Optimal strategies for modeling anatomy in a hybrid intelligence framework for auto-segmentation of organs.

Organ segmentation is a fundamental requirement in medical image analysis. Many methods have been proposed over the past 6 decades for segmentation. A unique feature of medical images is the anatomical information hidden within the image itself. To bring natural intelligence (NI) in the form of anatomical information accumulated over centuries into deep learning (DL) AI methods effectively, we have recently introduced the idea of hybrid intelligence (HI) that combines NI and AI and a system based on HI to perform medical image segmentation. This HI system has shown remarkable robustness to image artifacts, pathology, deformations, etc. in segmenting organs in the Thorax body region in a multicenter clinical study. The HI system utilizes an anatomy modeling strategy to encode NI and to identify a rough container region in the shape of each object via a non-DL-based approach so that DL training and execution are applied only to the fuzzy container region. In this paper, we introduce several advances related to modeling of the NI component so that it becomes substantially more efficient computationally, and at the same time, is well integrated with the DL portion (AI component) of the system. We demonstrate a 9-40 fold computational improvement in the auto-segmentation task for radiation therapy (RT) planning via clinical studies obtained from 4 different RT centers, while retaining state-of-the-art accuracy of the previous system in segmenting 11 objects in the Thorax body region.

Short-Chain Fatty Acids and Preeclampsia: A Scoping Review.

BACKGROUND: Preeclampsia (PE) is a pregnancy-associated hypertension disorder with high morbidity and mortality. Short-chain fatty acids (SCFAs)-molecules produced by gut microbes-have been associated with hypertension, yet their relation to PE remains uncertain. OBJECTIVES: The aim was to review existing human studies that examined associations of the major SCFAs (acetate, propionate, butyrate) in pregnancy with PE development. METHODS: Two reviewers independently searched online databases (EMBASE, PubMed, Web of Science, and Cochrane Database of Systematic Reviews) in January 2024 using the following terms: "short-chain fatty acids," "acetic acid," "butyric acid," "propionic acid," and "preeclampsia." The final set of included studies had to report associations of SCFAs with PE, be peer-reviewed, be written in English, and be conducted in humans. RESULTS: The abstracts of 907 studies were screened; 43 underwent full-text screening and 11 (1318 total participants, 352 with PE) were included in the final review. All studies used a case-control design. SCFAs were measured in a range of biospecimens (eg, serum, plasma, feces, placentas, and amniotic fluid) that were collected at distinct time points in pregnancy. All 7 studies that investigated butyrate found that it was lower in PE cases than in controls, with 6 of these showing statistical significance (P < .05). Five studies showed that acetate was significantly lower in individuals with PE compared with healthy individuals, while 1 study found that acetate was significantly higher in PE cases. One study reported significantly higher propionate among PE cases vs controls, while 2 studies reported significantly lower propionate levels in PE cases. The nuance in results for acetate and propionate may owe to reasons such as differences in distributions of population characteristics associated with SCFA level and PE or type of PE (early vs late). CONCLUSION: Current epidemiologic evidence, which derives only from case-control studies, suggests that SCFAs, particularly butyrate (protective), in pregnancy are related to the development of PE. Large-cohort studies are warranted to investigate the temporality and potential causality of these associations.

Auto-segmentation of thoraco-abdominal organs in pediatric dynamic MRI.

