Pathogen detection via inductively coupled plasma mass spectrometry analysis with nanoparticles.

Infections caused by viruses and bacteria pose a significant threat to global public health, emphasizing the critical importance of timely and precise detection methods. Inductively coupled plasma mass spectrometry (ICP-MS), a contemporary approach for pathogen detection, offers distinct advantages such as high sensitivity, a wide linear range, and multi-index capabilities. This review elucidates the underexplored application of ICP-MS in conjunction with functional nanoparticles (NPs) for the identification of viruses and bacteria. The review commences with an elucidation of the underlying principles, procedures, target pathogens, and NP requirements for this innovative approach. Subsequently, a thorough analysis of the advantages and limitations associated with these techniques is provided. Furthermore, the review delves into a comprehensive examination of the challenges encountered when utilizing NPs and ICP-MS for pathogen detection, culminating in a forward-looking assessment of the potential pathways for advancement in this domain. Thus, this review contributes novel perspectives to the field of pathogen detection in biomedicine by showcasing the promising synergy of ICP-MS and NPs.

Nano-enzyme hydrogels for cartilage repair effectiveness based on ternary strategy therapy.

Designing artificial nano-enzymes for scavenging reactive oxygen species (ROS) in chondrocytes (CHOs) is considered the most feasible pathway for the treatment of osteoarthritis (OA). However, the accumulation of ROS due to the amount of nano-enzymatic catalytic site exposure and insufficient oxygen supply seriously threatens the clinical application of this therapy. Although metal-organic framework (MOF) immobilization of artificial nano-enzymes to enhance active site exposure has been extensively studied, artificial nano-enzymes/MOFs for ROS scavenging in OA treatment are still lacking. In this study, a biocompatible lubricating hydrogel-loaded iron-doped zeolitic imidazolate framework-8 (Fe/ZIF-8/Gel) centrase was engineered to scavenge endogenous overexpressed ROS synergistically generating dissolved oxygen and enhancing sustained lubrication for CHOs as a ternary artificial nano-enzyme. This property enabled the nano-enzymatic hydrogels to mitigate OA hypoxia and inhibit oxidative stress damage successfully. Ternary strategy-based therapies show excellent cartilage repair in vivo. The experimental results suggest that nano-enzyme-enhanced lubricating hydrogels are a potentially effective OA treatment and a novel strategy.

AI-Enabled Portable E-Nose Regression Predicting Harmful Molecules in a Gas Mixture.

The biomimetic electronic nose (e-nose) technology is a novel technology used for the identification and monitoring of complex gas molecules, and it is gaining significance in this field. However, due to the complexity and multiplicity of gas mixtures, the accuracy of electronic noses in predicting gas concentrations using traditional regression algorithms is not ideal. This paper presents a solution to the difficulty by introducing a fusion network model that utilizes a transformer-based multikernel feature fusion (TMKFF) module combined with a 1DCNN_LSTM network to enhance the accuracy of regression prediction for gas mixture concentrations using a portable electronic nose. The experimental findings demonstrate that the regression prediction performance of the fusion network is significantly superior to that of single models such as convolutional neural network (CNN) and long short-term memory (LSTM). The present study demonstrates the efficacy of our fusion network model in accurately predicting the concentrations of multiple target gases, such as SO2, NO2, and CO, in a gas mixture. Specifically, our algorithm exhibits substantial benefits in enhancing the prediction performance of low-concentration SO2 gas, which is a noteworthy achievement. The determination coefficient (R2) values of 93, 98, and 99% correspondingly demonstrate that the model is very capable of explaining the variation in the concentration of the target gases. The root-mean-square errors (RMSE) are 0.0760, 0.0711, and 3.3825, respectively, while the mean absolute errors (MAE) are 0.0507, 0.0549, and 2.5874, respectively. These results indicate that the model has relatively small prediction errors. The method we have developed holds significant potential for practical applications in detecting atmospheric pollution detection and other molecular detection areas in complex environments.

Bioinspired integrated triboelectric electronic tongue.

