Identifying Links Between Productivity and Biobehavioral Rhythms Modeled From Multimodal Sensor Streams: Exploratory Quantitative Study.

BACKGROUND: Biobehavioral rhythms are biological, behavioral, and psychosocial processes with repeating cycles. Abnormal rhythms have been linked to various health issues, such as sleep disorders, obesity, and depression. OBJECTIVE: This study aims to identify links between productivity and biobehavioral rhythms modeled from passively collected mobile data streams. METHODS: In this study, we used a multimodal mobile sensing data set consisting of data collected from smartphones and Fitbits worn by 188 college students over a continuous period of 16 weeks. The participants reported their self-evaluated daily productivity score (ranging from 0 to 4) during weeks 1, 6, and 15. To analyze the data, we modeled cyclic human behavior patterns based on multimodal mobile sensing data gathered during weeks 1, 6, 15, and the adjacent weeks. Our methodology resulted in the creation of a rhythm model for each sensor feature. Additionally, we developed a correlation-based approach to identify connections between rhythm stability and high or low productivity levels. RESULTS: Differences exist in the biobehavioral rhythms of high- and low-productivity students, with those demonstrating greater rhythm stability also exhibiting higher productivity levels. Notably, a negative correlation (C=-0.16) was observed between productivity and the SE of the phase for the 24-hour period during week 1, with a higher SE indicative of lower rhythm stability. CONCLUSIONS: Modeling biobehavioral rhythms has the potential to quantify and forecast productivity. The findings have implications for building novel cyber-human systems that align with human beings' biobehavioral rhythms to improve health, well-being, and work performance.

VIBRANT: spectral profiling for single-cell drug responses.

High-content cell profiling has proven invaluable for single-cell phenotyping in response to chemical perturbations. However, methods with improved throughput, information content and affordability are still needed. We present a new high-content spectral profiling method named vibrational painting (VIBRANT), integrating mid-infrared vibrational imaging, multiplexed vibrational probes and an optimized data analysis pipeline for measuring single-cell drug responses. Three infrared-active vibrational probes were designed to measure distinct essential metabolic activities in human cancer cells. More than 20,000 single-cell drug responses were collected, corresponding to 23 drug treatments. The resulting spectral profile is highly sensitive to phenotypic changes under drug perturbation. Using this property, we built a machine learning classifier to accurately predict drug mechanism of action at single-cell level with minimal batch effects. We further designed an algorithm to discover drug candidates with new mechanisms of action and evaluate drug combinations. Overall, VIBRANT has demonstrated great potential across multiple areas of phenotypic screening.

The mediation role of resilience and postpartum traumatic stress disorder on parental attachment and the maternal-infant bonding.

AIMS: This study aimed to evaluate the correlation between parental attachment, resilience, postpartum traumatic stress disorder (PTSD), and maternal-infant bonding at 1 to 3 months postpartum. The mediation effect of resilience and PTSD on the postpartum parental attachment and maternal-infant bond was also evaluated. DESIGN: A cross-sectional research design was used. METHODS: A total of 400 postpartum women examined at a tertiary hospital in Wuhan from January 2021 to June 2021 were enrolled in the study. At about 1 to 3 months after giving birth, the women were asked to complete the Postpartum Bonding Questionnaire (PBQ), Connor-Davidson Resilience scale(CD-RISC), PTSD CheckList-Civilian version (PCL-C), and the Parental Bonding Instrument (PBI). The data were summarized using descriptive statistics. Mediation analyse and the Spearman correlation (r) were used to correlate the resilience and PTSD questionnaire scores. RESULTS: The care attachment dimension was significantly associated with resilience (r = 0.24, p < 0.01), PTSD (r = - 0.27, p < 0.01), and maternal-infant bonding (r = 0.10, p < 0.01), and the overprotection attachment dimension was significantly associated with resilience (r = - 0.11, p < 0.01), PTSD (r = 0.33, p < 0.01), and maternal-infant bonding (r = 0.16, p < 0.01). Resilience and PTSD can mediate the relationship between attachment and maternal-infant bonding. CONCLUSION: Parental attachment, resilience, and PTSD significantly affect maternal-infant bonding at 1 to 3 months postpartum. IMPACT: This study demonstrated that new interventions aimed at addressing PTSD symptoms and improving resilience might increase parental attachment and maternal-infant bonding after birth. However, further research is required to evaluate the success of these interventions.

