Children's Hospital of Zhejiang University

Advances in high-throughput sequencing (HTS) have fostered rapid developments in the field of microbiome research, and massive microbiome datasets are now being generated. However, the diversity of software tools and the complexity of analysis pipelines make it difficult to access this field. Here, we systematically summarize the advantages and limitations of microbiome methods. Then, we recommend specific pipelines for amplicon and metagenomic analyses, and describe commonly-used software and databases, to help researchers select the appropriate tools. Furthermore, we introduce statistical and visualization methods suitable for microbiome analysis, including alpha- and beta-diversity, taxonomic composition, difference comparisons, correlation, networks, machine learning, evolution, source tracing, and common visualization styles to help researchers make informed choices. Finally, a step-by-step reproducible analysis guide is introduced. We hope this review will allow researchers to carry out data analysis more effectively and to quickly select the appropriate tools in order to efficiently mine the biological significance behind the data.

Recently, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron variant (B.1.1.529) was first identified in Botswana in November 2021. It was first reported to the World Health Organization (WHO) on November 24. On November 26, 2021, according to the advice of scientists who are part of the WHO's Technical Advisory Group on SARS-CoV-2 Virus Evolution (TAG-VE), the WHO defined the strain as a variant of concern (VOC) and named it Omicron. Compared to the other four VOCs (Alpha, Beta, Gamma, and Delta), the Omicron variant was the most highly mutated strain, with 50 mutations accumulated throughout the genome. The Omicron variant contains at least 32 mutations in the spike protein, which was twice as many as the Delta variant. Studies have shown that carrying many mutations can increase infectivity and immune escape of the Omicron variant compared with the early wild-type strain and the other four VOCs. The Omicron variant is becoming the dominant strain in many countries worldwide and brings new challenges to preventing and controlling coronavirus disease 2019 (COVID-19). The current review article aims to analyze and summarize information data about the biological characteristics of amino acid mutations, the epidemic characteristics, immune escape, and vaccine reactivity of the Omicron variant, hoping to provide a scientific reference for monitoring, prevention, and vaccine development strategies for the Omicron variant.

BACKGROUND: Rhesus (Rh) disease and extreme hyperbilirubinemia (EHB) result in neonatal mortality and long-term neurodevelopmental impairment, yet there are no estimates of their burden. METHODS: Systematic reviews and meta-analyses were undertaken of national prevalence, mortality, and kernicterus due to Rh disease and EHB. We applied a compartmental model to estimate neonatal survivors and impairment cases for 2010. RESULTS: Twenty-four million (18% of 134 million live births ≥ 32 wk gestational age from 184 countries; uncertainty range: 23-26 million) were at risk for neonatal hyperbilirubinemia-related adverse outcomes. Of these, 480,700 (0.36%) had either Rh disease (373,300; uncertainty range: 271,800-477,500) or developed EHB from other causes (107,400; uncertainty range: 57,000-131,000), with a 24% risk for death (114,100; uncertainty range: 59,700-172,000), 13% for kernicterus (75,400), and 11% for stillbirths. Three-quarters of mortality occurred in sub-Saharan Africa and South Asia. Kernicterus with Rh disease ranged from 38, 28, 28, and 25/100,000 live births for Eastern Europe/Central Asian, sub-Saharan African, South Asian, and Latin American regions, respectively. More than 83% of survivors with kernicterus had one or more impairments. CONCLUSION: Failure to prevent Rh sensitization and manage neonatal hyperbilirubinemia results in 114,100 avoidable neonatal deaths and many children grow up with disabilities. Proven solutions remain underused, especially in low-income countries.