An electronic tongue (E-tongue) comprises a series of sensors that simulate human perception of taste and embedded artificial intelligence (AI) for data analysis and recognition. Traditional E-tongues based on electrochemical methods suffer from a bulky size and require larger sample volumes and extra power sources, limiting their applications in in vivo medical diagnosis and analytical chemistry. Inspired by the mechanics of the human tongue, triboelectric components have been incorporated into E-tongue platforms to overcome these limitations. In this study, an integrated multichannel triboelectric bioinspired E-tongue (TBIET) device was developed on a single glass slide chip to improve the device's taste classification accuracy by utilizing numerous sensory signals. The detection capability of the TBIET was further validated using various test samples, including representative human body, environmental, and beverage samples. The TBIET achieved a remarkably high classification accuracy. For instance, chemical solutions showed 100% identification accuracy, environmental samples reached 98.3% accuracy, and four typical teas demonstrated 97.0% accuracy. Additionally, the classification accuracy of NaCl solutions with five different concentrations reached 96.9%. The innovative TBIET exhibits a remarkable capacity to detect and analyze droplets with ultrahigh sensitivity to their electrical properties. Moreover, it offers a high degree of reliability in accurately detecting and analyzing various liquid samples within a short timeframe. The development of a self-powered portable triboelectric E-tongue prototype is a notable advancement in the field and is one that can greatly enhance the feasibility of rapid on-site detection of liquid samples in various settings.

Overexpression of Fgf18 in cranial neural crest cells recapitulates Pierre Robin sequence in mice.

The pivotal role of FGF18 in the regulation of craniofacial and skeletal development has been well established. Previous studies have demonstrated that mice with deficiency in Fgf18 exhibit severe craniofacial dysplasia. Recent clinical reports have revealed that the duplication of chromosome 5q32-35.3, which encompasses the Fgf18 gene, can lead to cranial bone dysplasia and congenital craniosynostosis, implicating the consequence of possible overdosed FGF18 signaling. This study aimed to test the effects of augmented FGF18 signaling by specifically overexpressing the Fgf18 gene in cranial neural crest cells using the Wnt1-Cre;pMes-Fgf18 mouse model. The results showed that overexpression of Fgf18 leads to craniofacial abnormalities in mice similar to the Pierre Robin sequence in humans, including abnormal tongue morphology, micrognathia, and cleft palate. Further examination revealed that elevated levels of Fgf18 activated the Akt and Erk signaling pathways, leading to an increase in the proliferation level of tongue tendon cells and alterations in the contraction pattern of the genioglossus muscle. Additionally, we observed that excessive FGF18 signaling contributed to the reduction in the length of Meckel's cartilage and disrupted the development of condylar cartilage, ultimately resulting in mandibular defects. These anomalies involve changes in several downstream signals, including Runx2, p21, Akt, Erk, p38, Wnt, and Ihh. This study highlights the crucial role of maintaining the balance of endogenous FGF18 signaling for proper craniofacial development and offers insights into potential formation mechanisms of the Pierre Robin sequence.

Robust antibacterial activity of rare-earth ions on planktonic and biofilm bacteria.

Bacterial infections pose a serious threat to human health, with emerging antibiotic resistance, necessitating the development of new antibacterial agents. Cu2+and Ag+are widely recognized antibacterial agents with a low propensity for inducing bacterial resistance; however, their considerable cytotoxicity constrains their clinical applications. Rare-earth ions, owing to their unique electronic layer structure, hold promise as promising alternatives. However, their antibacterial efficacy and biocompatibility relative to conventional antibacterial agents remain underexplored, and the variations in activity across different rare-earth ions remain unclear. Here, we systematically evaluate the antibacterial activity of five rare-earth ions (Yb3+, Gd3+, Sm3+, Tb3+, and La3+) againstStaphylococcus aureusandPseudomonas aeruginosa, benchmarked against well-established antibacterial agents (Cu2+, Ag+) and the antibiotic norfloxacin. Cytotoxicity is also assessed via live/dead staining of fibroblasts after 24 h rare-earth ion exposure. Our findings reveal that rare-earth ions require higher concentrations to match the antibacterial effects of traditional agents but offer the advantage of significantly lower cytotoxicity. In particular, Gd3+demonstrates potent bactericidal efficacy against both planktonic and biofilm bacteria, while maintaining the lowest cytotoxicity toward mammalian cells. Moreover, the tested rare-earth ions also exhibited excellent antifungal activity againstCandida albicans. This study provides a critical empirical framework to guide the selection of rare-earth ions for biomedical applications, offering a strategic direction for the development of novel antimicrobial agents.