Normal saline: Past, present, and future.

Normal saline (NS) is the most widely used agent in the medical field. However, from its origin to its widespread application, it remains a mystery. Moreover, there is an ongoing debate on whether its existence is reasonable, harmful to the human body, or will still exist in the future. The current review traces back to the origins of NS and provides a brief overview of the current situation of infusion. The purpose may shed some light on the possibility of the existence of NS in the future by elaborating on the origin of NS and the research status of the impact of NS on the human body.

Genetic identification of tissues and cell types underlying attention-deficit/hyperactivity disorder.

Background: Genome-wide association studies (GWASs) have identified numerous genetic variants associated with attention-deficit/hyperactivity disorder (ADHD), which is considered highly genetically heritable. However, because most of the variants located in the non-coding region of the human genome, the onset of ADHD requires further exploration. Methods: The risk genes involved in ADHD were identified by integrating GWAS summary data and expression quantitative trait locus (eQTL) data using summary-data-based Mendelian randomization (SMR) method. We then used a stratified linkage disequilibrium score regression (LDSR) method to estimate the contribution of ADHD-relevant tissues to its heritability to screen out disease-relevant tissues. To determine the ADHD-relevant cell types, we used an R package for expression-weighted cell type enrichment (EWCE) analysis. Results: By integrating the brain eQTL data and ADHD GWAS data using SMR, we identified 247 genes associated with ADHD. The LDSR applied to specifically expressed genes results showed that the ADHD risk genes were mainly enriched in brain tissue, especially in the mesencephalon, visual cortex, and frontal lobe regions. Further cell-type-specific analysis suggested that ADHD risk genes were highly expressed in excitatory neurons. Conclusion: The study showed that the etiology of ADHD is associated with excitatory neurons in the midbrain, visual cortex, and frontal lobe regions.

Towards Mapping Mouse Metabolic Tissue Atlas by Mid-Infrared Imaging with Heavy Water Labeling.

Understanding metabolism is of great significance to decipher various physiological and pathogenic processes. While great progress has been made to profile gene expression, how to capture organ-, tissue-, and cell-type-specific metabolic profile (i.e., metabolic tissue atlas) in complex mammalian systems is lagging behind, largely owing to the lack of metabolic imaging tools with high resolution and high throughput. Here, the authors applied mid-infrared imaging coupled with heavy water (D2 O) metabolic labeling to a scope of mouse organs and tissues. The premise is that, as D2 O participates in the biosynthesis of various macromolecules, the resulting broad C-D vibrational spectrum should interrogate a wide range of metabolic pathways. Applying multivariate analysis to the C-D spectrum, the authors successfully identified both inter-organ and intra-tissue metabolic signatures of mice. A large-scale metabolic atlas map between different organs from the same mice is thus generated. Moreover, leveraging the power of unsupervised clustering methods, spatially-resolved metabolic signatures of brain tissues are discovered, revealing tissue and cell-type specific metabolic profile in situ. As a demonstration of this technique, the authors captured metabolic changes during brain development and characterized intratumoral metabolic heterogeneity of glioblastoma. Altogether, the integrated platform paves a way to map the metabolic tissue atlas for complex mammalian systems.

Integrative analysis of DNA methylation and gene expression reveals key molecular signatures in acute myocardial infarction.

