A Novel Network Framework on Simultaneous Road Segmentation and Vehicle Detection for UAV Aerial Traffic Images.

Unmanned Aerial Vehicle (UAV) aerial sensors are an important means of collecting ground image data. Through the road segmentation and vehicle detection of drivable areas in UAV aerial images, they can be applied to monitoring roads, traffic flow detection, traffic management, etc. As well, they can be integrated with intelligent transportation systems to support the related work of transportation departments. Existing algorithms only realize a single task, while intelligent transportation requires the simultaneous processing of multiple tasks, which cannot meet complex practical needs. However, UAV aerial images have the characteristics of variable road scenes, a large number of small targets, and dense vehicles, which make it difficult to complete the tasks. In response to these issues, we propose to implement road segmentation and on-road vehicle detection tasks in the same framework for UAV aerial images, and we conduct experiments on a self-constructed dataset based on the DroneVehicle dataset. For road segmentation, we propose a new algorithm C-DeepLabV3+. The new algorithm introduces the coordinate attention (CA) module, which can obtain more accurate segmentation target location information and make the segmentation target edges more continuous. Also, the improved algorithm introduces the cascade feature fusion module to prevent the loss of detail information in road segmentation and to obtain better segmentation performance. For vehicle detection, we propose an improved algorithm S-YOLOv5 by adding a parameter-free lightweight attention module SimAM. Finally, the proposed road segmentation-vehicle detection framework is utilized to unite the C-DeepLabV3+ and S-YOLOv5 algorithms for the implementation of the serial tasks. The experimental results show that on the constructed ViDroneVehicle dataset, the C-DeepLabV3+ algorithm has an mPA value of 98.75% and an mIoU value of 97.53%, which can better segment the road area and solve the problem of occlusion. The mAP value of the S-YOLOv5 algorithm has an mAP value of 97.40%, which is more than YOLOv5's 96.95%, which effectively reduces the vehicle omission and false detection rates. By comparison, the results of both algorithms are superior to multiple state-of-the-art methods. The overall framework proposed in this paper has superior performance and is capable of realizing high-quality and high-precision road segmentation and vehicle detection from UAV aerial images.

Colorimetric detection of electrolyte ions in blood based on biphasic microdroplet extraction.

BACKGROUND: Timely diagnosis and prevention of diseases require rapid and sensitive detection of biomarkers from blood samples without external interference. Abnormal electrolyte ion levels in the blood are closely linked to various physiological disorders, including hypertension. Therefore, accurate, interference-free, and precise measurement of electrolyte ion concentrations in the blood is particularly important. RESULTS: In this work, a colorimetric sensor based on a biphasic microdroplet extraction is proposed for the detection of electrolyte ions in the blood. This sensor employs mini-pillar arrays to facilitate contact between adjacent blood microdroplets and organic microdroplets serving as sensing phases, with any color changes being monitored through a smartphone's colorimetric software. The sensor is highly resistant to interference and does not require pre-treatment of the blood samples. Remarkably, the sensor exhibits exceptional reliability and stability, allowing for rapid enrichment and detection of K+, Na+, and Cl- in the blood within 10 s (Cl-), 15 s (K+) and 40 s (Na+) respectively. SIGNIFICANCE: The colorimetric sensor based on biphasic microdroplet extraction offers portability due to its compact size and ease of operation without the need for large instruments. Additionally, it is location-independent, making it a promising tool for real-time biomarker detection in body fluids such as blood.

Hydrogel-Functionalized Bandages with Janus Wettability for Efficient Unidirectional Drug Delivery and Wound Care.

Chronic wounds have imposed a severe physical and economic burden on the global healthcare system, which are usually treated by the delivery of drugs or bioactive molecules to the wound bed through wound dressings. In this work, we have demonstrated a hydrogel-functionalized bandage with Janus wettability in a bilayer structure to achieve unidirectional drug delivery and multifunctional wound care. The Janus patterned bandage with porous gradient wetting channels on the upper layer is responsible for the unidirectional transport of the drug from the outside to the wound bed (up to 90% drug transport efficiency) while preventing drug diffusion in unwanted directions (<8%). The hydrogel composed of chitosan quaternary ammonium salt (HACC), poly(vinyl alcohol) (PVA), and poly(acrylic acid) (PAA) at the bottom layer further functionalized such a bandage with biocompatibility, excellent antibacterial properties, and hemostatic ability to promote wound healing. Especially, the hydrogel-functionalized bandage with Janus wettability exhibits excellent mechanical flexibility (∼198% strain), which can comply well with skin deformation (stretching, bending, or twisting) and maintain unidirectional drug delivery behavior without any leakage. The in vivo full-thickness skin wound model confirms that the hydrogel-functionalized bandage can significantly facilitate epithelialization and collagen deposition and improve drug delivery efficiency, thus promoting wound closure and healing (the wound healing ratio was 98.10% at day 15). Such a synergistic strategy of unidirectional drug delivery and multifunctional wound care provides a more efficient, economical, and direct method to promote wound healing, which could be used as a potential high-performance wound dressing for clinical application.