Purpose: Analysis of the abnormal motion of thoraco-abdominal organs in respiratory disorders such as the Thoracic Insufficiency Syndrome (TIS) and scoliosis such as adolescent idiopathic scoliosis (AIS) or early onset scoliosis (EOS) can lead to better surgical plans. We can use healthy subjects to find out the normal architecture and motion of a rib cage and associated organs and attempt to modify the patient's deformed anatomy to match to it. Dynamic magnetic resonance imaging (dMRI) is a practical and preferred imaging modality for capturing dynamic images of healthy pediatric subjects. In this paper, we propose an auto-segmentation set-up for the lungs, kidneys, liver, spleen, and thoraco-abdominal skin in these dMRI images which have their own challenges such as poor contrast, image non-standardness, and similarity in texture amongst gas, bone, and connective tissue at several inter-object interfaces. Methods: The segmentation set-up has been implemented in two steps: recognition and delineation using two deep neural network (DL) architectures (say DL-R and DL-D) for the recognition step and delineation step, respectively. The encoder-decoder framework in DL-D utilizes features at four different resolution levels to counter the challenges involved in the segmentation. We have evaluated on dMRI sagittal acquisitions of 189 (near-)normal subjects. The spatial resolution in all dMRI acquisitions is 1.46 mm in a sagittal slice and 6.00 mm between sagittal slices. We utilized images of 89 (10) subjects at end inspiration for training (validation). For testing we experimented with three scenarios: utilizing (1) the images of 90 (=189-89-10) different (remaining) subjects at end inspiration for testing, (2) the images of the aforementioned 90 subjects at end expiration for testing, and (3) the images of the aforesaid 99 (=89+10) subjects but at end expiration for testing. In some situations, we can take advantage of already available ground truth (GT) of a subject at a particular respiratory phase to automatically segment the object in the image of the same subject at a different respiratory phase and then refining the segmentation to create the final GT. We anticipate that this process of creating GT would require minimal post hoc correction. In this spirit, we conducted separate experiments where we assume to have the ground truth of the test subjects at end expiration for scenario (1), end inspiration for (2), and end inspiration for (3). Results: Amongst these three scenarios of testing, for the DL-R, we achieve a best average location error (LE) of about 1 voxel for the lungs, kidneys, and spleen and 1.5 voxels for the liver and the thoraco- abdominal skin. The standard deviation (SD) of LE is about 1 or 2 voxels. For the delineation approach, we achieve an average Dice coefficient (DC) of about 0.92 to 0.94 for the lungs, 0.82 for the kidneys, 0.90 for the liver, 0.81 for the spleen, and 0.93 for the thoraco-abdominal skin. The SD of DC is lower for the lungs, liver, and the thoraco-abdominal skin, and slightly higher for the spleen and kidneys. Conclusions: Motivated by applications in surgical planning for disorders such as TIS, AIS, and EOS, we have shown an auto-segmentation system for thoraco-abdominal organs in dMRI acquisitions. This proposed setup copes with the challenges posed by low resolution, motion blur, inadequate contrast, and image intensity non-standardness quite well. We are in the process of testing its effectiveness on TIS patient dMRI data.

Reclaiming Agriceuticals from Sweetpotato (Ipomoea batatas [L.] Lam.) By-Products.

Sweetpotato (SP, Ipomoea batatas [L.] Lam.) is a globally significant food crop known for its high nutritional and functional values. Although the contents and compositions of bioactive constituents vary among SP varieties, sweetpotato by-products (SPBs), including aerial parts, storage root peels, and wastes generated from starch processing, are considered as excellent sources of polyphenols (e.g., chlorogenic acid, caffeoylquinic acid, and dicaffeoylquinic acid), lutein, functional carbohydrates (e.g., pectin, polysaccharides, and resin glycosides) or proteins (e.g., polyphenol oxidase, ß-amylase, and sporamins). This review summarises the health benefits of these ingredients specifically derived from SPBs in vitro and/or in vivo, such as anti-obesity, anti-cancer, antioxidant, cardioprotective, and anti-diabetic, evidencing their potential to regenerate value-added bio-products in the fields of food and nutraceutical. Accordingly, conventional and novel technologies have been developed and sometimes combined for the pretreatment and extraction processes aimed at optimising the recovery efficiency of bioactive ingredients from SPBs while ensuring sustainability. However, so far, advanced extraction technologies have not been extensively applied for recovering bioactive compounds from SPBs except for SP leaves. Furthermore, the incorporation of reclaimed bioactive ingredients from SPBs into foods or other healthcare products remains limited. This review also briefly discusses current challenges faced by the SPB recycling industry while suggesting that more efforts should be made to facilitate the transition from scientific advances to commercialisation for reutilising and valorising SPBs.

Policies, Practices, and Environmental Characteristics Among Family Child Care Homes in South Carolina.