BACKGROUND: Recent studies have demonstrated that mesenchymal stem cells (MSCs) modulate the immune response and reduce lung injury in animal models. Currently, no clinical studies of the effects of MSCs in acute respiratory distress syndrome (ARDS) exist. The objectives of this study were first to examine the possible adverse events after systemic administration of allogeneic adipose-derived MSCs in ARDS patients and second to determine potential efficacy of MSCs on ARDS. METHODS: Twelve adult patients meeting the Berlin definition of acute respiratory distress syndrome with a PaO2/FiO2 ratio of < 200 were randomized to receive allogeneic adipose-derived MSCs or placebo in a 1:1 fashion. Patients received one intravenous dose of 1 × 106 cells/kg of body weight or saline. Possible side effects were monitored after treatment. Acute lung injury biomarkers, including IL-6, IL-8 and surfactant protein D (SP-D), were examined to determine the effects of MSCs on lung injury and inflammation. RESULTS: There were no infusion toxicities or serious adverse events related to MSCs administration and there were no significant differences in the overall number of adverse events between the two groups. Length of hospital stay, ventilator-free days and ICU-free days at day 28 after treatment were similar. There were no changes in biomarkers examined in the placebo group. In the MSCs group, serum SP-D levels at day 5 were significantly lower than those at day 0 (p = 0.027) while the changes in IL-8 levels were not significant. The IL-6 levels at day 5 showed a trend towards lower levels as compared with day 0, but this trend was not statistically significant (p = 0.06). CONCLUSIONS: Administration of allogeneic adipose-derived MSCs appears to be safe and feasible in the treatment of ARDS. However, the clinical effect with the doses of MSCs used is weak, and further optimization of this strategy will probably be required to reach the goal of reduced alveolar epithelial injury in ARDS. TRIAL REGISTRATION: Clinical trials.gov, NCT01902082.

Autophagy refers to a lysosomal degradative pathway or a process of self-cannibalization. This pathway maintains nutrients levels for vital cellular functions during periods of starvation and it provides cells with survival advantages under various stress situations. However, the mechanisms responsible for the induction and regulation of autophagy are poorly understood. The c-Jun NH2-terminal kinase (JNK) signal transduction pathway functions to induce defence mechanisms that protect organisms against acute oxidative and xenobiotic insults. This pathway has also been repeatedly linked to the molecular events involved in autophagy regulation. The present review will focus on recent advances in understanding of the relationship between mitogen-activated protein kinase (MAPK)/JNK signalling and autophagic cell death.

BACKGROUND: Systemic lupus erythematosis (SLE) is a complex and clinically heterogeneous autoimmune disease. A variety of immunological defects contribute to SLE, including dysregulated innate and adaptive immune response. A clearer understanding of the mechanisms driving disease pathogenesis combined with recent advances in medical science is predicted to enable accelerated progress towards improved SLE-personalized approaches to treatment. The aim of this review was to clarify the immunological pathogenesis and treatment of SLE. DATA SOURCES: Literature reviews and original research articles were collected from database, including PubMed and Wanfang. Relevant articles about SLE were included. RESULTS: Breakdown of self-tolerance is the main pathogenesis of SLE. The innate and adaptive immune networks are interlinked with each other through cytokines, complements, immune complexes and kinases of the intracellular machinery. Treatments targeted at possible targets of immunity have been assessed in clinical trials. Most of them did not show better safety and efficacy than traditional treatments. However, novel targeting treatments are still being explored. CONCLUSIONS: Dysregulated immune response plays a critical role in SLE, including innate immunity and adaptive immunity. Biologic agents that aim to specifically target abnormal immune processes were assessing and may bring new hope to SLE patients.

