reduction (DR) methods to generate a 2D-map for visualization of these data. We also compared the effects of different parameter settings and commonly used manifold learning methods on the results. The result indicates that the nonlinear method can better preserve the small differences between different subtypes of placental tissue than linear method. It led the foundation for the study on accurate computational modeling of placental tissue development in the future. The datasets and analysis provide a useful source for the researchers in the field of the maternal-fetal interface and the establishment of pregnancy.

To perform a bibliometric visualization in lipidomics-related research with two decades. The primary data was retrieved from the Web of Science, three sotwares (VOSviewer, CiteSpace, and R) provided an overview of this field. The countries, institutions, authors, key terms, and keywords were tracked and corresponding mapping was generated. From January 1st in 2001 to March 21th in 2022, 45,325 authors from 234 organizations in 101 countries published 7,338 publications in 382 journals were found. Journal of Lipid Research was the most productive (284 publications) and highly cited journal (18,293 citations). We clustered four keywords themes. The niche theme were shotgun lipidomics, tandem mass-spectrometry, and electrospray-ionization. The motor theme were expression, diseases, and inflammation. The emerging or decling theme were identification, mass-spectrometry, and fatty acids.The basic theme were metabolism, cell, and plasma. Though eight categories the lipid were classified, the keywords showed two of which were got more attention for research, fatty acyls and glycerophospholipids. The top 3 lipidomics-favoured diseases were insulin resistance, obesity, and Alzheimers disease. The top 3 lipidomics-favoured tissue was plasma, brain, and adipose tissue. Burst citations show "women" and "pregnancy" with the strength of 8.91 and 7.1, both topics may be a potential hotspot in the future.

BACKGROUNDEndometriosis is a complex and polygenic disease characterized by the growth of endometrial-like tissue outside the uterus. Associated conditions include pelvic pain and infertility among others. Current treatments of endometriosis are not effective in many patients and are prone of side effects, being laparoscopy the main solution for them. However, laparoscopy is an invasive solution that can negatively affect ovarian reserve among others fertility consequences. Therefore, more effective treatments are needed for inhibiting the disease progression. Some approaches as systems pharmacology can offer new avenues to study complex diseases at a systematic level, allowing us the integration of information at different levels for prioritizing effective therapies. OBJECTIVETo use systems pharmacology and in vitro analyses to predict more effective endometriosis treatment among approved drugs on the market for further clinical trials. STUDY DESIGNData-driven discovery study integrating protein-protein interactions, transcriptomics, approved drugs and endometriosis-related gene targets to predict therapies that can be repositioned in endometriosis. An in-silico molecular drug efficacy screening was performed to compare proposed treatments with the current ones. The candidate drugs with the safest clinical profiles were prioritized for in-vitro validation in an endometriosis cell model. RESULTSOur endometriosis molecular model showed that the endometriosis progression and the infertility associated are molecularly highly related since they share 210 genes through direct protein-protein interactions from 553 infertility-related and 2,931 progression-related endometriosis genes, an overlap that is higher than randomly expected (p-value = 1.9 x 10-{superscript 1}{superscript 1}). In addition, these genes are also closer physically interacting in the network than random expectation (z-score = -1.75, p-value = 0.03). Both results highlighted the similar molecular basis of endometriosis progression and infertility associated to endometrium. Leveraging this endometriosis disease network, we have identified sixteen potential candidate drugs, which had a superior molecular efficacy treating the disease genes as they target more endometriosis-related genes (p-value = 0.01) than randomly expected compared with current endometriosis treatments (p-value = 0.97). The drug safety analysis deemed adenine, copper, zinc, NADH, glutathione, and resveratrol as the safest candidate drugs for endometriosis, with no or hardly any clinical side effects. Zinc, copper and resveratrol suppressed endometriosis-related phenotypes in 12Z cells, including cell proliferation, migration, and overexpression of endometriosis-related biomarkers (ERB, IL-6 and VEGF). CONCLUSIONThis approach highlights the nutraceutics zinc and copper for having superior molecular efficacy than current endometriosis treatments, no considerable side effects, and the ability to reduce endometriosis cell phenotypes. Long-term clinical trials will be needed to confirm women in vivo effectiveness and provide new treatment opportunities for refractory endometriosis. Condensation pageO_ST_ABSTweetable StatementC_ST_ABSA drug repurposing approach based on systems pharmacology and in vitro validation predict that zinc and copper nutraceuticals target 76 endometriosis-related genes, reverse endometriosis phenotypes, and exceed efficacy of current treatments AJOG at a GlanceO_ST_ABSA. Why was this study conducted?C_ST_ABSCurrent endometriosis treatments do not effectively address root causes of the disease and have side effects. Systems pharmacology-based models help uncover existing drugs that can be repurposed for complex diseases such as endometriosis. B. What are the key findings?A systems pharmacology-based model prioritized sixteen approved pharmacotherapies with superior molecular efficacy than current endometriosis therapies. Experimental validation of three of the safest prioritized compounds (resveratrol, copper, and zinc) showed a significant reduction of endometriotic cell phenotypes. C. What does this study add to what is already known?There are several opportunities to improve the standard of care for patients suffering endometriosis with existing drugs on the market. Multi-target pharmacotherapies are likely more effective and safer than current endometriosis treatments and surgical interventions.

