TMH: Two-Tower Multi-Head Attention neural network for CTR prediction.

Click-through rate (CTR) prediction is a term used to predict the probability of a user clicking on an ad or item and has become a popular research area in advertising. As the volume of Internet data increases, the labor costs of traditional feature engineering continue to rise. To reduce the dependence on feature interactions, this paper proposes a fusion model that combines explicit and implicit feature interactions, called the Two-Tower Multi-Head Attention Neural Network (TMH) approach. The model integrates multiple components such as multi-head attention, residual network, and deep neural networks into an end-to-end model that automatically obtains vector-level combinations of explicit and implicit features to predict click-through rates through higher-order explicit and implicit interactions. We evaluated the effectiveness of TMH in CTR prediction through numerous experiments using three real datasets. The results demonstrate that our proposed method not only outperforms existing prediction methods but also offers good interpretability.

Wireless and battery-free wearable biosensing of riboflavin in sweat for precision nutrition.

Nutrition assessment is crucial for dietary guidance and prevention of malnutrition. Recent endeavors in wearable biochemical sensors have enabled real-time, in situ analysis of nutrients in sweat. However, the monitoring of riboflavin, an indispensable vitamin B involved in energy metabolism, remains challenging due to its trace level and variations in the sweat matrix. Herein, we report a wireless, battery-free, and flexible wearable biosensing system for the in situ monitoring of sweat riboflavin. Highly sensitive and selective electrochemical voltammetric detection is realized based on the synergistic effect of electrodeposited reduced graphene oxide (rGO) and platinum nanoparticles (PtNPs) with a low detection limit of 1.2 nM. The fully integrated system is capable of sweat sampling with the microfluidic patch, real-time riboflavin analysis and pH calibration with the flexible electrode array, as well as wirelessly simultaneous near field communication (NFC) energy harvesting and data transmission with the flexible circuit and a smartphone. On-body human sweat analysis demonstrates high accuracy cross-validated with gold-standard measurements, and reveals a strong correlation between sweat and urine riboflavin levels. The proposed wearable platform opens up attractive possibilities for noninvasive nutrient tracking, providing strong potential for personalized dietary guidance towards precision nutrition.

Smartphone-based portable photoelectrochemical biosensing system for point-of-care detection of urine creatinine and albumin.

Creatinine and albumin are crucial biomarkers for health monitoring and their ratio in urine is an effective approach for albuminuria assessment. Herein, to address the challenges of point-of-care and efficient analysis of the biomarkers simultaneously, we developed a fully integrated handheld smartphone-based photoelectrochemical biosensing system. A miniaturized printed circuit board included a potentiostat for photocurrent measurements and single-wavelength light-emitting diodes (LEDs) for photo-excitation, which was controlled with a Bluetooth-enabled smartphone. Graphitic carbon nitride (g-C3N4)/chitosan nanocomposites were modified on a transparent indium tin oxide (ITO) electrode as photoactive materials. Creatinine was detected through chelate formation with copper ion probes, while albumin was recognized specifically by an antigen-antibody reaction based on immunoassay. The biosensing system demonstrated good linearity and high sensitivity, with detection ranges of 100 µg mL-1 to 1500 µg mL-1 for creatinine, and 9.9 µg mL-1 to 500 µg mL-1 for albumin. Spiked artificial urine samples with different concentrations were tested to confirm the practical validity of the biosensing system, where an acceptable recovery rate ranged from 98.7% to 105.3%. This portable photoelectrochemical biosensing platform provides a convenient and cost-effective method for biofluid analysis, which has an extensive prospect in point-of-care testing (POCT) for mobile health.

Wireless, noninvasive therapeutic drug monitoring system for saliva measurement toward medication management of schizophrenia.

As an efficient patient management tool of precision medicine, decentralized therapeutic drug monitoring (TDM) provides new vision for therapy adherence and health management of schizophrenia in a convenient manner. To dispense with psychologically burdensome blood sampling and to achieve real-time, noninvasive, and continual circulating tracking of drugs with narrow therapeutic window, we study the temporal metabolism of clozapine, an antipsychotic with severe side effect, in rat saliva by a wireless, integrated and patient-friendly smart lollipop sensing system. Highly sensitive and efficient sensing performance with acceptable anti-biofouling property was realized based on the synergistic effect of electrodeposited reduced graphene oxide and ionic liquids in pretreatment-free saliva with low detection limit and good accuracy cross-validated with conventional method. On this basis, continual salivary drug levels with distinctive pharmacokinetics were found in different routes of drug administration. Pilot experiment reveals a strong correlation between blood and saliva clozapine and a positive relationship between drug dosage and salivary drug level, indicating potential applications presented by noninvasive saliva analysis towards patient-centered and personalized pharmacotherapy and adherence management via proposed smart lollipop system.