Emerging open-channel droplet arrays for biosensing.

Open-channel droplet arrays have attracted much attention in the fields of biochemical analysis, biofluid monitoring, biomarker recognition and cell interactions, as they have advantages with regard to miniaturization, parallelization, high-throughput, simplicity and accessibility. Such droplet arrays not only improve the sensitivity and accuracy of a biosensor, but also do not require sophisticated equipment or tedious processes, showing great potential in next-generation miniaturized sensing platforms. This review summarizes typical examples of open-channel microdroplet arrays and focuses on diversified biosensing integrated with multiple signal-output approaches (fluorescence, colorimetric, surface-enhanced Raman scattering (SERS), electrochemical, etc.). The limitations and development prospects of open-channel droplet arrays in biosensing are also discussed with regard to the increasing demand for biosensors.

Active Enrichment of Nanoparticles for Ultra-Trace Point-of-Care COVID-19 Detection.

Active enrichment can detect nucleic acid at ultra-low concentrations without relatively time-consuming polymerase chain reaction (PCR), which is an important development direction for future rapid nucleic acid detection. Here, we reported an integrated active enrichment platform for direct hand-held detection of nucleic acid of COVID-19 in nanoliter samples without PCR. The platform consists of a capillary-assisted liquid-carrying system for sampling, integrated circuit system for ultrasound output, and cell-phone-based surface-enhanced Raman scattering (SERS) system. Considering the acoustic responsiveness and SERS-enhanced performance, gold nanorods were selected for biomedical applications. Functionalized gold nanorods can effectively capture and enrich biomarkers under ultrasonic aggregation. Such approaches can actively assemble gold nanorods in 1-2 s and achieved highly sensitive (6.15 × 10-13 M) SERS detection of COVID-19 biomarkers in nanoliter (10-7 L) samples within 5 min. We further demonstrated the high stability, repeatability, and selectivity of the platform, and validated its potential for the detection of throat swab samples. This simple, portable, and ultra-trace integrated active enrichment detection platform is a promising diagnostic tool for the direct and rapid detection of COVID-19.

One-click investigation of shape influence of silver nanostructures on SERS performance for sensitive detection of COVID-19.

Sensitive and accurate detection of SARS-CoV-2 methods is meaningful for preventing and controlling the novel coronavirus. The detection techniques supporting portable, onsite, in-time, and online data transfer are urgently needed. Here, we one-click investigated the shape influence of silver nanostructures on SERS performance and their applications in the sensitive detection of SARS-CoV-2. Such investigation is achieved by adjusting multiple parameters (concentration, potential, and time) on the integrated electrochemical array, thus various morphologies (e.g., bulk, dendritic, globular, and spiky) can be one-click synthesized. The SERS performance results indicated that dendritic nanostructures are superior to the other three with an order of magnitude signal enhancement. Such on-electrode dendritic silver substrate also represents high sensitivity (LOD = 7.42 × 10-14 M) and high reproducibility (RSD = 3.67%) toward the SARS-CoV-2 RNA sequence detection. Such approach provides great potentials for rapid diagnosis and prevention of diverse infectious diseases.

End-to-End Network for Pedestrian Detection, Tracking and Re-Identification in Real-Time Surveillance System.