BackgroundExisting proposed pathogenesis for preeclampsia (PE) was only applied for early-onset PE (EOPE). Our previous work identified the transcriptome to decipher EOPE and late-onset PE (LOPE), but the pathogenesis models were not validated. We developed and validated the pathogenesis models by hierarchical representation learning of interactomes connecting endometrial maturation, placentation, chorioamnionitis, and hemolysis, elevated liver enzyme, and low platelet (HELLP) syndrome. MethodsWe utilized 19 gene sets from our previous work to infer interactomes to develop (n=177) and validate (n=352) explainable artificial intelligence models for each PE subtype using deep-insight visible neural network and gradient-weighted class activation mapping. ResultsThe hierarchically learned representations identified novel genes for LOPE, similar to endometrial maturation (MRPL34, DYNLL1), chorioamnionitis (ANKRD13A, SLA), and HELLP syndrome (FAM43A). We also identified novel genes for EOPE, similar to endometrial maturation (SNAP23, PPL, LRRC32), placentation (GPT2, UBE2H, NIPAL3, NIN, KIAA0232, MT1F, DKK3, SLC24A3), and HELLP syndrome (SWAP70, GREM2, GPR146, PIP5K1B, EZR). Nonetheless, a gene for each subtype was frequently studied, i.e., IGF1 (chorioamnionitis) and PAPPA2 (placentation), including LOPE and EOPE samples. ConclusionsOur pathogenesis models connected both endometrial maturation and HELLP syndrome with LOPE and EOPE. However, they were differently connected with chorioamnionitis and placentation.

Large language models can help with clinical decision-making tasks. Complex oncology cases are best managed through multidisciplinary tumor boards but are difficult to do so due to their expense. The MDAT framework is proposed to mimic tumor board-style collaboration. Some LLMs are prompted to act like experts. They first analyze the prompt from their respective perspectives. Then the decision-making takes place to vote, agree, deal with discord and unify. We evaluated this framework on 3 LLMs: ChatGPT-4o, DeepSeek-R1 and Llama-4 (1,056 clinical questions; 182 cases). It focused on staging and selecting the treatment, managing the complications and following up. Fixed-size MDATs, across all baselines and configurations of MDAT, outperformed six different strong prompting baseline algorithms across all models. DeepSeek-R1 benefits from a five-agent MDAT that performs best (98.26/100). The fixed-size MDATs showed the greatest gains in higher-complexity questions, demonstrating robustness where accurate, multidisciplinary reasoning is most needed. As our MDT-inspired agent enhances LLM accuracy for oncology decision-making, it provides a pragmatic approach to integrating AI into complex cancer management. Author summaryExperts, such as surgeon, oncologist and radiologist get together to decide on the best treatment for a cancer patient but it is rather costly and time consuming. Large language models exhibit excellent performance when it comes to answering questions, particularly in medical contexts. We wanted to discover if large language models can work collaboratively in the same manner. We designed a framework named MDAT, standing for Multidisciplinary Agent Team. Rather than asking a question to one generic AI, our system builds a team of AIs "experts". Each agent considers a patients case from its perspective before all of them work together in a well-designed system to provide the expert recommendation. Through multidisciplinary team (MDT) gynecologic cancer scenarios, we examined this team-based artificial intelligence on a custom-built dataset of more than 1000 questions. According to our findings, this method always provides more reliable and accurate answers than decision-making methods, especially for the toughest cases. According to our findings, we can create safer and more effective ways of helping doctors make crucial decisions for cancer patients by making AI act like a real medical team.