Assessing the role of dental insurance in oral health care disparities in Canadian adults.

Background: This study examines the association of dental insurance with oral health care access and utilization in Canada while accounting for income and sociodemographic factors. It contributes to a baseline of oral health care disparities before the implementation of the Canadian Dental Care Plan (CDCP). Data and methods: This retrospective study of Canadians aged 18 to 64 years is based on data from the 2022 Canadian Community Health Survey. Multivariable logistic regression was employed to evaluate the association of dental insurance with the recency and frequency of dental visits, as well as avoidance of dental care because of cost. Results: Overall, 65.7% of Canadians reported visiting a dental professional in the previous year: 74.6% of those with private insurance, 62.8% with public insurance, and 49.8% uninsured. Cost-related avoidance of dental care was 16.0%, 20.9%, and 47.4% for the privately insured, publicly insured, and uninsured, respectively. After adjustment, adults with private (odds ratio [OR]=2.54; 95% confidence interval [CI]: 2.32 to 2.78) and public (OR=2.17; 95% CI: 1.75 to 2.68) insurance were more likely to have visited a dental professional in the last year compared with those without insurance. Similarly, both private (OR=0.22; 95% CI: 0.20 to 0.25) and public (OR=0.22; 95% CI: 0.17 to 0.29) insurance holders showed a significantly lower likelihood of avoiding dental visits because of cost when compared with uninsured individuals. Interpretation: This study showed the significant association of dental insurance with access to oral health care in Canada, contributing to setting a critical benchmark for assessments of the CDCP's effectiveness in addressing oral health disparities.

A review of copper-based Fenton reactions for the removal of organic pollutants from wastewater over the last decade: different reaction systems.

In recent years, as global industrialization has intensified, environmental pollution has become an increasingly serious problem. Improving water quality and achieving wastewater purification remain top priorities for environmental health initiatives. The Fenton process is favored by researchers due to its high efficiency and ease of operation. Central to the Fenton process is a catalyst used to activate hydrogen peroxide, rapidly degrading pollutants, improving water quality. Among various catalysts developed, copper-based catalysts have attracted considerable attention due to their affordability, high activity, and stable performance. Based on this, this paper reviews the development of copper-based Fenton systems over the past decade. It mainly involves the research and application of copper-based catalysts in different Fenton systems, including photo-Fenton, electro-Fenton, microwave-Fenton, and ultrasonic-Fenton. This review provides a fundamental reference for the subsequent studies of copper-based Fenton systems, contributing to the goal of transitioning these systems from laboratory research into practical environmental applications.

Enhancing Tranexamic Acid Penetration through AQP-3 Protein Triggering via ZIF-8 Encapsulation for Melasma and Rosacea Therapy.

The systemic use of tranexamic acid (TA) as an oral drug can bring adverse reactions, while intradermal injection leads to pain and a risk of infection. Moreover, it is difficult for highly hydrophilic TA to penetrate the skin barrier that contains lots of hydrophobic lipid compounds, which poses enormous restrictions on its topical application. Current transdermal TA delivery strategies are suffering from low drug load rates, plus their synthesis complexity, time-consumption, etc. adding to the difficulty of TA topical application in clinical therapeutics. To increase the penetration of TA, a novel approach using TA-loaded ZIF-8 (TA@ZIF-8) is developed. The encapsulation efficiency of TA@ZIF-8 reaches ≈25% through physical adsorption and chemical bonding of TA indicates by theoretical simulation and the improved TA penetration is elevated through activating the aquaporin-3 (AQP-3) protein. Additionally, in vivo and in vitro experiments demonstrate the preponderance of TA@ZIF-8 for penetration ability and the advantages in intracellular uptake, minor cytotoxicity, and inhibition of melanogenesis and inflammatory factors. Moreover, clinical trials demonstrate the safety and efficacy of TA@ZIF-8 in the treatment of melasma and rosacea. This work presents a potential topical application of TA, free from the safety concerns associated with systemic drug administration.