Surveillance video has been widely used in business, security, search, and other fields. Identifying and locating specific pedestrians in surveillance video has an important application value in criminal investigation, search and rescue, etc. However, the requirements for real-time capturing and accuracy are high for these applications. It is essential to build a complete and smooth system to combine pedestrian detection, tracking and re-identification to achieve the goal of maximizing efficiency by balancing real-time capture and accuracy. This paper combined the detector and Re-ID models into a single end-to-end network by introducing a new track branch to YOLOv5 architecture for tracking. For pedestrian detection, we employed the weighted bi-directional feature pyramid network (BiFPN) to enhance the network structure based on the YOLOv5-Lite, which is able to further improve the ability of feature extraction. For tracking, based on Deepsort, this paper enhanced the tracker, which uses the Noise Scale Adaptive (NSA) Kalman filter to track, and adds adaptive noise to strengthen the anti-interference of the tracking model. In addition, the matching strategy is further updated. For pedestrian re-identification, the network structure of Fastreid was modified, which can increase the feature extraction speed of the improved algorithm by leaps and bounds. Using the proposed unified network, the parameters of the entire model can be trained in an end-to-end method with the multi-loss function, which has been demonstrated to be quite valuable in some other recent works. Experimental results demonstrate that pedestrians detection can obtain a 97% mean Average Precision (mAP) and that it can track the pedestrians well with a 98.3% MOTA and a 99.8% MOTP on the MOT16 dataset; furthermore, high pedestrian re-identification performance can be achieved on the VERI-Wild dataset with a 77.3% mAP. The overall framework proposed in this paper has remarkable performance in terms of the precise localization and real-time detection of specific pedestrians across time, regions, and cameras.

Programmable Microparticle Array for In Situ Modification and Multiple miRNA Detection.

Simultaneous detection of multiple miRNAs of one disease can greatly reduce misdiagnosis and improve the detection rate, which is helpful for early cancer diagnosis. Here, a programmable microparticle-array-based acoustic microchip for in situ simultaneous multiple miRNAs detection is developed. On this microchip, the multiple probes-labeled microparticle array can be procedurally arranged in a microfluidic reaction chamber when four orthogonally piezoelectric transducers are applied. The probes-labeled microparticle array offers a platform for full molecular contact under dynamic ultrasonic streaming, and the array supplies a multipoint data correction to reduce the false positive of the detection results for more precisely visible fluorescence multiple target miRNAs sensing. We employed miRNA-21, miRNA-210, and miRNA-155 as specific biomarkers of pancreatic cancer and successfully finished the multiple miRNAs simultaneous detection in the microchip with a detection limit of 139.1, 179.9, and 111.4 pM, respectively. Such a device is programmable by adjusting the imputing frequency and voltage, and target biomarkers can be easily collected when the ultrasound force is released for further analysis, which shows great potential in multiple miRNAs enrichment and simultaneous detection for cancer clinical diagnosis.

Real-Time and Efficient Multi-Scale Traffic Sign Detection Method for Driverless Cars.

Traffic signs detection and recognition is an essential and challenging task for driverless cars. However, the detection of traffic signs in most scenarios belongs to small target detection, and most existing object detection methods show poor performance in these cases, which increases the difficulty of detection. To further improve the accuracy of small object detection for traffic signs, this paper proposed an optimization strategy based on the YOLOv4 network. Firstly, an improved triplet attention mechanism was added to the backbone network. It was combined with optimized weights to make the network focus more on the acquisition of channel and spatial features. Secondly, a bidirectional feature pyramid network (BiFPN) was used in the neck network to enhance feature fusion, which can effectively improve the feature perception field of small objects. The improved model and some state-of-the-art (SOTA) methods were compared on the joint dataset TT100K-COCO. Experimental results show that the enhanced network can achieve 60.4% mAP(Mean Average Precision), surpassing the YOLOv4 by 8% with the same input size. With a larger input size, it can achieve a best performance capability of 66.4% mAP. This work provides a reference for research on obtaining higher accuracy for traffic sign detection in autonomous driving.

Safety-oriented planning of expressway truck service areas based on driver demand.

The rapid development of the economy has promoted the growth of freight transportation. The truck service areas on expressways, as the main places for truck drivers to rest, play an important role in ensuring the driving safety of trucks. If these service areas are constructed densely or provide a plentiful supply of parking areas, they are costly to construct. However, if the distance between two adjacent truck service areas is very large or the number of truck parking spaces in service areas is small, the supply will fail to meet the parking needs of truck drivers. In this situation, the continuous working time of truck drivers will be longer, and this is likely to cause driver fatigue and even traffic accidents. To address these issues, this paper established a non-linear optimization model for truck service area planning of expressways to optimize truck driving safety. An improved genetic algorithm is proposed to solve the model. A case study of a 215.5-kilometers-length section of the Guang-Kun expressway in China was used to demonstrate the effectiveness of the model and algorithm. As validated by this specific case, the proposed model and solution algorithm can provide an optimal plan for the layout of truck service areas that meet the parking needs of truck drivers while minimizing the service loss rate. The research results of this paper can contribute to the construction of truck service areas and the parking management of trucks on expressways.