Importance: Early diagnosis of retinoblastoma, the most common intraocular cancer, can save both a child's life and vision. However, anecdotal evidence suggests that many children across the world are diagnosed late. To our knowledge, the clinical presentation of retinoblastoma has never been assessed on a global scale. Objectives: To report the retinoblastoma stage at diagnosis in patients across the world during a single year, to investigate associations between clinical variables and national income level, and to investigate risk factors for advanced disease at diagnosis. Design, Setting, and Participants: A total of 278 retinoblastoma treatment centers were recruited from June 2017 through December 2018 to participate in a cross-sectional analysis of treatment-naive patients with retinoblastoma who were diagnosed in 2017. Main Outcomes and Measures: Age at presentation, proportion of familial history of retinoblastoma, and tumor stage and metastasis. Results: The cohort included 4351 new patients from 153 countries; the median age at diagnosis was 30.5 (interquartile range, 18.3-45.9) months, and 1976 patients (45.4%) were female. Most patients (n = 3685 [84.7%]) were from low- and middle-income countries (LMICs). Globally, the most common indication for referral was leukocoria (n = 2638 [62.8%]), followed by strabismus (n = 429 [10.2%]) and proptosis (n = 309 [7.4%]). Patients from high-income countries (HICs) were diagnosed at a median age of 14.1 months, with 656 of 666 (98.5%) patients having intraocular retinoblastoma and 2 (0.3%) having metastasis. Patients from low-income countries were diagnosed at a median age of 30.5 months, with 256 of 521 (49.1%) having extraocular retinoblastoma and 94 of 498 (18.9%) having metastasis. Lower national income level was associated with older presentation age, higher proportion of locally advanced disease and distant metastasis, and smaller proportion of familial history of retinoblastoma. Advanced disease at diagnosis was more common in LMICs even after adjusting for age (odds ratio for low-income countries vs upper-middle-income countries and HICs, 17.92 [95% CI, 12.94-24.80], and for lower-middle-income countries vs upper-middle-income countries and HICs, 5.74 [95% CI, 4.30-7.68]). Conclusions and Relevance: This study is estimated to have included more than half of all new retinoblastoma cases worldwide in 2017. Children from LMICs, where the main global retinoblastoma burden lies, presented at an older age with more advanced disease and demonstrated a smaller proportion of familial history of retinoblastoma, likely because many do not reach a childbearing age. Given that retinoblastoma is curable, these data are concerning and mandate intervention at national and international levels. Further studies are needed to investigate factors, other than age at presentation, that may be associated with advanced disease in LMICs.

Mesenchymal stem cells (MSCs) are progenitors of connective tissues, which have emerged as important tools for tissue engineering due to their differentiation potential along various cell types. In recent years, accumulating evidence has suggested that the regulation of mitochondria dynamics and function is essential for successful differentiation of MSCs. In this paper, we review and provide an integrated view on the role of mitochondria in MSC differentiation. The mitochondria are maintained at a relatively low activity level in MSCs, and upon induction, mtDNA copy number, protein levels of respiratory enzymes, the oxygen consumption rate, mRNA levels of mitochondrial biogenesis-associated genes, and intracellular ATP content are increased. The regulated level of mitochondrial ROS is found not only to influence differentiation but also to contribute to the direction determination of differentiation. Understanding the roles of mitochondrial dynamics during MSC differentiation will facilitate the optimization of differentiation protocols by adjusting biochemical properties, such as energy production or the redox status of stem cells, and ultimately, benefit the development of new pharmacologic strategies in regenerative medicine.

Since December 2019, there has been an outbreak of novel coronavirus (2019-nCoV) infection in China. Two cases of neonates with positive 2019-nCoV tests have been reported. Due to the immature immune system and the possibility of vertical transmission from mother to infant, neonates have become a high-risk group susceptible to 2019-nCoV, which emphasize a close cooperation from both perinatal and neonatal pediatrics. In neonatal intensive care unit (NICU), to prevent and control infection, there should be practical measures to ensure the optimal management of children potentially to be infected. According to the latest 2019-nCoV national management plan and the actual situation, the Chinese Neonatal 2019-nCoV expert working Group has put forward measures on the prevention and control of neonatal 2019-nCoV infection.