Systematic review of epidemiological and toxicological evidence on health effects of fluoride in drinking water.

INTRODUCTION: Fluoride is a naturally occurring substance that is also added to drinking water, dental hygiene products, and food supplements for preventing dental caries. Concerns have been raised about several other potential health risks of fluoride. OBJECTIVE: To conduct a robust synthesis of evidence regarding human health risks due to exposure to fluoride in drinking water, and to develop a point of departure (POD) for setting a health-based value (HBV) for fluoride in drinking water. METHODS: A systematic review of evidence published since recent reviews of human, animal, and in vitro data was carried out. Bradford Hill considerations were used to weigh the evidence for causality. Several key studies were considered for deriving PODs. RESULTS: The current review identified 89 human studies, 199 animal studies, and 10 major in vitro reviews. The weight of evidence on 39 health endpoints was presented. In addition to dental fluorosis, evidence was considered strong for reduction in IQ scores in children, moderate for thyroid dysfunction, weak for kidney dysfunction, and limited for sex hormone disruptions. CONCLUSION: The current review identified moderate dental fluorosis and reduction in IQ scores in children as the most relevant endpoints for establishing an HBV for fluoride in drinking water. PODs were derived for these two endpoints, although there is still some uncertainty in the causal weight of evidence for causality for reducing IQ scores in children and considerable uncertainty in the derivation of its POD. Given our evaluation of the overall weight of evidence, moderate dental fluorosis is suggested as the key endpoint until more evidence is accumulated on possible reduction of IQ scores effects. A POD of 1.56 mg fluoride/L for moderate dental fluorosis may be preferred as a starting point for setting an HBV for fluoride in drinking water to protect against moderate and severe dental fluorosis. Although outside the scope of the current review, precautionary concerns for potential neurodevelopmental cognitive effects may warrant special consideration in the derivation of the HBV for fluoride in drinking water.

A review of nanomaterials for biosensing applications.

A biosensor is a device that reacts with the analyte to be analyzed, detects its concentration, and generates readable information, which plays an important role in medical diagnosis, detection of physiological indicators, and disease prevention. Nanomaterials have received increasing attention in the fabrication and improvement of biosensors due to their unique physicochemical and optical properties. In this paper, the properties of nanomaterials such as the size effect, optical and electrical properties, and their advantages in the field of biosensing are briefly summarized, and the application of nanomaterials can effectively improve the sensitivity and reduce the detection limit of biosensors. The advantages of commonly used nanomaterials such as gold nanoparticles (AuNPs), carbon nanotubes (CNTs), quantum dots (QDs), graphene, and magnetic nanobeads for biosensor applications are also reviewed. Besides, the two main types of biosensors using nanomaterials involved in their construction and their working principles are described, and the toxicity and biocompatibility of nanomaterials and the future direction of nanomaterial biosensors are discussed.

Using a new human milk fortifier to optimize human milk feeding among very preterm and/or very low birth weight infants: a multicenter study in China.