BV (bacterial vaginosis) influences 20-40% of women but its etiology is still poorly understood. An open question about the BV is which of the hundreds of bacteria found in the human vaginal microbiome (HVM) are the causal pathogens of BV? Existing attempts to identify them have failed for at least two reasons: (i) a focus on species per se that ignores species interactions; (ii) a lack of systems-level understanding of the HVM. Here, we recast the question of microbial causality of BV by asking if there are any reliable signatures of community composition (or poly-microbial cults) associated with it? We apply a new framework (species dominance network analysis by Ma & Ellison (2019: Ecological Monographs) to detect critical structures in HVM networks associated with BV risks and etiology. We reanalyzed the metagenomic datasets of a mixed-cohort of 25 BV patients and healthy women. In these datasets, we detected 15 trio-motifs that occurred exclusively in BV patients. We failed to find any of these 15 trio-motifs in three additional cohorts of 1535 healthy women. Most member-species of the 15 trio motifs are BV-associated anaerobic bacteria (BVAB), Ravels community-state type indicators, or the most dominant species; virtually all species interactions in these trios are high-salience skeletons, suggesting that those trios are strongly connected cults. The presence of trio motifs unique to BV may act as indicators for its personalized diagnosis and could help elucidate a more mechanistic interpretation of its risks and etiology. Single sentence summary15 trio motifs unique to the BV (bacterial vaginosis) may act as indicators for personalized BV diagnosis, risk prediction and etiological study.

BackgroundNowadays, the internal fixation has been an effective way for calcaneal fractures treatment. However, high risk of infection was found after the internal fixation, and the mechnism remains unclear. ObjectiveIn this work, we systematically preformed a comparative proteomic analysis between necrotic tissues and normal soft tissues aiming to find the molecular changes of the tissue during the fixation. MethodThe necrotic tissues (NTs) samples (n = 3) and the normal soft tissues control (NC) samples (n = 3) which was 2 -3 cm away from the NT were collected after the surgery. ALC-MS/MS analysis. A label free quantitation stragy was used to compare the proteome alterations followed by detailed bioinformatic analysis. ResultsA total of 902 and 1286 protein groups were quantified in the NT group and the NC group separately, with 233 proteins upregulated and 484 proteins downregulated in the NT group. Those differently expressed proteins were highly correlated with the metabolic pathways, especially those downregulated proteins in the necrotic tissue indicating an inacitive cell states in the nearby of the plate internal fixation. In addition, the detailed functiona analysis showed that the the upregutated proteins in necrotic tissue were highly enriched in the disease-related functions. ConclusionThis alerted us to clean the wound in time and found a safer strategy for internal fixation. Altogether, the emerging understanding of the proteomic properties in the necrotic tissue will guide the development of new strategies for internal fixation of calcaneal fractures

In this study, we report first high concentrations of a ketone body, 3-hydroxybutyrate (3HB) in the amniotic fluid in humans. Although 3HB concentrations in the maternal blood are approximately 0.1, those in the amniotic fluid are approximately 0.6 mM. High placental 3HB production is potentially key for producing and maintaining high 3HB levels in the amniotic fluid. The rate-limiting enzyme, mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase 2 (HMGCS2), is highly expressed in the cells of the chorionic plate and responsible for 3HB production. Therefore, high HMGCS2 expression maintenance is supposed to be pivotal for maintaining the 3HB supply for the human fetus. Here, we propose that humans display two pathways, an amniotic fluid- and another umbilical vein-mediated, for supplying 3HB to the human fetus. These supply pathways are supposedly essential for human brain development during the late phase of pregnancy. Graphical abstractHuman fetuses are supported by 3HB from the amniotic fluid for their brain development. HIGHLIGTS3-Hydroxybutyrate concentrations are high in the amniotic fluid in the human uterus. The chorionic plate of the placenta highly expresses 3-hydroxy-3-methylglutaryl-CoA synthase 2. Human fetuses may be supplied with 3HB for brain development through the amniotic fluid.