Ultra-trace enriching biosensing in nanoliter sample.

Rapid detection and accurate analysis of trace samples is an important prerequisite for precision medicine. Here we integrated capillary with ultrasound to induce biomarkers enrichment in nanoliter samples, and developed a nanoliter sample enrichment analysis method for ultra-trace miRNA biosensing. The interaction between ultrasonic field and capillary provides a gradient ultrasound field, which is essential for the aggregation of functionalized microspheres along with the enrichment of specific biomarkers. The results indicated that the enrichment of the biomarkers effectively enhanced the fluorescence intensity, and the limit of detection reaches 7.8✕10-12 M in 100 nL. Such integrated device can realize ultrasonic enrichment and visual analysis of target samples, and provides a new idea for rapid and highly sensitive detection of ultra-trace biomarkers in clinical diagnosis.

On-demand mixing and dispersion in mini-pillar based microdroplets.

The analysis and detection of ultra-trace biomarkers are often carried out in microliter droplets. Common stirring approaches have some difficulties in precise and contactless mixing and dispersion in microdroplets. In this work, an open mini-pillar-based platform that integrates with ultrasound units is developed to achieve contactless mixing and dispersion in microliter samples. On such a platform, mini-pillars can anchor microdroplets as individual microreactors, and each ultrasound unit can be remotely controlled to achieve on-demand contactless micro-stirring, which is also confirmed by mixing and dispersing of Fe3O4 nanoparticles (1 µm) in microdroplets (10 µL). Such on-demand high-throughput mixing and dispersion that integrates ultrasound mixing with microdroplet technology provides a potential robot-based platform for achieving high-throughput and ultra-trace biosensing in microliter droplets.

Droplet array for open-channel high-throughput SERS biosensing.

Open-channel and high throughput are two important aspects of clinical diagnosis, correlation biochemical analysis, cell culture techniques and food safety. Here, we propose the mini-pillar based array for open-channel and high-throughput SERS detection of miRNA. The polydimethylsiloxane (PDMS) mini-pillars are used as a high-throughput platform, which have good anchoring and aggregation effects on microdroplets, greatly reducing the amount of analytical solution and facilitate the homogeneous sample distribution after evaporation. The deposited gold nanorods (Au NRs) on the pillars with optimized diameter served as SERS-active substrate, can greatly improve the sensitivity of SERS signal compared to other planar substrates. On the open-channel biological chip, sensitive, simultaneous, and specific detection of breast cancer marker miRNA-1246 can be performed. In this mini-pillar array SERS system, the limit of detection (LOD) is 10-12 M. The mini-pillar array shows enormous potential for open channel, high-throughput biomolecular detection, providing an opportunity for biomedical point-of-care testing (POCT) and drug screening.

Integrated individually electrochemical array for simultaneously detecting multiple Alzheimer's biomarkers.

Simultaneous detection of multiple biomarkers is benefit for reducing the detection cycles, avoiding the false-positive signals, and providing the cross validation, which provide the opportunity to understand the pathogenic mechanisms and achieve precise early diagnosis. Here, we demonstrate the mini-pillar-based individual electrochemical array for simultaneous detection of multiple biomarkers. On such platform, the mini-pillar could confine the microdroplet as individual and open-channel microreactor, which is extremely helpful for reducing reagent consumption and extracting internal information, and the electrodes array embedded in mini-pillar are integrated on one side to achieve multiple and simultaneous electrochemical sensing. The introduction of gold nanodendrites by electrodeposition has greatly enhanced sensitivity via improving probe-binding capacity and response signals. Sensitive and selective detection of multiple Alzheimer's biomarkers including Tau, ApoE4, Amyloid-ß and miRNA-101 on such mini-pillar-based biosensor is also achieved. Such biosensor platform with the advantages of high-yield, high sensitivity, low-waste and multiple signals output shows great promise in sensing multiple biomolecules for disease diagnosis and health monitoring.

Integrated Smart Janus Textile Bands for Self-Pumping Sweat Sampling and Analysis.