Fat mass and obesity-associated gene (FTO) is a member of the Fe (II)- and oxoglutarate-dependent AlkB dioxygenase family and is linked to both obesity and intellectual disability. The role of FTO in neurodevelopment and neurogenesis, however, remains largely unknown. Here we show that FTO is expressed in adult neural stem cells and neurons and displays dynamic expression during postnatal neurodevelopment. The loss of FTO leads to decreased brain size and body weight. We find that FTO deficiency could reduce the proliferation and neuronal differentiation of adult neural stem cells in vivo, which leads to impaired learning and memory. Given the role of FTO as a demethylase of N6-methyladenosine (m6A), we went on to perform genome-wide m6A profiling and observed dynamic m6A modification during postnatal neurodevelopment. The loss of FTO led to the altered expression of several key components of the brain derived neurotrophic factor pathway that were marked by m6A. These results together suggest FTO plays important roles in neurogenesis, as well as in learning and memory.

Many cancer cells display enhanced glycolysis and suppressed mitochondrial metabolism. This phenomenon, known as the Warburg effect, is critical for tumor development. However, how cancer cells coordinate glucose metabolism through glycolysis and the mitochondrial tricarboxylic acid (TCA) cycle is largely unknown. We demonstrate here that phosphoglycerate kinase 1 (PGK1), the first ATP-producing enzyme in glycolysis, is reversibly and dynamically modified with O-linked N-acetylglucosamine (O-GlcNAc) at threonine 255 (T255). O-GlcNAcylation activates PGK1 activity to enhance lactate production, and simultaneously induces PGK1 translocation into mitochondria. Inside mitochondria, PGK1 acts as a kinase to inhibit pyruvate dehydrogenase (PDH) complex to reduce oxidative phosphorylation. Blocking T255 O-GlcNAcylation of PGK1 decreases colon cancer cell proliferation, suppresses glycolysis, enhances the TCA cycle, and inhibits tumor growth in xenograft models. Furthermore, PGK1 O-GlcNAcylation levels are elevated in human colon cancers. This study highlights O-GlcNAcylation as an important signal for coordinating glycolysis and the TCA cycle to promote tumorigenesis.

BACKGROUND: Dyslipidemia in pregnancy are associated with gestational diabetes mellitus (GDM), preeclampsia, preterm birth and other adverse outcomes, which has been extensively studied in western countries. However, similar studies have rarely been conducted in Asian countries. Our study was aimed at investigating the associations between maternal dyslipidemia and adverse pregnancy outcomes among Chinese population. METHODS: Data were derived from 934 pairs of non-diabetic mothers and neonates between 2010 and 2011. Serum blood samples were assayed for fasting total cholesterol (TC), triglycerides (TG), high-density lipoprotein-cholesterol (HDL-C), and low-density lipoprotein-cholesterol (LDL-C) concentrations during the first, second and third trimesters. The present study explored the associations between maternal lipid profile and pregnancy complications and perinatal outcomes. The pregnancy complications included GDM, preeclampsia and intrahepatic cholestasis of pregnancy (ICP); the perinatal outcomes included preterm birth, small/large for gestational age (SGA/LGA) infants and macrosomia. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated and adjusted via stepwise logistic regression analysis. Optimal cut-off points were determined by ROC curve analysis. RESULTS: After adjustments for confounders, every unit elevation in third-trimester TG concentration was associated with increased risk for GDM (OR = 1.37, 95% CI: 1.18-1.58), preeclampsia (OR = 1.50, 95% CI: 1.16-1.93), ICP (OR = 1.28, 95% CI: 1.09-1.51), LGA (OR = 1.13, 95% CI: 1.02-1.26), macrosomia (OR = 1.19, 95% CI: 1.02-1.39) and decreased risk for SGA (OR = 0.63, 95% CI: 0.40-0.99); every unit increase in HDL-C concentration was associated with decreased risk for GDM and macrosomia, especially during the second trimester (GDM: OR = 0.10, 95% CI: 0.03-0.31; macrosomia: OR = 0.25, 95% CI: 0.09-0.73). The optimal cut-off points for third-trimester TG predicting GDM, preeclampsia, ICP, LGA and SGA were separately ≥ 3.871, 3.528, 3.177, 3.534 and ≤ 2.530 mmol/L. The optimal cut-off points for third-trimester HDL-C identifying GDM, macrosomia and SGA were respectively ≤ 1.712, 1.817 and ≥ 2.238 mmol/L. CONCLUSIONS: Among Chinese population, maternal high TG in late pregnancy was independently associated with increased risk of GDM, preeclampsia, ICP, LGA, macrosomia and decreased risk of SGA. Relative low maternal HDL-C during pregnancy was significantly associated with increased risk of GDM and macrosomia; whereas relative high HDL-C was a protective factor for both of them.