The low success rate in in vitro fertilization (IVF) may be related to our inability to select embryos with good implantation potential by traditional morphology grading and remains a great challenge to clinical practice. Multiple deep learning-based methods have been introduced to improve embryo selection. However, existing methods only achieve limited prediction power and generally ignore the repeated embryo transfers from one stimulated IVF cycle. To improve the deep learning-based models, we introduce Embryo2live, which assesses the multifaceted qualities of embryos from static images taken under standard inverted microscope, primarily in vision transformer frameworks to integrate global features. We first demonstrated its superior performance in predicting Gardners blastocyst grades with up to 9% improvement from the best existing method. We further validated its high capability of supporting transfer learning using the large clinical dataset of the Centre. Remarkably, when applying Embryo2live to the clinical dataset for embryo prioritization, we found it improved the live birth rates of the Top 1 embryo in patients with multiple embryos available for transfer from 23.0% with conventional morphology grading to 71.3% using Embryo2live, reducing the average number of embryo transfers from 2.1 to 1.4 to attain a live birth.

Preeclampsia is a "placenta-derived disorder" characterized disease, which remains a major unaddressed public health problem. The differentiation and development of cytotrophoblasts and extravillous trophoblasts are two crucial processes, which are tightly regulated. And any abnormal regulation can lead to the occurrence of pregnancy-related diseases such as preeclampsia. In this article, we performed Spatial transcriptomics on tissues of PE and Normal, and found a large amount of extravillous trophoblasts (EVTs) were accumulated in the cluster 8 (cytotrophoblast shell) of PE_decidua, and Trajectory analysis revealed KRT8 was over-expressed in the cytotrophoblast shell of PE_decidua. The accumulation of EVTs caused by the increase of KRT8 promotes the development of preeclampsia.

ObjectiveOn the basis of retrospectively analysis the trans-cerebellar section showing the Sylvian fissure of normal fetus is better than trans-thalamic section in middle and late trimester, we prospectively studied the morphological changes of the Sylvian fissure of normal fetus on the trans-cerebellar section in order to provide valuable information for the diagnosis of fetal cerebral cortical dysplasia. MethodsA prospective cross-sectional study was conducted on 845 normal fetuses at 21 to 32 weeks of gestation from January 2019 to September 2020. The angle of the posterosuperior horn of the Sylvian fissure was measured. Based on the angle, the morphology of the Sylvian fissure was divided into four shapes included [Formula] shape (Angle >90{degrees}), [Pcy] shape (Angle {approx}90{degrees}), {pi} shape (Angle <90{degrees}) and shape (Angle {approx}0{degrees}). ResultsThe angle of the posterosuperior horn of the Sylvian fissure was negatively correlated with gestational age which correlation coefficient was 0.966 (P < 0.001). Taking gestational age as the independent variable and the angle of posterosuperior horn of the Sylvian fissure as the dependent variable, it showed that there was a linear relationship between the gestational age and the angle of posterosuperior horn of the Sylvian fissure. We got a simple correlation formula that the angle of posterosuperior horn of the Sylvian fissure ({degrees}) =140-13x(gestational weeks-21). It was found that the morphological changes of the Sylvian fissure were related to the gestational age. The morphology of the Sylvian fissure was [Formula]-shaped at 21 to 24 weeks of gestation, [Pcy]-shaped at 25 to 26 weeks of gestation, {pi}-shaped at 26 to 30 weeks of gestation, and the Sylvian fissure was almost closed and appeared shape after 31 to 32 weeks of gestation. ConclusionThis study preliminarily elucidates that the morphological changes of the posterosuperior horn of the Sylvian fissure of normal fetuses at 21 to 32 weeks of gestation through the trans-cerebellar section, which could provide valuable information for the evaluation of the normal development of the Sylvian fissure and the prenatal diagnosis of cerebral cortical hypoplasia of fetuses.