All-MXene-Printed RF Resonators as Wireless Plant Wearable Sensors for In Situ Ethylene Detection.

Printed flexible electronics have emerged as versatile functional components of wearable intelligent devices that bridge the digital information networks with biointerfaces. Recent endeavors in plant wearable sensors provide real-time and in situ insights to study phenotyping traits of crops, whereas monitoring of ethylene, the fundamental phytohormone, remains challenging due to the lack of flexible and scalable manufacturing of plant wearable ethylene sensors. Here the all-MXene-printed flexible radio frequency (RF) resonators are presented as plant wearable sensors for wireless ethylene detection. The facile formation of additive-free MXene ink enables rapid, scalable manufacturing of printed electronics, demonstrating decent printing resolution (2.5% variation), ≈30000 S m-1 conductivity and mechanical robustness. Incorporation of MXene-reduced palladium nanoparticles (MXene@PdNPs) facilitates 1.16% ethylene response at 1 ppm with 0.084 ppm limit of detection. The wireless sensor tags are attached on plant organ surfaces for in situ and continuously profiling of plant ethylene emission to inform the key transition of plant biochemistry, potentially extending the application of printed MXene electronics to enable real-time plant hormone monitoring for precision agriculture and food industrial management.

High-Performance Stretchable Thermoelectric Generator for Self-Powered Wearable Electronics.

Wearable thermoelectric generators (TEGs), which can convert human body heat to electricity, provide a promising solution for self-powered wearable electronics. However, their power densities still need to be improved aiming at broad practical applications. Here, a stretchable TEG that achieves comfortable wearability and outstanding output performance simultaneously is reported. When worn on the forehead at an ambient temperature of 15 °C, the stretchable TEG exhibits excellent power densities with a maximum value of 13.8 µW cm-2 under the breezeless condition, and even as high as 71.8 µW cm-2 at an air speed of 2 m s-1 , being one of the highest values for wearable TEGs. Furthermore, this study demonstrates that this stretchable TEG can effectively power a commercial light-emitting diode and stably drive an electrocardiogram module in real-time without the assistance of any additional power supply. These results highlight the great potential of these stretchable TEGs for power generation applications.

MXene-based wireless facemask enabled wearable breath acetone detection for lipid metabolic monitoring.

Breath acetone (BrAC) detection presents a promising scheme for noninvasive monitoring of metabolic health due to its close correlation to diets and exercise-regulated lipolysis. Herein, we report a Ti3C2Tx MXene-based wireless facemask for on-body BrAC detection and real-time tracking of lipid metabolism, where Ti3C2Tx MXene serves as a versatile nanoplatform for not only acetone detection but also breath interference filtration. The incorporation of in situ grown TiO2 and short peptides with Ti3C2Tx MXene further improves the acetone sensitivity and selectivity, while TiO2-MXene interfaces facilitate light-assisted response calibration. To further realize wearable breath monitoring, a miniaturized flexible detection tag has been integrated with a commercially available facemask, which enables facile BrAC detection and wireless data transmission. Through the hierarchically designed filtration-detection-calibration-transmission system, we realize BrAC detection down to 0.31 ppm (part per million) in breath. On-body breath tests validate the facemask in dynamically monitoring of lipid metabolism, which could guide dieter, athletes, and fitness enthusiasts to arrange diets and exercise activities. The proposed wearable platform opens up new possibility toward the practice of breath analysis as well as daily lipid metabolic management.

Dementia Research on Facebook and Twitter: Current Practice and Challenges.