BACKGROUNDS: Acute myocardial infarction (AMI) has been one of the most fatal diseases among all types of heart diseases due to its rapid onset and high rates of fatality. Understanding accurately how multi-omics molecular features change at the early stage of AMI is crucial for its treatment. Currently, the changes involved in DNA methylation modification and gene expression of multiple genes have remained unexplored. RESULTS: We used the RNA-seq and MeDIP-seq on heart tissues from AMI mouse models at series of time points (Sham, AMI 10-min, 1-h, 6-h, 24-h and 72-h), to comprehensively describe the transcriptome and genome-wide DNA methylation changes at above time points. We identified 18814, 18614, 23587, 26018 and 33788 differential methylation positions (DMPs) and 123, 135, 731, 1419 and 2779 differentially expressed genes (DEGs) at 10-min, 1-h, 6-h, 24-h and 72-h AMI, respectively, compared with the sham group. Remarkably, the 6-h AMI with the drastic changes of DEGs and a large number of enriched functional pathways in KEGG may be the most critical stage of AMI process. The 4, 9, 40, 26, and 183 genes were further identified at each time point, based on the negative correlation (P < 0.05) between the differential mRNA expression and the differential DNA methylation. The mRNA and the promoter methylation expressions of five genes (Ptpn6, Csf1r, Col6a1, Cyba, and Map3k14) were validated by qRT-PCR and BSP methods, and the mRNA expressions were further confirmed to be regulated by DNA methylation in cardiomyocytes in vitro. CONCLUSIONS: Our findings profiled the molecular variations from the perspective of DNA methylation in the early stage of AMI and provided promising epigenetic-based biomarkers for the early clinical diagnosis and therapeutic targets of AMI.

Effect of aerobic training on exercise capacity and quality of life in patients older than 75 years with acute coronary syndrome undergoing percutaneous coronary intervention.

BACKGROUND: Limited clinical studies are available on early exercise-based cardiac rehabilitation in elderly acute coronary syndrome (ACS) patients. OBJECTIVE: To evaluate the effect of aerobic exercise on exercise capacity and quality of life (QoL) in such patients. METHODS: Seventy elderly patients with ACS undergoing percutaneous coronary intervention in Zhejiang Hospital during August 2016-June 2017 were randomly divided into the control (n = 35) or cardiac rehabilitation group (CR, n = 35). The control group was treated with standard medical treatments without exercise, whereas the CR group was treated with standard medical treatments and exercise-based cardiac rehabilitation. General information, cardiopulmonary exercise test (CPET) results, responses to QoL and mental health questionnaires, and clinical outcomes and safety were collected. RESULTS: The CR group safely finished CPET and the 12-week exercise-based cardiac rehabilitation. After the 12-week intervention, the CR group showed significant differences in maximal oxygen uptake (VO2max) and greater improvements in VO2max, compared with the control group. The CR group showed statistically significant differences in QoL and mental health compared with the control group. CONCLUSION: CPET-based exercise in cardiac rehabilitation can safely increase exercise capacity and QoL in such patients.

An Approach for Deep Learning in ECG Classification Tasks in the Presence of Noisy Labels.

Cardiovascular disease (CVD) is a serial of diseases with global leading causes of death. Electrocardiogram (ECG) is the most commonly used basis for CVD diagnosis due to its low cost and no injury. Due to the great performance shown in classification tasks with large-scale data sets, deep learning has been widely applied in ECG diagnosis. Manual labeling is a time-consuming and labor-intensive job, which makes it error-prone and easy to labeled wrongly. These noisy labels cause deterioration in performance since deep neural network is easy to over-fitting with noisy labels. However, currently, only limited studies have been concerned with this problem. To alleviate the performance degradation caused by noisy labels, we come up with an optimization method combining data clean and anti-noise loss function. Our method filters the noisy data by data-clean method, followed by training the network with boot-hard loss function. The experiment is carried on MIT-BIH arrhythmia database and we take a 1-D CNN model for test. The result indicates that our optimization method can produce an effective improvement for noisy label problems when the proportion of incorrect labels ranging from 10% to 50%.Clinical Relevance- The proposed algorithm can be potentially applied to deal with the noisy label problem in ECG diagnosis task.