BACKGROUND: Human milk fortifier (HMF) composition has been optimized recently. But clinical evidence of its safety and efficacy is limited in Chinese population. The aim of this study was to evaluate effects of a new HMF in growth, nutritional status, feeding intolerance, and major morbidities among very preterm (VPT) or very low birth weight (VLBW) infants in China. METHODS: VPT/VLBW infants admitted from March 2020 to April 2021 were prospectively included in the experimental (new HMF, nHMF) group, who received a new powdered HMF as a breast milk feeding supplement during hospitalization. Infants in the control group (cHMF) admitted from January 2018 to December 2019, were retrospective included, and matched with nHMF group infants for gestational age and birth weight. They received other kinds of commercially available HMFs. Weight gain velocity, concentrations of nutritional biomarkers, incidence of major morbidities, and measures of feeding intolerance were compared between the two groups. RESULTS: Demographic and clinical characteristics of infants in nHMF and cHMF groups were comparable. Weight gain velocity had no significant difference between the nHMF (14.0 ± 3.5 g/kg/d) and the cHMF group (14.2 ± 3.8 g/kg/d; P = 0.46). Incidence of morbidities, including necrotizing enterocolitis, bronchopulmonary dysplasia, retinopathy of prematurity, culture-confirmed sepsis, and feeding intolerance during hospitalization between nHMF and cHMF, were similar (all P-values > 0.05). The time to achieve full enteral feeding [13.5 (10, 21) days] in the nHMF group was significantly shorter than that in the cHMF group [17 (12, 23) days, HR = 0.67, 95%CI: 0.49, 0.92; P = 0.01]. Compared with cHMF group, the decrease of blood urea nitrogen level over time in nHMF group was smaller (ß = 0.6, 95%CI:0.1, 1.0; P = 0.01). CONCLUSIONS: The new HMF can promote growth of preterm infants effectively without increasing the incidence of major morbidity and feeding intolerance. It can be used feasible in Chinese VPT/VLBW infants. TRIAL REGISTRATION: This study was registered on ClinicalTrials.gov (NCT04283799).

Downregulated BMP-Smad1/5/8 signaling causes emphysema via dysfunction of alveolar type II epithelial cells.

Bone morphogenetic protein (BMP)-Smad1/5/8 signaling plays a crucial regulatory role in lung development and adult lung homeostasis. However, it remains elusive whether BMP-Smad1/5/8 signaling is involved in the pathogenesis of emphysema. In this study, we downregulated BMP-Smad1/5/8 signaling by overexpressing its antagonist Noggin in adult mouse alveolar type II epithelial cells (AT2s), resulting in an emphysematous phenotype mimicking the typical pathological features of human emphysema, including distal airspace enlargement, pulmonary inflammation, extracellular matrix remodeling, and impaired lung function. Dysregulation of BMP-Smad1/5/8 signaling in AT2s leads to inflammatory destruction dominated by macrophage infiltration, associated with reduced secretion of surfactant proteins and inhibition of AT2 proliferation and differentiation. Reactivation of BMP-Smad1/5/8 signaling by genetics or chemotherapy significantly attenuated the morphology and pathophysiology of emphysema and improved the lung function in Noggin-overexpressing lungs. We also found that BMP-Smad1/5/8 signaling was downregulated in cigarette smoke-induced emphysema, and that enhancing its activity in AT2s prevented or even reversed emphysema in the mouse model. Our data suggest that BMP-Smad1/5/8 signaling, located at the top of the signaling cascade that regulates lung homeostasis, represents a key molecular regulator of alveolar stem cell secretory and regenerative function, and could serve as a potential target for future prevention and treatment of pulmonary emphysema. © 2023 The Pathological Society of Great Britain and Ireland.

Fully integrated wearable microneedle biosensing platform for wide-range and real-time continuous glucose monitoring.

Wearable microneedle sensors for continuous glucose monitoring (CGM) have great potential for clinical impact by allowing access to large data sets to provide individualized treatment plans. To date, their development has been challenged by the accurate wide linear range tracking of interstitial fluid (ISF) glucose (Glu) levels. Here, we present a CGM platform consisting of a three-electrode microneedle electrochemical biosensor and a fully integrated radio-chemical analysis system. The long-term performance of the robust CGM on diabetic rats was achieved by electrodepositing Prussian blue (PB), and crosslinking glucose oxidase (GOx) and chitosan to form a 3D network using glutaraldehyde (GA). After redox by GOx, PB rapidly decomposes hydrogen peroxide and mediates charge transfer, while the 3D network and graphite powder provide enrichment and release sites for Glu and catalytic products, enabling a sensing range of 0.25-35 mM. Microneedle CGM has high sensitivity, good stability, and anti-interference ability. In diabetic rats, CGM can accurately monitor Glu levels in the ISF in real-time, which are highly consistent with levels measured by commercial Glu meters. These results indicate the feasibility and application prospects of the PB-based CGM for the clinical management of diabetes. STATEMENT OF SIGNIFICANCE: This study addresses the challenge of continuous glucose monitoring system design where the narrow linear range of sensing due to the miniaturization of sensors fails to meet the monitoring needs of clinical diabetic patients. This was achieved by utilizing a three-dimensional network of glutaraldehyde cross-linked glucose oxidase and chitosan. The unique topology of the 3D network provides a large number of sites for glucose enrichment and anchors the enzyme to the sensing medium and the conductive substrate through covalent bonding, successfully blocking the escape of the enzyme and the sensing medium and shortening the electron transfer and transmission path.