Autism spectrum disorder (ASD) is characterized by stereotyped behavior and deficits in communication and social interactions. Gastrointestinal (GI) dysfunction is an ASDassociated comorbidity, implying a potential role of the gut microbiota in ASD GI pathophysiology. Several recent studies found that autistic individuals harbor an altered bacterial gut microbiota. In some cases, remodeling the gut microbiota by antibiotic administration and microbiota transfer therapy reportedly alleviated the symptoms of ASD. However, there is little consensus on specific bacterial species that are similarly altered across individual studies. The aim of this study is to summarize previously published data and analyze the alteration of the relative abundance of bacterial genera in the gut microbiota in controls and individuals with ASD using meta-analysis. We analyzed nine studies, including 254 patients with ASD, and found that children with ASD had lower percentages of Akkermansia, Bacteroides, Bifidobacterium, and Parabacteroides and a higher percentage of Faecalibacterium in the total detected microflora compared to controls. In contrast, children with ASD had lower abundance of Enterococcus, Escherichia coli, Bacteroides, and Bifidobacterium and higher abundance of Lactobacillus. This metaanalysis suggests an association between ASD and alteration of microbiota composition and warrants additional prospective cohort studies to evaluate the association of bacterial changes with ASD symptoms, which would provide further evidence for the precise microbiological treatment of ASD.

Despite recent progress in our understanding of the association between the gut microbiome and colorectal cancer (CRC), multi-kingdom gut microbiome dysbiosis in CRC across cohorts is unexplored. We investigated four-kingdom microbiota alterations using CRC metagenomic datasets of 1,368 samples from 8 distinct geographical cohorts. Integrated analysis identified 20 archaeal, 27 bacterial, 20 fungal and 21 viral species for each single-kingdom diagnostic model. However, our data revealed superior diagnostic accuracy for models constructed with multi-kingdom markers, in particular the addition of fungal species. Specifically, 16 multi-kingdom markers including 11 bacterial, 4 fungal and 1 archaeal feature, achieved good performance in diagnosing patients with CRC (area under the receiver operating characteristic curve (AUROC) = 0.83) and maintained accuracy across 3 independent cohorts. Coabundance analysis of the ecological network revealed associations between bacterial and fungal species, such as Talaromyces islandicus and Clostridium saccharobutylicum. Using metagenome shotgun sequencing data, the predictive power of the microbial functional potential was explored and elevated D-amino acid metabolism and butanoate metabolism were observed in CRC. Interestingly, the diagnostic model based on functional EggNOG genes achieved high accuracy (AUROC = 0.86). Collectively, our findings uncovered CRC-associated microbiota common across cohorts and demonstrate the applicability of multi-kingdom and functional markers as CRC diagnostic tools and, potentially, as therapeutic targets for the treatment of CRC.

-assisted, template-free synthetic strategy, we developed hollow Prussian blue nanozymes (HPBZs) with multienzyme activity to scavenge RONS in a rat model of ischemic stroke. The comprehensive characteristics of HPBZs against RONS were explored. Apart from attenuating oxidative stress, HPBZs also suppressed apoptosis and counteracted inflammation both in vitro and in vivo, thereby contributing to increased brain tolerance of ischemic injury with minimal side effects. This study provides a proof of concept for a novel class of neuroprotective nanoagents that might be beneficial for treatment of ischemic stroke and other RONS-related disorders.