A One-Dimensional Siamese Few-Shot Learning Approach for ECG Classification under Limited Data.

Electrocardiogram (ECG) is mainly used by medical domain to diagnose arrhythmia. With the development of deep learning algorithms in the ECG classification field, related algorithms have achieved very high accuracy. However, the training of deep learning algorithms always requires large amounts of samples, while the labeled samples are often lacked in the field of medical signals. Therefore, the performance of deep learning algorithms will be greatly restricted. To overcome the sample scarcity problem, we propose a few-shot ECG classification approach based on the Siamese network. This network architecture first uses two one-dimensional convolutional neural network (CNN) that share weights to extract feature vectors of the paired input signals. Then, L1-distance between the two feature vectors is calculated and inputted into the fully connected layer with an activation function sigmoid to determine whether the input pairs belong to same category. We validated our method on the MIT-BIH arrhythmia database. By experiments, our method performs better than existing networks under the circumstance of extremely few amounts of data.

Dual Attention Convolutional Neural Network Based on Adaptive Parametric ReLU for Denoising ECG Signals with Strong Noise.

Electrocardiogram (ECG) signal is one of the most important methods for diagnosing cardiovascular diseases but is usually affected by noises. Denoising is therefore necessary before further analysis. Deep learning-related methods have been applied to image processing and other domains with great success but are rarely used for denoising ECG signals. This paper proposes an effective and simple model of encoder-decoder structure for denoising ECG signals (APR-CNN). Specifically, Adaptive Parametric ReLU (APReLU) and Dual Attention Module (DAM) are introduced in the model. Rectified Linear Unit (ReLU) is replaced with the APReLU for better negative information retainment. The DAM is an attention-based module consisting of a channel attention module and spatial attention module, through which the inter-spatial and inter-channel relationship of the input data are exploited. We tested our model on the MIT-BIH dataset, and the results show that the APR-CNN can handle ECG signals with a different signal-to-noise ratio (SNR). The comparative experiment proves our model is better than other deep learning and traditional methods.Clinical Relevance- This paper proposed a method capable of denoising ECG signals with strong noise to alleviate difficulties for further medical analysis.

Panax Notoginseng Saponins Protect H9c2 Cells From Hypoxia-reoxygenation Injury Through the Forkhead Box O3a Hypoxia-inducible Factor-1 Alpha Cell Signaling Pathway.

ABSTRACT: Panax notoginseng saponins (PNS) are commonly used in the treatment of cardiovascular diseases. Whether PNS can protect myocardial ischemia-reperfusion injury by regulating the forkhead box O3a hypoxia-inducible factor-1 alpha (FOXO3a/HIF-1α) cell signaling pathway remains unclear. The purpose of this study was to investigate the protective effect of PNS on H9c2 cardiomyocytes through the FOXO3a/HIF-1α cell signaling pathway. Hypoxia and reoxygenation of H9C2 cells were used to mimic MIRI in vitro, and the cells were treated with PNS, 2-methoxyestradiol (2ME2), and LY294002." Cell proliferation, lactate dehydrogenase, and malonaldehyde were used to evaluate the degree of cell injury. The level of reactive oxygen species was detected with a fluorescence microscope. The apoptosis rate was detected by flow cytometry. The expression of autophagy-related proteins and apoptosis-related proteins was detected by western blot assay. PNS could reduce H9c2 hypoxia-reoxygenation injury by promoting autophagy and inhibiting apoptosis through the HIF-1α/FOXO3a cell signaling pathway. Furthermore, the protective effects of PNS were abolished by HIF-1α inhibitor 2ME2 and PI3K/Akt inhibitor LY294002. PNS could reduce H9c2 hypoxia-reoxygenation injury by promoting autophagy and inhibiting apoptosis through the HIF-1α/FOXO3a cell signaling pathway.