Background: Child care program requirements have adopted nutrition and physical activity standards to address childhood obesity, but few studies have examined the effects of these standards in family child care homes (FCCHs). Methods: In a cross-sectional study (2017-2019), the Childcare Home Eating and Exercise study examined self-reported provider characteristics and observed policies and practices related to physical activity and nutrition in FCCHs in South Carolina. Two-sample t-tests were used to compare observed nutrition and physical activity policy, practice, and environment scores in child care homes that participated in versus did not participate in the state's ABC Quality program, which is designed to improve child care and includes policies and practices intended to increase physical activity levels and improve diet quality. Results: Environment and Policy Assessment and Observation results for nutrition and physical activity were 7.5 out of 21 and 11.8 out of 30, respectively, indicating much room for improvement in nutrition and physical activity policies, practices, and environment in South Carolina FCCHs. The study found one difference between FCCHs that did and did not participate in the ABC Quality program; non-ABC homes provided more time for physical activity. Conclusions: Future research should develop ways to strengthen the guidelines and improve the implementation of obesity prevention standards in FCCHs.

Three-Dimensional Scaffolds for Intestinal Cell Culture: Fabrication, Utilization, and Prospects.

The intestine is a visceral organ that integrates absorption, metabolism, and immunity, which is vulnerable to external stimulus. Researchers in the fields such as food science, immunology, and pharmacology have committed to developing appropriate in vitro intestinal cell models to study the intestinal absorption and metabolism mechanisms of various nutrients and drugs, or pathogenesis of intestinal diseases. In the past three decades, the intestinal cell models have undergone a significant transformation from conventional two-dimensional cultures to three-dimensional (3D) systems, and the achievements of 3D cell culture have been greatly contributed by the fabrication of different scaffolds. In this review, we first introduce the developing trend of existing intestinal models. Then, four types of scaffolds, including Transwell, hydrogel, tubular scaffolds, and intestine-on-a-chip, are discussed for their 3D structure, composition, advantages, and limitations in the establishment of intestinal cell models. Excitingly, some of the in vitro intestinal cell models based on these scaffolds could successfully mimic the 3D structure, microenvironment, mechanical peristalsis, fluid system, signaling gradients, or other important aspects of the original human intestine. Furthermore, we discuss the potential applications of the intestinal cell models in drug screening, disease modeling, and even regenerative repair of intestinal tissues. This review presents an overview of state-of-the-art scaffold-based cell models within the context of intestines, and highlights their major advances and applications contributing to a better knowledge of intestinal diseases. Impact statement The intestine tract is crucial in the absorption and metabolism of nutrients and drugs, as well as immune responses against external pathogens or antigens in a complex microenvironment. The appropriate experimental cell model in vitro is needed for in-depth studies of intestines, due to the limitation of animal models in dynamic control and real-time assessment of key intestinal physiological and pathological processes, as well as the "R" principles in laboratory animal experiments. Three-dimensional (3D) scaffold-based cell cultivation has become a developing tendency because of the superior cell proliferation and differentiation and more physiologically relevant environment supported by the customized 3D scaffolds. In this review, we summarize four types of up-to-date 3D cell culture scaffolds fabricated by various materials and techniques for a better recapitulation of some essential physiological and functional characteristics of original intestines compared to conventional cell models. These emerging 3D intestinal models have shown promising results in not only evaluating the pharmacokinetic characteristics, security, and effectiveness of drugs, but also studying the pathological mechanisms of intestinal diseases at cellular and molecular levels. Importantly, the weakness of the representative 3D models for intestines is also discussed.

GA-Net: A geographical attention neural network for the segmentation of body torso tissue composition.