Mesenchymal stem cells (MSCs) are adult stromal cells with the capacity to differentiate into multiple types of cells. MSCs represent an attractive option in regenerative medicine due to their multifaceted abilities for tissue repair, immunosuppression, and anti-inflammation. Recent studies demonstrate that MSCs exert their effects via paracrine activity, which is at least partially mediated by extracellular vesicles (EVs). MSC-derived EVs (MSC-EVs) could mimic the function of parental MSCs by transferring their components such as DNA, proteins/peptides, mRNA, microRNA (miRNA), lipids, and organelles to recipient cells. In this review, we aim to summarize the mechanism and role of miRNA transfer in mediating the effects of MSC-EVs in the models of human diseases. The first three sections of the review discuss the sorting of miRNAs into EVs, uptake of EVs by target cells, and functional transfer of miRNAs via EVs. Then, we describe the composition of miRNAs in MSC-EVs. Next, we provide the existing evidence that MSC-EVs affect the outcomes of renal, liver, heart, and brain diseases by transferring their miRNA contents. In conclusion, EV-mediated miRNA transfer plays an important role in disease-modulating capacity of MSCs.

N 6 -methyladenosine (m 6 A) is the most abundant internal modification on mammalian messenger RNA. It is installed by a writer complex and can be reversed by erasers such as the fat mass and obesity-associated protein FTO. Despite extensive research, the primary physiological substrates of FTO in mammalian tissues and development remain elusive. Here, we show that FTO mediates m 6 A demethylation of long-interspersed element-1 (LINE1) RNA in mouse embryonic stem cells (mESCs), regulating LINE1 RNA abundance and the local chromatin state, which in turn modulates the transcription of LINE1-containing genes. FTO-mediated LINE1 RNA m 6 A demethylation also plays regulatory roles in shaping chromatin state and gene expression during mouse oocyte and embryonic development. Our results suggest broad effects of LINE1 RNA m 6 A demethylation by FTO in mammals.

Externally controlling the excitation of a neuronal subset through ion channels activation can modulate the firing pattern of an entire neural circuit in vivo. As nanovalves in the cell membrane, ion channels can be opened by light (optogenetics) or ultrasonic (sonogenetics) means. A thoroughly analyzed force sensor is the Escherichia coli mechano sensitive channel of large conductance (MscL). Here we expressed MscL in rat hippocampal neurons in a primary culture and showed that it could be activated by low-pressure ultrasound pulses. The gain-of-function mutation, I92L, sensitized MscL's sonic response, triggering action potentials at a peak negative pressure as low as 0.25 MPa. Further, the I92L MscL reliably elicited individual spikes by timed brief pulses, making excitation programmable. Because MscL opens to tension in the lipid bilayer, requiring no other proteins or ligands, it could be developed into a general noninvasive sonogenetic tool to manipulate the activities of neurons or other cells and potential nanodevices.