Deep learning in magnetic resonance image reconstruction.

Magnetic resonance (MR) imaging visualises soft tissue contrast in exquisite detail without harmful ionising radiation. In this work, we provide a state-of-the-art review on the use of deep learning in MR image reconstruction from different image acquisition types involving compressed sensing techniques, parallel image acquisition and multi-contrast imaging. Publications with deep learning-based image reconstruction for MR imaging were identified from the literature (PubMed and Google Scholar), and a comprehensive description of each of the works was provided. A detailed comparison that highlights the differences, the data used and the performance of each of these works were also made. A discussion of the potential use cases for each of these methods is provided. The sparse image reconstruction methods were found to be most popular in using deep learning for improved performance, accelerating acquisitions by around 4-8 times. Multi-contrast image reconstruction methods rely on at least one pre-acquired image, but can achieve 16-fold, and even up to 32- to 50-fold acceleration depending on the set-up. Parallel imaging provides frameworks to be integrated in many of these methods for additional speed-up potential. The successful use of compressed sensing techniques and multi-contrast imaging with deep learning and parallel acquisition methods could yield significant MR acquisition speed-ups within clinical routines in the near future.

Recurrence Plot-Based Approach for Cardiac Arrhythmia Classification Using Inception-ResNet-v2.

The present study addresses the cardiac arrhythmia (CA) classification problem using the deep learning (DL)-based method for electrocardiography (ECG) data analysis. Recently, various DL techniques have been utilized to classify arrhythmias, with one typical approach to developing a one-dimensional (1D) convolutional neural network (CNN) model to handle the ECG signals in the time domain. Although the CA classification in the time domain is very prevalent, current methods' performances are still not robust or satisfactory. This study aims to develop a solution for CA classification in two dimensions by introducing the recurrence plot (RP) combined with an Inception-ResNet-v2 network. The proposed method for nine types of CA classification was tested on the 1st China Physiological Signal Challenge 2018 dataset. During implementation, the optimal leads (lead II and lead aVR) were selected, and then 1D ECG segments were transformed into 2D texture images by the RP approach. These RP-based images as input signals were passed into the Inception-ResNet-v2 for CA classification. In the CPSC, Georgia, and the PTB_XL ECG databases of the PhysioNet/Computing in Cardiology Challenge 2020, the RP-based method achieved an average F1-score of 0.8521, 0.8529, and 0.8862, respectively. The results suggested the excellent generalization ability of the proposed method. To further assess the performance of the proposed method, we compared the 2D RP-image-based solution with the published 1D ECG-based works on the same dataset. Also, it was compared with two traditional ECG transform into 2D image methods, including the time waveform of the ECG recordings and time-frequency images based on continuous wavelet transform (CWT). The proposed method achieved the highest average F1-score of 0.844, with only two leads of the 12-lead ECG original data, which outperformed other works. Therefore, the promising results indicate that the 2D RP-based method has a high clinical potential for CA classification using fewer lead ECG signals.

Deep unregistered multi-contrast MRI reconstruction.

Multiple magnetic resonance images of different contrasts are normally acquired for clinical diagnosis. Recently, research has shown that the previously acquired multi-contrast (MC) images of the same patient can be used as anatomical prior to accelerating magnetic resonance imaging (MRI). However, current MC-MRI networks are based on the assumption that the images are perfectly registered, which is rarely the case in real-world applications. In this paper, we propose an end-to-end deep neural network to reconstruct highly accelerated images by exploiting the shareable information from potentially misaligned reference images of an arbitrary contrast. Specifically, a spatial transformation (ST) module is designed and integrated into the reconstruction network to align the pre-acquired reference images with the images to be reconstructed. The misalignment is further alleviated by maximizing the normalized cross-correlation (NCC) between the MC images. The visualization of feature maps demonstrates that the proposed method effectively reduces the misalignment between the images for shareable information extraction when applied to the publicly available brain datasets. Additionally, the experimental results on these datasets show the proposed network allows the robust exploitation of shareable information across the misaligned MC images, leading to improved reconstruction results.