PURPOSE: Body composition analysis (BCA) of the body torso plays a vital role in the study of physical health and pathology and provides biomarkers that facilitate the diagnosis and treatment of many diseases, such as type 2 diabetes mellitus, cardiovascular disease, obstructive sleep apnea, and osteoarthritis. In this work, we propose a body composition tissue segmentation method that can automatically delineate those key tissues, including subcutaneous adipose tissue, skeleton, skeletal muscle tissue, and visceral adipose tissue, on positron emission tomography/computed tomography scans of the body torso. METHODS: To provide appropriate and precise semantic and spatial information that is strongly related to body composition tissues for the deep neural network, first we introduce a new concept of the body area and integrate it into our proposed segmentation network called Geographical Attention Network (GA-Net). The body areas are defined following anatomical principles such that the whole body torso region is partitioned into three non-overlapping body areas. Each body composition tissue of interest is fully contained in exactly one specific minimal body area. Secondly, the proposed GA-Net has a novel dual-decoder schema that is composed of a tissue decoder and an area decoder. The tissue decoder segments the body composition tissues, while the area decoder segments the body areas as an auxiliary task. The features of body areas and body composition tissues are fused through a soft attention mechanism to gain geographical attention relevant to the body tissues. Thirdly, we propose a body composition tissue annotation approach that takes the body area labels as the region of interest, which significantly improves the reproducibility, precision, and efficiency of delineating body composition tissues. RESULTS: Our evaluations on 50 low-dose unenhanced CT images indicate that GA-Net outperforms other architectures statistically significantly based on the Dice metric. GA-Net also shows improvements for the 95% Hausdorff Distance metric in most comparisons. Notably, GA-Net exhibits more sensitivity to subtle boundary information and produces more reliable and robust predictions for such structures, which are the most challenging parts to manually mend in practice, with potentially significant time-savings in the post hoc correction of these subtle boundary placement errors. Due to the prior knowledge provided from body areas, GA-Net achieves competitive performance with less training data. Our extension of the dual-decoder schema to TransUNet and 3D U-Net demonstrates that the new schema significantly improves the performance of these classical neural networks as well. Heatmaps obtained from attention gate layers further illustrate the geographical guidance function of body areas for identifying body tissues. CONCLUSIONS: (i) Prior anatomic knowledge supplied in the form of appropriately designed anatomic container objects significantly improves the segmentation of bodily tissues. (ii) Of particular note are the improvements achieved in the delineation of subtle boundary features which otherwise would take much effort for manual correction. (iii) The method can be easily extended to existing networks to improve their accuracy for this application.

Urban Living Environment and Myopia in Children.

Importance: The global prevalence of myopia has shown a steady increase over recent decades, with urban areas seemingly experiencing a more significant impact. Objective: To assess the association between urbanization and the prevalence, incidence, progression, and severity of myopia. Design, Setting, and Participants: This cohort study included students in grades 1 to 6 in Tianjin, China, who underwent 3 vision examinations conducted over a 2-year period, from March 1, 2021, to March 31, 2023. Participants from grades 1 to 4 completed the 2-year follow-up. Exposures: Urban living environment. Main Outcomes and Measures: The association of urbanization with the incidence, progression, prevalence, and severity of myopia. To quantify urbanization, an urban score was constructed using satellite data and an iterative exploratory factor analysis. Results: Of 177 894 students (51.7% male; mean [SD] age, 10.27 [1.75] years) included in the study, 137 087 students (52.3% male; mean [SD] age, 8.97 [1.21] years) were followed up for 2 years. A positive association was identified between myopia incidence and urbanization. Specifically, each 1-unit increment in the urban score was associated with an increased risk of myopia over a 1-year period (odds ratio [OR], 1.09; 95% CI, 1.01-1.15; P = .02) and a 2-year period (OR, 1.53; 95% CI, 1.50-1.57; P < .001). Conversely, each 1-unit increase in the urban score was associated with a significant decrease in myopia progression at 1 year (OR, 0.84; 95% CI, 0.82-0.86; P < .001) and 2 years (OR, 0.73; 95% CI, 0.70-0.75, P < .001). In a cross-sectional data analysis, the urban score was positively associated with myopia prevalence (OR, 1.62; 95% CI, 1.08-2.42; P = .02) and negatively associated with myopia severity, as indicated by spherical equivalent refraction (OR, 1.46; 95% CI, 1.07-1.99; P = .02). Conclusions and Relevance: This study exploring urban living environments and myopia revealed dual associations of urban living with both the incidence and the progression of myopia. The observed patterns emphasize the urgency of promptly implementing myopia control strategies in less urbanized regions, where myopia progression may be accentuated.

Associations of Plastic Bottle Exposure with Infant Growth, Fecal Microbiota, and Short-Chain Fatty Acids.