Biomass-derived 2D carbon materials: structure, fabrication, and application in electrochemical sensors.

Biomass, a renewable hydrocarbon, is one of the favorable sources of advanced carbon materials owing to its abundant resources and diverse molecular structures. Biomass-based two-dimensional carbon nanomaterials (2D-BC) have attracted extensive attention due to their tunable structures and properties, and have been widely used in the design and fabrication of electrochemical sensing platforms. This review embarks on the thermal conversion process of biomass from different sources and the synthesis strategy of 2D-BC materials. The affinity between 2D-BC structure and properties is emphasized. The recent progress in 2D-BC-based electrochemical sensors for health and environmental monitoring is also presented. Finally, the challenges and future development directions related to such materials are proposed in order to promote their further application in the field of electrochemical sensing.

Acoustofluidics for simultaneous droplet transport and centrifugation facilitating ultrasensitive biomarker detection.

Trace biological sample detection is critical for the analysis of pathologies in biomedicine. Integration of microfluidic manipulation techniques typically strengthens biosensing performance. For instance, using isothermal amplification reactions to sense trace miRNA in peripheral circulation lacks a sufficiently complex pretreatment process that limits the sensitivity of on-chip detection. Herein we propose an orthogonal tunable acoustic tweezer (OTAT) to simultaneously actuate the transportation and centrifugation of µ-droplets on a single device. The OTAT enables diversified modes of droplet transportation such as unidirectional transport, multi-direction transport, round-trip transport, tilt angle movement, multi-droplet fusion, and continuous centrifugation of the dynamic droplets simultaneously. The multiplicity of modalities enables the focusing of a loaded analyte at the center of the droplet or constant rotation about the center axis of the droplet. We herein demonstrate the OTAT's ability to actuate transportation, fusion, and centrifugation-based pretreatment of two biological sample droplets loaded with miRNA biomarkers and multiple mixtures, as well as facilitating the increase of fluorescence detection sensitivity by an order of magnitude compared to traditional tube reaction methods. The results herein demonstrate the OTAT-based droplet acoustofluidic platform's ability to combine a wide range of biosensing mechanisms and provide a higher accuracy of detection for one-stop point-of-care disease diagnosis.

Swirl-like Acoustofluidic Stirring Facilitates Microscale Reactions in Sessile Droplets.

Sessile droplets play a crucial role in the microreactors of biochemical samples. Acoustofluidics provide a non-contact and label-free method for manipulating particles, cells, and chemical analytes in droplets. In the present study, we propose a micro-stirring application based on acoustic swirls in sessile droplets. The acoustic swirls are formed inside the droplets by asymmetric coupling of surface acoustic waves (SAWs). With the merits of the slanted design of the interdigital electrode, the excitation position of SAWs is selective by sweeping in wide frequency ranges, allowing for the droplet position to be customized within the aperture region. We verify the reasonable existence of acoustic swirls in sessile droplets by a combination of simulations and experiments. The different periphery of the droplet meeting with SAWs will produce acoustic streaming phenomena with different intensities. The experiments demonstrate that acoustic swirls formed after SAWs encountering droplet boundaries will be more obvious. The acoustic swirls have strong stirring abilities to rapidly dissolve the yeast cell powder granules. Therefore, acoustic swirls are expected to be an effective means for rapid stirring of biomolecules and chemicals, providing a new approach to micro-stirring in biomedicine and chemistry.

Antibacterial black phosphorus nanosheets for biomedical applications.