BACKGROUND: Social media is a powerful tool for engaging diverse audiences in dementia research. However, there is little data summarizing current content exchange in this context. OBJECTIVE: To inform ethical dementia research engagement on social media, we characterized current practices by analyzing public social media posts. METHODS: We retrieved Facebook (2-year period, N = 7,896) and Twitter (1-year period, N = 9,323) posts containing dementia research-related keywords using manual and machine learning-based search strategies. We performed qualitative and quantitative content and sentiment analyses on random samples (10%) of the posts. RESULTS: Top Facebook users were advocacy (45%) and health organizations (25%). On Twitter, academics/researchers were the largest user group. Prevention was the most frequently coded theme (Facebook 30%; Twitter 26%), followed by treatment (Facebook 15%; Twitter 18%). Diagnostics had the highest Facebook engagement. Sharing knowledge was the primary form of content exchange (Facebook 63%; Twitter 80%). Most shared journal articles were peer-reviewed and open access. Emotional tone was overall more positive on Facebook. Justice was a prominent ethics topic regarding inequalities related to identity and intersecting modes of marginalization in dementia research. CONCLUSION: The findings indicate the importance of social media as an engagement tool of current topics in health research and reveal areas of potential for increased engagement. These data can inform consensus-based best practices for ethical social media application in dementia research.

Elimination of oxygen interference in the photoelectrochemical sensor with ferricyanide shield oxygen reduction for point of care testing.

As a sensitive and promising detection method, photoelectrochemical (PEC) sensor has been widely used in biochemical analysis field. However, the interferences from environment, especially dissolved oxygen, often impact the stability and precision of PEC sensors, which limit its practical applications. Here, we report a dissolved oxygen insensitive PEC sensor based on a proposed indirect electron transfer model. Through the detailed study of the charge transfer process, we determined that the photocurrent mainly comes from the electrochemical reaction between the photochemical products and the electrode, rather than direct charge transfer between the photoelectric materials and the electrode. The newly designed PEC sensor used ferricyanide to shield oxygen reduction and eliminated the influence from variable oxygen solubility. This sensor maintained robust responses over an extremely wide range (1.0-7.5 mg/L) of dissolved oxygen concentrations. To further demonstrate its capability, a smartphone based portable immunosensor was constructed for the detection of human serum albumin (HSA), which exhibited excellent stability and accuracy. The relative error of current was reduced by 81.3% over traditional electron donor solution. This work effectively improves the stability of PEC sensors, and lays the foundation for the subsequently practical applications of PEC sensor in point-of-care testing.

Smartphone-based photoelectrochemical biosensing system with graphitic carbon nitride/gold nanoparticles modified electrodes for matrix metalloproteinase-2 detection.

Photoelectrochemical analysis has been widely used in the field of biosensing due to its high sensitivity and strong anti-interference ability. Herein, a portable and versatile smartphone-based photoelectrochemical biosensing platform was developed for the rapid and on-site biomedical analysis. In the system, light excitation and photocurrent measurements were accomplished by a miniaturized and integrated circuit board. Smartphone with a specifically designed application was utilized to wirelessly control the system via Bluetooth. For photoelectrochemical sensor, graphitic carbon nitride (g-C3N4) and gold nanoparticles loaded on indium tin oxide electrodes were utilized as photoactive materials and signal amplification elements, respectively. The gold nanoparticles were also used to immobilized matrix metalloproteinase-2 (MMP-2) specific cleavage peptide that modified with bovine serum albumin (BSA) on the terminal. In the presence of MMP-2, the peptide was specifically hydrolyzed and cleaved. Thus, parts of the peptide chain and BSA were detached from the electrode resulting in the decrease of steric hindrance and the increase of photoelectrochemical currents. The photocurrents changed linearly with the logarithm of MMP-2 concentrations ranging from 1 pg/mL to 100 ng/mL in both buffer and artificial serum with correlation coefficient of 0.9943 and 0.9698. The limit of detections were as low as 0.48 pg/mL in buffer and 0.55 pg/mL in artifical serum. It indicated that the biosensor has good linearity and high sensitivity, which also verified the effectiveness of the portable instrument. This system provides a pioneering solution for the development of miniaturized and portable photoelectrochemical analysis instruments used for the field monitoring of different analytes.

Downregulation of ceramide synthase 1 promotes oral cancer through endoplasmic reticulum stress.