Optimizing multicontrast MRI reconstruction with shareable feature aggregation and selection.

This paper proposes a new method for optimizing feature sharing in deep neural network-based, rapid, multicontrast magnetic resonance imaging (MC-MRI). Using the shareable information of MC images for accelerated MC-MRI reconstruction, current algorithms stack the MC images or features without optimizing the sharing protocols, leading to suboptimal reconstruction results. In this paper, we propose a novel feature aggregation and selection scheme in a deep neural network to better leverage the MC features and improve the reconstruction results. First, we propose to extract and use the shareable information by mapping the MC images into multiresolution feature maps with multilevel layers of the neural network. In this way, the extracted features capture complementary image properties, including local patterns from the shallow layers and semantic information from the deep layers. Then, an explicit selection module is designed to compile the extracted features optimally. That is, larger weights are learned to incorporate the constructive, shareable features; and smaller weights are assigned to the unshareable information. We conduct comparative studies on publicly available T2-weighted and T2-weighted fluid attenuated inversion recovery brain images, and the results show that the proposed network consistently outperforms existing algorithms. In addition, the proposed method can recover the images with high fidelity under 16 times acceleration. The ablation studies are conducted to evaluate the effectiveness of the proposed feature aggregation and selection mechanism. The results and the visualization of the weighted features show that the proposed method does effectively improve the usage of the useful features and suppress useless information, leading to overall enhanced reconstruction results. Additionally, the selection module can zero-out repeated and redundant features and improve network efficiency.

Facile synthesis of the crescent-like SnS nanocrystals capped by polyvinyl pyrrolidone and its performance of adsorbing dyes.

With using Sn2+ as tin source, l-cysteine as sulphur source and polyvinyl pyrrolidone (PVP, Mw = 1300000) as surfactant, a novel three-dimensional and crescent-like SnS nanocrystal (NCs) was successfully synthesized in a one-pot hydrothermal method. The as-prepared SnS NCs displayed uniform crescent-like morphological structure, and demonstrated excellent efficiency for the adsorption of cationic dyes such as rhodamine B (RhB) and methylene blue (MB). Kinetic analysis indicated that the adsorption process followed the pseudo second-order model, and the maximum capacity of the SnS NCs to adsorb MB was determined by Langmuir equation to be 252 mg⋅g-1 at 298 K. The pH dependence of SnS NCs on the adsorption of cationic dyes and the characterization of zeta potential jointly suggested the existence of electrostatic attraction in the process. Overall, this study showed that electrostatic field of functional groups and the capping of PVP could significantly enhance the adsorption performance of the SnS NCs, and also provides a novel insight into the development of highly efficient inorganic adsorbents for cationic dyes.

Differences in Immune Responses between Children and Adults with COVID-19.

Over 85 590 000 individuals have been infected with severe acute respiratory syndrome Coronavirus 2 (SARS-CoV-2). Although there have been an increasing number of reports on coronavirus disease 2019 (COVID-19), it is unclear why infected children show milder symptoms than adults. A retrospective case study was performed at two designated hospitals for COVID-19. Patients (56 children and 63 adults) with confirmed SARS-CoV-2 infection and mild pneumonia were randomly enrolled in this study. The median age of the children was 7.0 years, and 51.79% of them were boys. The median age of the adults was 57 years, and 47.62% were men. The most common symptoms were fever, cough, sputum and diarrhoea. There were no significant differences in symptoms between children and adult patients. In terms of immunological indices on admission, adult patients displayed typical leukopenia and markedly higher levels of IL-2, IL-4, and IL-6 than child patients. The elevation of IL-2, IL-4 and IL-6 in adults induced more extensive lung injury. The effective and non-aggressive immune response successfully resisted SARS-CoV-2 invasion and maintained mild symptoms in child patients. The correlation of higher IL-2, IL-4, and IL-6 with the lung injury might be evidence that preventing excessive cytokine production can avoid further lung damage in these patients.

