An explainable deep learning framework for characterizing and interpreting human brain states.

Deep learning approaches have been widely adopted in the medical image analysis field. However, a most of existing deep learning approaches focus on achieving promising performances such as classification, detection, and segmentation, and much less effort is devoted to the explanation of the designed models. Similarly, in the brain imaging field, many deep learning approaches have been designed and applied to characterize and predict human brain states. However, these models lack interpretation. In response, we propose a novel domain knowledge informed self-attention graph pooling-based (SAGPool) graph convolutional neural network to study human brain states. Specifically, the dense individualized and common connectivity-based cortical landmarks system (DICCCOL, structural brain connectivity profiles) and holistic atlases of functional networks and interactions system (HAFNI, functional brain connectivity profiles) are integrated with the SAGPool model to better characterize and interpret the brain states. Extensive experiments are designed and carried out on the large-scale human connectome project (HCP) Q1 and S1200 dataset. Promising brain state classification performances are observed (e.g., an average of 93.7% for seven-task classification and 100% for binary classification). In addition, the importance of the brain regions, which contributes most to the accurate classification, is successfully quantified and visualized. A thorough neuroscientific interpretation suggests that these extracted brain regions and their importance calculated from self-attention graph pooling layer offer substantial explainability.

Visualizing Research Trends and Identifying Hotspots of Traditional Chinese Medicine (TCM) Nursing Technology for Insomnia: A 18-Years Bibliometric Analysis of Web of Science Core Collection.

Objective: To explore the research hotspot and frontier direction of TCM nursing of insomnia and provide reference for the follow-up study of the optimal scheme of TCM nursing of insomnia. Background: Insomnia is a common sleep-wake disorder, affects 6-10% of adults and was associated with independent higher risks of cardiovascular disease and diabetes. TCM Nursing Techniques of insomnia has a long history and has shown a definite impact. However, it's still lack of analysis in the field of the most commonly used and effective techniques, as well as the co-morbidities associated with insomnia. Therefore, the database was searched and analyzed to find effective TCM Nursing Techniques for insomnia and related diseases related to insomnia. Method: Randomized controlled trials on the intervention of TCM Nursing Techniques in insomnia were retrieved from Web of Science Core Collection and imported into CiteSpace 5.6.R5 visualization software. The works of literature were co-cited by keywords authors and institutions for visual analysis, and the co-morbidities associated with insomnia of TCM Nursing Techniques in literature was extracted manually. The symptoms of co-morbidities associated with insomnia were imported into Cytoscape 3.9.0 software and clustered by CytoHubba. Result: As of October 20, 2021, the literature published in the last 20 years from Web of Science Core Collection was screened, and the publication period of the included literature was from 2004 to 2021. From 2016 to now, the total number of articles has been increasing. A total of 146 articles were included, and the highest production year was 2020. There is little cooperation between states, institutions, and authors. China (including Hong Kong and Taiwan) and Hong Kong Polytech University are leading countries and institutions in this area. MYUNGHAENG HUR is the most cited author, and J ALTERN COMPLEM MED is the most cited journal. According to cluster analysis and keyword frequency, auricular therapy, aromatherapy, and acupressure are the three most commonly used techniques. While the top five co-morbidities are fatigue, anxiety, depression, pain and hemodialysis. The three frontier topics and the main research directions are sleep quality, comorbid insomnia and clinical trial design. Conclusion: We found that acupressure, aromatherapy, and auricular acupoint therapy are the most commonly used nursing methods of TCM to intervene in insomnia. However, these studies have limitations such as small sample size, lack of objectivity in evaluating sleep quality, and high heterogeneity of intervention measures, which are not conducive to forming TCM clinical nursing guidelines. Therefore, it is necessary to adopt objectified sleep quality evaluation methods, select suitable acupoints according to TCM theories, and design multi-center large-sample clinical trials based on the safety principle of randomized blind control. This study provides an in-depth perspective for insomnia research on TCM Nursing Techniques and includes information for follow-up research on TCM Nursing Techniques of insomnia.

Listening for the therapeutic window: Advances in drug delivery utilizing photoacoustic imaging.

The preclinical landscape of photoacoustic imaging has experienced tremendous growth in the past decade. This non-invasive imaging modality augments the spatiotemporal capabilities of ultrasound with optical contrast. While it has principally been investigated for diagnostic applications, many recent reports have described theranostic delivery systems and drug monitoring strategies using photoacoustics. Here, we provide an overview of the progress to date while highlighting work in three specific areas: theranostic nanoparticles, real-time drug monitoring, and stem cell ("living drug") tracking. Additionally, we discuss the challenges that remain to be addressed in this burgeoning field.

Copper Sulfide Nanodisks and Nanoprisms for Photoacoustic Ovarian Tumor Imaging.

Transvaginal ultrasound is widely used for ovarian cancer screening but has a high false positive rate. Photoacoustic imaging provides additional optical contrast to supplement ultrasound and might be able to improve the accuracy of screening. Here, we report two copper sulfide (CuS) nanoparticles types (nanodisks and triangular nanoprisms) as the photoacoustic contrast agents for imaging ovarian cancer. Both CuS nanoprisms and nanodisks were ~6 nm thick and ~26 nm wide and were coated with poly(ethylene glycol) to make them colloidally stable in phosphate buffered saline (PBS) for at least 2 weeks. The CuS nanodisks and nanoprisms revealed strong localized surface plasmon resonances with peak maxima at 1145 nm and 1098 nm, respectively. Both nanoparticles types had strong and stable photoacoustic intensity with detection limits below 120 pM. The circular CuS nanodisk remained in the circulation of nude mice (n=4) and xenograft 2008 ovarian tumors (n=4) 17.9-fold and 1.8-fold more than the triangular nanoprisms, respectively. Finally, the photoacoustic intensity of the tumors from the mice (n=3) treated with CuS nanodisks was 3.0-fold higher than the baseline. The tumors treated with nanodisks had a characteristic peak at 920 nm in the spectrum to potentially differentiate the tumor from adjacent tissues.

Smart Albumin-Biomineralized Nanocomposites for Multimodal Imaging and Photothermal Tumor Ablation.

Smart cyanine-grafted gadolinium oxide nanocrystals (Cy-GdNCs) obtained by albumin-based biomineralization are shown to be theranostic nanocomposites, with promising properties for trimodal near-infrared fluorescence/photoacoustics/magnetic-resonance imaging-guided photothermal tumor ablation.