Bacterial infections pose a significant threat to human health and a heavy burden on the global healthcare system. Antibiotics are the primary treatment, but they can lead to bacterial resistance and adverse side effects. Two-dimensional (2D) nanomaterials such as graphene, MoS2, and MXene have emerged as novel antibacterial agents due to their potential to circumvent bacterial resistance. Among the 2D nanomaterials, black phosphorus nanosheets (BPNs) have attracted great research interest due to their excellent biocompatibility. BPNs possess unique properties, such as a high specific surface area, tunable bandgap, and easy surface functionalization, enabling them to combat bacteria through physical disruption of bacterial membranes, photothermal and photodynamic therapies. However, the low preparation efficiency and inevitable oxidative degradation of BPNs have limited their wide application. This review provides a comprehensive overview of recent advances in antibacterial research on BPNs, encompassing their preparation methods, structural and physicochemical properties, antibacterial mechanisms, and potential applications. By addressing the challenges and prospects of using BPNs as an alternative to antibiotics, this review provides valuable insights and guidance for utilizing BPNs in shaping the future of antibacterial therapy.

Multi-modal recommendation algorithm fusing visual and textual features.

In recommender systems, the lack of interaction data between users and items tends to lead to the problem of data sparsity and cold starts. Recently, the interest modeling frameworks incorporating multi-modal features are widely used in recommendation algorithms. These algorithms use image features and text features to extend the available information, which alleviate the data sparsity problem effectively, but they also have some limitations. On the one hand, multi-modal features of user interaction sequences are not considered in the interest modeling process. On the other hand, the aggregation of multi-modal features often employs simple aggregators, such as sums and concatenation, which do not distinguish the importance of different feature interactions. In this paper, to tackle this, we propose the FVTF (Fusing Visual and Textual Features) algorithm. First, we design a user history visual preference extraction module based on the Query-Key-Value attention to model users' historical interests by using of visual features. Second, we design a feature fusion and interaction module based on the multi-head bit-wise attention to adaptively mine important feature combinations and update the higher-order attention fusion representation of features. We conduct experiments on the Movielens-1M dataset, and the experiments show that FVTF achieved the best performance compared with the benchmark recommendation algorithms.

Deep learning-based precise prediction and early detection of radiation-induced temporal lobe injury for nasopharyngeal carcinoma.

Background: Radiotherapy is the mainstay of treatment for nasopharyngeal carcinoma. Radiation-induced temporal lobe injury (TLI) can regress or resolve in the early phase, but it is irreversible at a later stage. However, no study has proposed a risk-based follow-up schedule for its early detection. Planning evaluation is difficult when dose-volume histogram (DVH) parameters are similar and optimization is terminated. Methods: This multicenter retrospective study included 6065 patients between 2014 and 2018. A 3D ResNet-based deep learning model was developed in training and validation cohorts and independently tested using concordance index in internal and external test cohorts. Accordingly, the patients were stratified into risk groups, and the model-predicted risks were used to develop risk-based follow-up schedules. The schedule was compared with the Radiation Therapy Oncology Group (RTOG) recommendation (every 3 months during the first 2 years and every 6 months in 3-5 years). Additionally, the model was used to evaluate plans with similar DVH parameters. Findings: Our model achieved concordance indexes of 0.831, 0.818, and 0.804, respectively, which outperformed conventional prediction models (all P < 0.001). The temporal lobes in all the cohorts were stratified into three groups with discrepant TLI-free survival. Personalized follow-up schedules developed for each risk group could detect TLI 1.9 months earlier than the RTOG recommendation. According to a higher median predicted 3-year TLI-free survival (99.25% vs. 99.15%, P < 0.001), the model identified a better plan than previous models. Interpretation: The deep learning model predicted TLI more precisely. The model-determined risk-based follow-up schedule detected the TLI earlier. The planning evaluation was refined because the model identified a better plan with a lower risk of TLI. Funding: The Sun Yat-sen University Clinical Research 5010 Program (2015020), Guangdong Basic and Applied Basic Research Foundation (2022A1515110356), Medical Scientific Research Foundation of Guangdong Province (A2022367), and Guangzhou Science and Technology Program (2023A04J1788).