PurposeLow back pain is the worlds leading cause of disability and pathology of the lumbar intervertebral discs is frequently considered a driver of pain. The geometric characteristics of intervertebral discs offer valuable insights into their mechanical behavior and pathological conditions. In this study, we present a convolutional neural network (CNN) autoencoder to extract latent features from segmented disc MRI. Additionally, we interpret these latent features and demonstrate their utility in identifying disc pathology, providing a complementary perspective to standard geometric measures. MethodsWe examined 195 sagittal T1-weighted MRI of the lumbar spine from a publicly available multi-institutional dataset. The proposed pipeline includes five main steps: 1) segmenting MRI, 2) training the CNN autoencoder and extracting latent geometric features, 3) measuring standard geometric features, 4) predicting disc narrowing with latent and/or standard geometric features and 5) determining the relationship between latent and standard geometric features. ResultsOur segmentation model achieved an IoU of 0.82 (95% CI: 0.80-0.84) and DSC of 0.90 (95% CI: 0.89-0.91). The minimum bottleneck size for which the CNN autoencoder converged was 4x1 after 350 epochs (IoU of 0.9984 - 95% CI: 0.9979-0.9989). Combining latent and geometric features improved predictions of disc narrowing compared to using either feature set alone. Latent geometric features encoded for disc shape and angular orientation. ConclusionsThis study presents a CNN-autoencoder to extract latent features from segmented lumbar disc MRI, enhancing disc narrowing prediction and feature interpretability. Future work will integrate disc voxel intensity to analyze composition.

BackgroundInflammation is currently recognized as one of the major causes of premature delivery. As a member of the IL-1{beta} family, interleukin-33 (IL-33) has been shown to be involved in a variety of pregnancy-related diseases. This study aims to investigate the potential function of IL-33 in uterine smooth muscle cells during labor. MethodsSamples of myometrium from term pregnant ([&ge;]37 weeks gestation) women were frozen or cells were isolated and cultured. Immunohistochemistry and western blotting were used to identify the distribution of IL-33. Cultured cells were incubated with LPS to mimic inflammation as well as 48C to block endoplasmic reticulum (ER) stress and BAPTA-AM, a calcium chelator. ResultsSimilar with onset of labor, LPS could reduce the expression of nuclear IL-33 in a time-limited manner and induced ER stress. Meanwhile, Knockdown of IL-33 increased LPS-induced calcium concentration, ER stress and phosphorylation of NF-{kappa}B and p38. In addition, siRNA IL-33 further simulates LPS enhanced COX-2 expression via NF-{kappa}B and p38 pathways. IL-33 expression was decreased in the nucleus with the onset of labor. LPS induced ER stress and increased expression of the labor-associated gene, COX-2, as well as IL-6 and IL-8 in cultured myometrial cells. IL-33 also increased COX-2 expression. However, after IL-33 was knockdown, the stimulating effect of LPS on calcium was enhanced. 48C also inhibited the expression of COX-2 markedly. The expression of calcium channels on the membrane and intracellular free calcium ion were both increased accompanied by phosphorylated NF-{kappa}B and p38. ConclusionsThese data suggest that IL-33 may be involved in initiation of labor by leading to stress of the ER via an influx of calcium ions in human uterine smooth muscle cells. FundingThis study was supported by grants from the National Natural Science Foundation of China (Nos. 81300507).

Measuring blood pressure during pregnancy is an essential component of antenatal care, and is critical for detecting adverse conditions such as pre-eclampsia. The standard approach for measuring blood pressure is via manual auscultation by a trained expert or via an oscillometric self-inflating cuff. While both methods can provide reasonably accurate blood pressure measurements when used correctly, non-expert use can lead to significant error. Moreover, such techniques are uncomfortable and can cause bruising, pain and consequential resistance to use / low compliance. In this work, we propose a low-cost onedimensional Doppler-based method for the detection of hypertension in pregnancy. Using a sample of 653 pregnant women of Mayan descent in highland Guatemala, we recorded up to 10 minutes of 1D Doppler data of the fetus, and blood pressure from both arms using an Omron M7 oscillometric cuff. A hierarchical LSTM network with attention mechanism was trained to classify hypertension in pregnancy, producing an area under the receiveroperator curve of 0.94. A projection of the data into lower dimensions indicates hypertensive cases are located at the periphery of the distribution of the output of the distribution. This work presents the first demonstration that blood pressure can be measured using Doppler (without occlusion) and may lead to a novel class of blood pressure monitors which allow rapid blood pressure estimation from multiple body locations. Moreover, the association of the predictor with the fetal blood flow indicates that hypertension in the mother has a significant effect on the fetal blood flow.