Importance: The Chinese Neonatal Network was established in 2018 and maintains a standardized national clinical database of very preterm or very low-birth-weight infants in tertiary neonatal intensive care units (NICUs) throughout China. National-level data on outcomes and care practices of very preterm infants (VPIs) in China are lacking. Objective: To assess the care practices in NICUs and outcomes among VPIs in China. Design, Setting, and Participants: A cohort study was conducted comprising 57 tertiary hospitals from 25 provinces throughout China. All infants with gestational age (GA) less than 32 weeks who were admitted to the 57 NICUs between January 1 and December 31, 2019, were included. Main Outcomes and Measures: Care practices, morbidities, and survival were the primary outcomes of the study. Major morbidities included bronchopulmonary dysplasia, severe intraventricular hemorrhage (grade ≥3) and/or periventricular leukomalacia, necrotizing enterocolitis (stage ≥2), sepsis, and severe retinopathy of prematurity (stage ≥3). Results: A total of 9552 VPIs were included, with mean (SD) GA of 29.5 (1.7) weeks and mean (SD) birth weight of 1321 (321) g; 5404 infants (56.6%) were male. Antenatal corticosteroids were used in 75.6% (6505 of 8601) of VPIs, and 54.8% (5211 of 9503)were born through cesarean delivery. In the delivery room, 12.1% of VPIs received continuous positive airway pressure and 26.7% (2378 or 8923) were intubated. Surfactant was prescribed for 52.7% of the infants, and postnatal dexamethasone was prescribed to 9.5% (636 of 6675) of the infants. A total of 85.5% (8171) of the infants received complete care, and 14.5% (1381) were discharged against medical advice. The incidences of the major morbidities were bronchopulmonary dysplasia, 29.2% (2379 of 8148); severe intraventricular hemorrhage and/or periventricular leukomalacia, 10.4% (745 of 7189); necrotizing enterocolitis, 4.9% (403 of 8171 ); sepsis, 9.4% (764 of 8171); and severe retinopathy of prematurity, 4.3% (296 of 6851) among infants who received complete care. Among VPIs with complete care, 95.4% (7792 of 8171) survived: 65.6% (155 of 236) at 25 weeks' or less GA, 89.0% (880 of 988) at 26 to 27 weeks' GA, 94.9% (2635 of 2755)at 28 to 29 weeks' GA, and 98.3% (4122 of 4192) at 30 to 31 weeks' GA. Only 57.2% (4677 of 8171) of infants survived without major morbidity: 10.5% (25 of 236) at 25 weeks' or less GA, 26.8% (48 of 179) at 26 to 27 weeks' GA, 51.1% (1409 of 2755) at 28 to 29 weeks' GA, and 69.3% (2904 of 4192) at 30 to 31 weeks' GA. Among all infants admitted, the survival rate was 87.6% (8370 of 9552)and survival without major morbidities was 51.8% (4947 of 9552). Conclusions and Relevance: The findings of this study suggest that survival and survival without major morbidity of VPIs in Chinese NICUs have improved but remain lower than in high-income countries. Comprehensive and targeted quality improvement efforts are needed to provide complete care for all VPIs, optimize obstetrical and neonatal care practices, and improve outcomes.

Most recent methods for RGB (red-green-blue)-thermal salient object detection (SOD) involve several floating-point operations and have numerous parameters, resulting in slow inference, especially on common processors, and impeding their deployment on mobile devices for practical applications. To address these problems, we propose a lightweight spatial boosting network (LSNet) for efficient RGB-thermal SOD with a lightweight MobileNetV2 backbone to replace a conventional backbone (e.g., VGG, ResNet). To improve feature extraction using a lightweight backbone, we propose a boundary boosting algorithm that optimizes the predicted saliency maps and reduces information collapse in low-dimensional features. The algorithm generates boundary maps based on predicted saliency maps without incurring additional calculations or complexity. As multimodality processing is essential for high-performance SOD, we adopt attentive feature distillation and selection and propose semantic and geometric transfer learning to enhance the backbone without increasing the complexity during testing. Experimental results demonstrate that the proposed LSNet achieves state-of-the-art performance compared with 14 RGB-thermal SOD methods on three datasets while improving the numbers of floating-point operations (1.025G) and parameters (5.39M), model size (22.1 MB), and inference speed (9.95 fps for PyTorch, batch size of 1, and Intel i5-7500 processor; 93.53 fps for PyTorch, batch size of 1, and NVIDIA TITAN V graphics processor; 936.68 fps for PyTorch, batch size of 20, and graphics processor; 538.01 fps for TensorRT and batch size of 1; and 903.01 fps for TensorRT/FP16 and batch size of 1). The code and results can be found from the link of https://github.com/zyrant/LSNet.