BACKGROUND/OBJECTIVES: Murine models show that plastics, via their chemical constituents (e.g., phthalates), influence microbiota, metabolism, and growth. However, research on plastics in humans is lacking. Here, we examine how the frequency of plastic bottle exposure is associated with fecal microbiota, short-chain fatty acids (SCFAs), and anthropometry in the first year of life. SUBJECTS/METHODS: In 442 infants from the prospective Nurture birth cohort, we examined the association of frequency of plastic bottle feeding at 3 months with anthropometric outcomes (skinfolds, length-for-age, and weight-for-length) at 12 months of age and growth trajectories between 3 and 12 months. Furthermore, in a subset of infants (n = 70) that contributed fecal samples at 3 months and 12 months of age, we examined plastic bottle frequency in relation to fecal microbiota composition and diversity (measured by 16S rRNA gene sequencing of V4 region), and fecal SCFA concentrations (quantified using gas chromatography mass spectrometry). RESULTS: At 3 months, 67.6% of infants were plastic bottle fed at every feeding, 15.4% were exclusively breast milk fed, and 48.9% were exclusively formula fed. After adjustment for potential confounders, infants who were plastic bottle fed less than every feeding compared to those who were plastic bottle fed at every feeding at 3 months did not show differences in anthropometry over the first 12 months of life, save for lower length-for-age z-score at 12 months (adjusted ß = -0.45, 95% CI: -0.76, -0.13). Infants who were plastic bottle fed less than every feeding versus every feeding had lower fecal microbiota alpha diversity at 3 months (mean difference for Shannon index: -0.59, 95% CI: -0.99, -0.20) and lower isovaleric acid concentration at 3 months (mean difference: -2.12 µmol/g, 95% CI: -3.64, -0.60), but these results were attenuated following adjustment for infant diet. Plastic bottle frequency was not strongly associated with microbiota diversity or SCFAs at 12 months after multivariable adjustment. Frequency of plastic bottle use was associated with differential abundance of some bacterial taxa, however, significance was not consistent between statistical approaches. CONCLUSIONS: Plastic bottle frequency at 3 months was not strongly associated with measures of adiposity or growth (save for length-for-age) over the first year of life, and while plastic bottle use was associated with some features of fecal microbiota and SCFAs in the first year, these findings were attenuated in multivariable models with infant diet. Future research is needed to assess health effects of exposure to other plastic-based products and objective measures of microplastics and plastic constituents like phthalates.

An Activated Dendritic-Cell-Related Gene Signature Indicative of Disease Prognosis and Chemotherapy and Immunotherapy Response in Colon Cancer Patients.

Accumulating evidence has underscored the prognostic value of tumor-infiltrating immune cells in the tumor microenvironment of colon cancer (CC). In this retrospective study, based on publicly available transcriptome profiles and clinical data from the Gene Expression Omnibus and The Cancer Genome Atlas databases, we derived and verified an activated dendritic cell (aDC)-related gene signature (aDCRS) for predicting the survival outcomes and chemotherapy and immunotherapy response of CC patients. We quantified the infiltration abundance of 22 immune cell subtypes via the "CIBERSORT" R script. Univariate Cox proportional hazards (PHs) regression was used to identify aDC as the most robust protective cell type for CC prognosis. After selecting differentially expressed genes (DEGs) significantly correlated with aDC infiltration, we performed univariate Cox-PH regression, LASSO regression, and stepwise multivariate Cox-PH regression successively to screen out prognosis-related genes from selected DEGs for constructing the aDCRS. Receiver operating characteristic (ROC) curves and Kaplan-Meier (KM) analysis were employed to assess the discriminatory ability and risk-stratification capacity. The "oncoPredict" package, Cancer Treatment Response gene signature DataBase, and Tumor Immune Dysfunction and Exclusion algorithm were utilized to estimate the practicability of the aDCRS in predicting response to chemotherapy and immune checkpoint blockade. Gene set enrichment analysis and single-cell RNA sequencing analysis were also implemented. Furthermore, an aDCRS-based nomogram was constructed and validated via ROC curves, calibration plots and decision curve analysis. In conclusion, aDCRS and an aDCRS-based nomogram will facilitate precise prognosis prediction and individualized therapeutic interventions, thus improving the survival outcomes of CC patients in the future.

Young Children's Dietary Quality in Family Child Care and in Their Own Home.

BACKGROUND: Some evidence suggests that children may have higher quality dietary intake in early care and education settings, compared with their respective homes, but no studies have explored these differences among children in less formal family child care. OBJECTIVE: The purpose of this study was to compare dietary quality via the Healthy Eating Index 2015 among children in family child care and in their own home. DESIGN: This was a cross-sectional analysis of baseline dietary intake data from the Childcare Home Eating and Exercise Research study, a natural experiment, using directly observed dietary data in child care and 24-hour recall data in homes among children in South Carolina. PARTICIPANTS/SETTING: Participants were 123 children in 52 family child-care homes between 2018 and 2019. MAIN OUTCOME MEASURE: The main outcome was total and component Healthy Eating Index 2015 scores. STATISTICAL ANALYSIS: The analysis was a hierarchical linear regression of children nested within family child care homes adjusting for child, provider, facility, and parent characteristics, including sex, age, race, ethnicity, and income, with parameters and SEs estimated via bootstrap sampling. RESULTS: Children had a mean ± SD Healthy Eating Index 2015 score of 60.3 ± 12.1 in family child-care homes and 54.3 ± 12.9 in their own home (P < 0.001). In adjusted analysis and after accounting for clustering of children in family child care homes, total HEI-2015 scores were lower at home than in care (ß = -5.18 ± 1.47; 95% CI -8.05 to -2.30; P = 0.003). CONCLUSIONS: Children had healthier dietary intake in family child-care homes vs their respective homes.