C18 ceramide plays an important role in the occurrence and development of oral squamous cell carcinoma. However, the function of ceramide synthase 1, a key enzyme in C18 ceramide synthesis, in oral squamous cell carcinoma is still unclear. The aim of our study was to investigate the relationship between ceramide synthase 1 and oral cancer. In this study, we found that the expression of ceramide synthase 1 was downregulated in oral cancer tissues and cell lines. In a mouse oral squamous cell carcinoma model induced by 4-nitroquinolin-1-oxide, ceramide synthase 1 knockout was associated with the severity of oral malignant transformation. Immunohistochemical studies showed significant upregulation of PCNA, MMP2, MMP9, and BCL2 expression and downregulation of BAX expression in the pathological hyperplastic area. In addition, ceramide synthase 1 knockdown promoted cell proliferation, migration, and invasion in vitro. Overexpression of CERS1 obtained the opposite effect. Ceramide synthase 1 knockdown caused endoplasmic reticulum stress and induced the VEGFA upregulation. Activating transcription factor 4 is responsible for ceramide synthase 1 knockdown caused VEGFA transcriptional upregulation. In addition, mild endoplasmic reticulum stress caused by ceramide synthase 1 knockdown could induce cisplatin resistance. Taken together, our study suggests that ceramide synthase 1 is downregulated in oral cancer and promotes the aggressiveness of oral squamous cell carcinoma and chemotherapeutic drug resistance.

Epidemiologic relationship between periodontitis and type 2 diabetes mellitus.

BACKGROUND: To systematically review the epidemiologic relationship between periodontitis and type 2 diabetes mellitus (T2DM). METHODS: Four electronic databases were searched up until December 2018. The manual search included the reference lists of the included studies and relevant journals. Observational studies evaluating the relationship between T2DM and periodontitis were included. Meta-analyses were conducted using STATA. RESULTS: A total of 53 observational studies were included. The Adjusted T2DM prevalence was significantly higher in periodontitis patients (OR = 4.04, p = 0.000), and vice versa (OR = 1.58, p = 0.000). T2DM patients had significantly worse periodontal status, as reflected in a 0.61 mm deeper periodontal pocket, a 0.89 mm higher attachment loss and approximately 2 more lost teeth (all p = 0.000), than those without T2DM. The results of the cohort studies found that T2DM could elevate the risk of developing periodontitis by 34% (p = 0.002). The glycemic control of T2DM patients might result in different periodontitis outcomes. Severe periodontitis increased the incidence of T2DM by 53% (p = 0.000), and this result was stable. In contrast, the impact of mild periodontitis on T2DM incidence (RR = 1.28, p = 0.007) was less robust. CONCLUSIONS: There is an evident bidirectional relationship between T2DM and periodontitis. Further well-designed cohort studies are needed to confirm this finding. Our results suggest that both dentists and physicians need to be aware of the strong connection between periodontitis and T2DM. Controlling these two diseases might help prevent each other's incidence.

Sensitive Detection of Exosomal Proteins via a Compact Surface Plasmon Resonance Biosensor for Cancer Diagnosis.

Exosomes are small extracellular vesicles released by cells for cell-cell communication. They play important roles in cancer development, metastasis, and drug resistance. Exosomal proteins have been demonstrated by many studies as promising biomarkers for cancer screening, diagnosis, and monitoring. Among many detection techniques, surface plasmon resonance (SPR) is a highly sensitive, label-free, and real-time optical detection method. Commercial prism-based wavelength/angular-modulated SPR sensors afford high sensitivity and resolution, but their large footprint and high cost limit their adaptability for clinical settings. Recently, a nanoplasmonic exosome (nPLEX) assay was developed to detect exosomal proteins for ovarian cancer diagnosis. However, comparing with conventional SPR biosensors, the broad applications of nanoplasmonic biosensors are limited by the difficult and expensive fabrication of nanostructures. We have developed an intensity-modulated, compact SPR biosensor (25 cm × 10 cm × 25 cm) which uses a conventional SPR sensing mechanism and does not require nanostructure fabrication. Calibration from glycerol showed that the compact SPR biosensor offered sensitivity of 9.258 × 103%/RIU and resolution of 8.311 × 10-6 RIU. We have demonstrated the feasibility of the compact SPR biosensor in lung cancer diagnosis using exosomal epidermal growth factor receptor (EGFR) and programmed death-ligand 1 (PD-L1) as biomarkers. It detected a higher level of exosomal EGFR from A549 nonsmall cell lung cancer (NSCLC) cells than BEAS-2B normal cells. With human serum samples, the compact SPR biosensor detected similar levels of exosomal EGFR in NSCLC patients and normal controls, and higher expression of exosomal PD-L1 in NSCLC patients than normal controls. The compact SPR biosensor showed higher detection sensitivity than ELISA and similar sensing accuracy as ELISA. It is a simple and user-friendly sensing platform, which may serve as an in vitro diagnostic test for cancer.