BackgroundIt has been reported that loss of PCBP2 led to increased reactive oxygen species (ROS) production and accelerated cell aging. Knockdown of PCBP2 in HCT116 cells leads to significant down-regulation of fibroblast growth factor 2 (FGF2). Here, we tried to elucidate the intrinsic factors and potential mechanisms of BMSCs aging from the interactions among PCBP2, ROS and FGF2. MethodsUnlabeled quantitative proteomics were performed to show differentially expressed proteins in the replicative senescent human-derived bone marrow stromal cells (RS-hBMSCs). ROS and FGF2 were detected in the loss-and-gain cell function experiments of PCBP2. The function recovery experiments were performed to verify whether PCBP2 regulates cell function through ROS/FGF2-dependent ways. ResultsPCBP2 expression was significantly lower in P10-hBMSCs. Knocking down the expression of PCBP2 inhibited the proliferation while accentuated the apoptosis and cell arrest of RS-hBMSCs. PCBP2 silence could increase the production of ROS. On the contrary, overexpression of PCBP2 increased the viability of both P3-hBMSCs and P10-hBMSCs significantly. Meanwhile, over-expression of PCBP2 led to significantly reduced expression of FGF2. Overexpression of FGF2 significantly offset the effect of PCBP2 overexpression in P10-hBMSCs, leading to decreased cell proliferation, increased apoptosis and reduced G0/G1 phase ratio of the cells. ConclusionThis study initially elucidates that PCBP2 as an intrinsic aging factor regulates the replicative senescence of hBMSCs through the ROS-FGF2 signaling axis.

The prevalence of knee osteoarthritis (OA) is widespread and the heterogeneous patient factors and clinical symptoms in OA patients impede developing personalized treatments for OA patients. In this study, we used unsupervised and supervised machine learning to organize the heterogeneity in knee OA patients and predict disease progression in individuals from the Osteoarthritis Initiative (OAI) dataset. We identified four distinct knee OA phenotypes using unsupervised learning that were defined by nutrition, disability, stiffness, and pain (knee and back) and were strongly related to disease fate. Interestingly, the absence of supplemental vitamins from an individuals diet was protective from disease progression. Moreover, we established a phenotyping tool and prognostic model from 5 variables (WOMAC disability score of the right knee, WOMAC total score of the right knee, WOMAC total score of the left knee, supplemental vitamins and minerals frequency, and antioxidant combination multivitamins frequency) that can be utilized in clinical practice to determine the risk of knee OA progression in individual patients. We also developed a prognostic model to estimate the risk for total knee replacement and provide suggestions for modifiable variables to improve long-term knee health. This combination of unsupervised and supervised data-driven tools provides a framework to identify knee OA phenotype in a clinical scenario and personalize treatment strategies.