On the regularization of feature fusion and mapping for fast MR multi-contrast imaging via iterative networks.

Multi-contrast (MC) Magnetic Resonance Imaging (MRI) of the same patient usually requires long scanning times, despite the images sharing redundant information. In this work, we propose a new iterative network that utilizes the sharable information among MC images for MRI acceleration. The proposed network has reinforced data fidelity control and anatomy guidance through an iterative optimization procedure of Gradient Descent, leading to reduced uncertainties and improved reconstruction results. Through a convolutional network, the new method incorporates a learnable regularization unit that is capable of extracting, fusing, and mapping shareable information among different contrasts. Specifically, a dilated inception block is proposed to promote multi-scale feature extractions and increase the receptive field diversity for contextual information incorporation. Lastly, an optimal MC information feeding protocol is built through the design of a complementary feature extractor block. Comprehensive experiments demonstrated the superiority of the proposed network, both qualitatively and quantitatively.

Factors associated with postpartum women's knowledge, attitude and practice regarding human milk banks and milk donation: A cross-sectional survey.

BACKGROUND: Breast milk is the optimal method of human nutrition, and donor human milk is often needed to reduce the incidence of necrotizing enterocolitis and septicemia in preterm infants and improve their survival rate. Donor human milk is recommended as the first alternative when mothers' milk is not available. The establishment of human milk banks is of great significance to promote the breastfeeding of preterm infants. However, there are insufficient studies on human milk banks and milk donation in China. OBJECTIVES: (1) To investigate postpartum women's knowledge, attitude and practice regarding human milk banks and milk donation and to analyze the influencing factors. (2) To explore reasons why postpartum women reject milk donation and donor milk. DESIGN AND SETTINGS: A cross-sectional survey was conducted from February 2019 to July 2019 at two hospitals in Wuhan, a large city in central China. PARTICIPANTS: Mothers who returned to hospital for postpartum follow-up within six months participated in this survey (N = 1078). METHODS: Questionnaires were used to obtain sociodemographic data and to determine participants' knowledge, attitude and practice regarding human milk banks and milk donation. FINDINGS: Of the respondents, 216 (20%) had prior knowledge of human milk banks and milk donation. For the sub-domain of knowledge, the item with the highest correct response rate was the benefit of breast milk, and the item with the lowest correct rate was the acceptance of donor human milk. For the sub-domain of attitude, 811(75.3%) of participants held a supportive attitude for the establishment of human milk banks, and 877(81.3%) were supportive of donating breast milk while 412 (38.3%) were supportive of accepting donor human milk. For the sub-domain of practice, the practice of milk donation was not optimistic as participants lacked interest in donating breast milk and spreading knowledge of breast milk banks, and only 28.3% of participants indicated that they would donate breast milk continuously. Participants' age, educational background, weight of the newborn and having prior knowledge of human milk banks were factors that could positively predict their knowledge, attitude and practice associated with human milk banks and milk donation; medication usage during pregnancy or lactation was a factor negatively predicting their knowledge about human milk banks and milk donation. CONCLUSION: This study reveals that a majority of postpartum women are supportive of human milk banks and more willing to donate breast milk than receive donor milk. Lack of knowledge about human milk banks and safety concerns are the main factors hindering postpartum women from donating or accepting donor milk. Findings suggest that it is important to enhance public awareness regarding human milk banks as potential resources for life-saving therapy for preterm infants.This information should be disseminated during the early stage of the establishment of human milk banks. Moreover, health education of pregnant women should include the importance of human milk as well as the alternative and safety of donor milk from milk banks, especially for promoting the health of preterm infants and infants who are unable to receive mothers' breastmilk.