A longitudinal study of infant 24-hour sleep: comparisons of sleep diary and accelerometer with different algorithms.

STUDY OBJECTIVES: To longitudinally compare sleep/wake identification and sleep parameter estimation from sleep diaries to accelerometers using different algorithms and epoch lengths in infants. METHODS: Mothers and other caregivers from the Nurture study (southeastern United States, 2013-2018) reported infants' 24-hour sleep in sleep diaries for 4 continuous days, while infants concurrently wore accelerometers on the left ankle at 3, 6, 9, and 12 months of age. We applied the Sadeh, Sadeh Infant, Cole, and Count-scaled algorithm to accelerometer data at 15 and 60 seconds epochs. For sleep/wake identification, we assessed agreement by calculating epoch-by-epoch percent agreement and kappas. We derived sleep parameters from sleep diaries and accelerometers separately and evaluated agreement using Bland-Altman plots. We estimated longitudinal trajectories of sleep parameters using marginal linear and Poisson regressions with generalized estimation equation estimation. RESULTS: Among the 477 infants, 66.2% were black and 49.5% were female. Agreement for sleep/wake identification varied by epoch length and algorithm. Relative to sleep diaries, we observed similar nighttime sleep offset, onset, and total nighttime sleep duration from accelerometers regardless of algorithm and epoch length. However, accelerometers consistently estimated about 1 less nap per day using the 15 seconds epoch, 70 and 50 minutes' shorter nap duration per day using the 15 and 60 seconds epoch, respectively; but accelerometers estimated over 3 times more wake after nighttime sleep onset (WASO) per night. Some consistent sleep parameter trajectories from 3 to 12 months from accelerometers and sleep diaries included fewer naps and WASOs, shorter total daytime sleep, longer total nighttime sleep, and higher nighttime sleep efficiency. CONCLUSIONS: Although there is no perfect measure of sleep in infancy, our findings suggest that a combination of accelerometer and diary may be needed to adequately measure infant sleep.

Auto-segmentation of thoracic brachial plexuses for radiation therapy planning.

Recently, deep learning networks have achieved considerable success in segmenting organs in medical images. Several methods have used volumetric information with deep networks to achieve segmentation accuracy. However, these networks suffer from interference, risk of overfitting, and low accuracy as a result of artifacts, in the case of very challenging objects like the brachial plexuses. In this paper, to address these issues, we synergize the strengths of high-level human knowledge (i.e., natural intelligence (NI)) with deep learning (i.e., artificial intelligence (AI)) for recognition and delineation of the thoracic brachial plexuses (BPs) in computed tomography (CT) images. We formulate an anatomy-guided deep learning hybrid intelligence approach for segmenting thoracic right and left brachial plexuses consisting of 2 key stages. In the first stage (AAR-R), objects are recognized based on a previously created fuzzy anatomy model of the body region with its key organs relevant for the task at hand wherein high-level human anatomic knowledge is precisely codified. The second stage (DL-D) uses information from AAR-R to limit the search region to just where each object is most likely to reside and performs encoder-decoder delineation in slices. The proposed method is tested on a dataset that consists of 125 images of the thorax acquired for radiation therapy planning of tumors in the thorax and achieves a Dice coefficient of 0.659.

A Deep Learning Based Geographic Attention Model for Body Composition Tissue Segmentation.