ObjectivesThe study aimed to conduct a bibliometric analysis of publications concerning lumbar spondylolisthesis, as well as explore its research topics and trends with machine-learning based text mining. MethodsThe data were extracted from the Web of Science Core Collection (WoSCC) database and analyzed in Rstudio1.3.1. Annual publication production and the top 20 productive authors over time were presented. Additionally, top 20 productive journals and top 20 impact journals were compared by spine-subspecialty or not. Similarly, top 20 productive countries/regions and top 20 impact countries/regions were compared by developed countries/regions or not. The collaborative relationship among countries and the research trends in the past decade were presented by R package "Bibliometrix". Latent Dirichlet allocation (LDA) analysis was conducted to classify main topics of lumbar spondylolisthesis. ResultUp to 2021, a total number of 4990 articles concerning lumbar spondylolisthesis were finally included for analysis. Spine-subspecialty journals were found to be dominant in productivity and impact of the field, and SPINE, EUROPEAN SPINE JOURNAL and JOURNAL OF NEUROSURGERY-SPINE were the top 3 productive and the top 3 impact journals in this field. USA, China and Japan have contributed to over half of the publication productivity, but European countries seemed to publish more influential articles. It seemed that developed countries/regions tended to produce more articles as well as high influential articles, and international collaborations focused on USA, Europe and eastern Asia. Publications concerning emerging surgical technique was the major topic, followed by radiographic measurement and epidemiology for this field. Cortical bone trajectory, oblique lumbar interbody fusion, oblique lateral lumbar interbody fusion, lateral lumbar interbody fusion, degenerative lumbar spondylolisthesis, adjacent segment disease, spinal canal stenosis, minimally invasive transforaminal lumbar interbody fusion and percutaneous pedicle screw were the recent hotspots over the past 5 years. ConclusionsThe study successfully summarized the productivity and impact of different countries/regions and journals, which should benefit the journal selection and pursuit of international collaboration for researcher who were interested in the field of lumbar spondylolisthesis. Additionally, the current study may encourage more researchers in the field and somewhat inform their research direction in the future.

PurposeThe ability to detect, monitor, and precisely time the morphokinetic stages of human pre -implantation embryo development plays a critical role in assessing their viability and potential for successful implantation. Therefore, there is a need for accurate and accessible tools to analyse embryos. This work describes a highly accurate, machine learning model designed to predict 17 morphokinetic stages of pre-implantation human development, an improvement on existing models. This model provides a robust tool for researchers and clinicians, enabling the automation of morphokinetic stage prediction, standardising the process, and reducing subjectivity between clinics. MethodA computer vision model was built on a publicly available dataset for embryo Morphokinetic stage detection. The dataset contained 273,438 labelled images based on Embryoscope/+(C) embryo images. The dataset was split 70/10/20 into training/validation/test sets. Two different deep learning architectures were trained and tested, one using EfficientNet-V2-Large and the other using EfficientNet-V2-Large with the addition of fertilisation time as input. A new postprocessing algorithm was developed to reduce noise in the predictions of the deep learning model and detect the exact time of each morphokinetic stage change. ResultsThe proposed model reached an overall test accuracy of 87% across 17 morphokinetic stages on an independent test set. ConclusionThe proposed model shows a 17% accuracy improvement, compared to the best models on the same dataset. Therefore, our model can accurately detect morphokinetic stages in static embryo images as well as detecting the exact timings of stage changes in a complete time-lapse video.

Currently, the number of patients with COVID-19 has significantly increased. Thus, there is an urgent need for developing treatments for COVID-19. Drug repurposing, which is the process of reusing already-approved drugs for new medical conditions, can be a good way to solve this problem quickly and broadly. Many clinical trials for COVID-19 patients using treatments for other diseases have already been in place or will be performed at clinical sites in the near future. Additionally, patients with comorbidities such as diabetes mellitus, obesity, liver cirrhosis, kidney diseases, hypertension, and asthma are at higher risk for severe illness from COVID-19. Thus, the relationship of comorbidity disease with COVID-19 may help to find repurposable drugs. To reduce trial and error in finding treatments for COVID-19, we propose building a network-based drug repurposing framework to prioritize repurposable drugs. First, we utilized knowledge of COVID-19 to construct a disease-gene-drug network (DGDr-Net) representing a COVID-19-centric interactome with components for diseases, genes, and drugs. DGDr-Net consisted of 592 diseases, 26,681 human genes and 2,173 drugs, and medical information for 18 common comorbidities. The DGDr-Net recommended candidate repurposable drugs for COVID-19 through network reinforcement driven scoring algorithms. The scoring algorithms determined the priority of recommendations by utilizing graph-based semi-supervised learning. From the predicted scores, we recommended 30 drugs, including dexamethasone, resveratrol, methotrexate, indomethacin, quercetin, etc., as repurposable drugs for COVID-19, and the results were verified with drugs that have been under clinical trials. The list of drugs via a data-driven computational approach could help reduce trial-and-error in finding treatment for COVID-19.