Measurement of body composition, including multiple types of adipose tissue, skeletal tissue, and skeletal muscle, on computed tomography (CT) images is practical given the powerful anatomical structure visualization ability of CT, and is useful for clinical and research applications related to health care and underlying pathology. In recent years, deep learning-based methods have contributed significantly to the development of automatic body composition analysis (BCA). However, the unsatisfactory segmentation performance for indistinguishable boundaries of multiple body composition tissues and the need for large-scale datasets for training deep neural networks still need to be addressed. This paper proposes a deep learning-based approach, called Geographic Attention Network (GA-Net), for body composition tissue segmentation on body torso positron emission tomography/computed tomography (PET/CT) images which leverages the body area information. The representation ability of GA-Net is significantly enhanced with the body area information as it strongly correlates with the target body composition tissue. This method achieves precise segmentation performance for multiple body composition tissues, especially for boundaries that are hard to distinguish, and effectively reduces the data requirements for training the network. We evaluate the proposed model on a dataset that includes 50 body torso PET/CT scans for segmenting 4 key bodily tissues - subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), skeletal muscle tissue (SMT), and skeleton (Sk). Experiments show that our proposed method increases segmentation accuracy, especially with a limited training dataset, by providing geographic information of target body composition tissues.

Pre-pregnancy body mass index and gut microbiota of mothers and children 5 years postpartum.

BACKGROUND: Maternal pre-pregnancy body mass index (BMI) has been linked to altered gut microbiota in women shortly after delivery and in their offspring in the first few years of life. But little is known about how long these differences persist. METHODS: We followed 180 mothers and children from pregnancy until 5-year postpartum in the Gen3G cohort (Canada, enrolled 2010-2013). At 5 years postpartum we collected stool samples from mothers and children and estimated the gut microbiota by 16 S rRNA sequencing (V4 region) using Illumina MiSeq, and assigning amplicon sequence variants (ASV). We examined whether overall microbiota composition (as measured by microbiota ß diversity) was more similar between mother-child pairs compared to between mothers or between children. We also assessed whether mother-child pair sharing of overall microbiota composition differed by the weight status of mothers before pregnancy and of children at 5-year. Furthermore, in mothers, we examined whether pre-pregnancy BMI, BMI 5-year postpartum, and change in BMI between time points was associated with maternal gut microbiota 5-year postpartum. In children, we further examined associations of maternal pre-pregnancy BMI and child 5-year BMI z-score with child 5-year gut microbiota. RESULTS: Mother-child pairs had greater similarity in overall microbiome composition compared to between mothers and between children. In mothers, higher pre-pregnancy BMI and 5-year postpartum BMI were associated with lower microbiota observed ASV richness and Chao 1 index; in children's gut microbiota, higher maternal pre-pregnancy BMI was weakly associated with lower microbiota Shannon index, whereas child's 5-year BMI z-score was associated with higher observed ASV richness. Pre-pregnancy BMI was also linked to differential abundances of several microbial ASVs in the Ruminococcaceae and Lachnospiraceae families, but no specific ASV had overlapping associations with BMI measures in both mothers and children. CONCLUSIONS: Pre-pregnancy BMI was associated with gut microbiota diversity and composition of mothers and children 5 years after birth, however, the nature and direction of most associations differed for mothers and children. Future studies are encouraged to confirm our findings and look into potential mechanisms or factors that may drive these associations.

The construction efficiency study of China National Ecological Civilization Pilot Zone with network SBM model: a city-based analysis.

The Chinese government proposed the establishment of China National Ecological Civilization Pilot Zone in 2016 to further explore the coordinated development of economy and environment. Fujian, Jiangxi, and Guizhou provinces were selected as the first batch of pilot zones. After years of exploration, it is necessary to discuss and summarize the construction progress of the three pilot zones from the perspective of the city. In this study, first, the ecological civilization pilot zone construction system was decomposed into an economic construction subsystem (ECS) and an environmental optimization subsystem (EOS). Then, a two-stage network SBM model was adopted to calculate the efficiencies of the subsystems, and the Kruskal-Wallis test was used to measure the efficiency difference. Finally, a panel data regression model was applied to explore the influencing factors of both subsystems. The results show that the ECS efficiency is higher than that of the EOS, and the ECS efficiency in Fujian is significantly better than that in Jiangxi and Guizhou. However, there is no significant difference in EOS efficiency in the three provinces. Furthermore, industrial structure and population agglomeration have a significant effect on ECS efficiency, environmental regulation has a significant impact on EOS, and the technology level has a significant impact on both subsystems. Based on the results, policy implications for improving the efficiency of the two subsystems were given respectively.