Wearable sweat sensors have spearheaded the thrust toward personalized health monitoring with continuous, real-time, and molecular-level insight in a noninvasive manner. However, effective sweat sampling still remains a huge challenge. Here, we introduce an intelligent Janus textile band that bridges the gap between self-pumping sweat collection, comfortable epidemic microclimate, and sensitive electrochemical biosensing via an integrated wearable platform. The dominant sweat sampling configuration is a textile with Janus wettability, which is fabricated by electrospinning a hydrophobic polyurethane (PU) nanofiber array onto superhydrophilic gauze. Based on a contact-pumping model, the Janus textile can unidirectionally and thoroughly transport sweat from skin (hydrophobic side) to embedded electrode surface (hydrophilic side) with epidemic comfort. On-body experimentation reveals that the sensitive detection of multiple biomarkers including glucose, lactate, K+, and Na+ is achieved in the pumped sweat. Such smart Janus textile bands can effectively drain epidermal sweat to targeted assay sites via interface modifications, representing a reinforced and controlled biofluids analysis pathway with physiological comfort.

Integrating modification and detection in acoustic microchip for in-situ analysis.

Ultrasound as a biocompatible and powerful approach has been advanced in biotechnology. Here we present an acoustic microchip integrating modification and detection for in-situ analysis. Such microchip employs two pairs of piezoelectric transducers (PZTs) for acoustic field generation and a polydimethylsiloxane (PDMS) microcavity on a polyethylene terephthalate (PET) substrate for producing microparticle array. The applying of acoustic field results in rapidly forming microparticle array by adjusting the inputting frequency and voltage. In-situ modification and detection are accelerated due to the dynamic ultrasonic streaming around the ultrasound induced microparticle array. Such array also benefits from reducing the detection errors by coupling of multiple points. With this strategy, biomarkers (e.g. miRNA) can be enriched, and achieve in-situ modification and detection via simple two steps with excellent specificity. After the detection, samples are regained from the output channel by releasing the acoustic field, which is benefit for further analysis. Such integrated modification and detection acoustic microchip shows great potential in visual in-situ analysis and enriching ultratrace biomarkers for clinical diagnosis.

Microdroplet-captured tapes for rapid sampling and SERS detection of food contaminants.

Simple, effective, and rapid detection of chemically relevant hazards is a highly desirable research goal, which can provide early-warning information to improve the patient-care outcomes for public health. Here, we introduce a microdroplet-captured tape toward rapid SERS screening of food contaminants. The dominant sensing unit lies on functionalized microwell in conductive carbon tapes, which is simply prepared by physical punching, magnetron sputtering and electrochemical deposition of Au nanodendrites. The tape-based sensors not only possess highly branched Au nanodendrites in microwell for promoting SERS activity, but also enable anchoring the microdroplets via direct dip-pulling from pristine analytes solutions upon sticky incorporated on a glove. Early-warning SERS detection of food contaminants including Sudan-1, thiram, and thiabendazole from the real samples can be achieved by such simple sampling method. These tape-based sensors with a facile operation module and accessible signal read-out represent an innovative point-of-care testing (POCT) device for forensic, military, consumer protection, environmental monitoring, and food safety applications.

Mini-pillar microarray for individually electrochemical sensing in microdroplets.

High throughput and high sensitivity are two important aspects in multiple biomarker recognition, drug discovery and relevant biochemical sensing. Here, we integrate mini-pillar microarray with the circuit components toward high-throughput individual electrochemical sensing in microdroplets. On such droplet-microarray-based electrochemical platform, the high adhesion of the mini-pillar can hold a microdroplet (hundreds nanoliter to a few microliter) regardless of rotation and deformation. Each pillar as a unit has a three-electrode to achieve individual electrochemical sensing, and electrodes are integrated on one side to achieve the sequential electrochemical read-out. Qualitative and quantitative electrochemical assessments of multiple glucose concentrations in individual microdroplets are also achieved. Such mini-pillar-based individual electrochemical platform shows great potential in high-throughput and high-sensitive biomolecular recognitions, provides an opportunity to develop miniaturized sensing platform for emerging biological and pathological applications.

Nanodendritic gold/graphene-based biosensor for tri-mode miRNA sensing.

A nanodendritic gold/graphene-based biosensor that can perform fluorescence, SERS and electrochemical tri-modal miRNA detection in a single microdroplet has been developed. The biosensor was used to successfully perform tri-modal quantitative trace miRNA-375 detection, which enormously reduces false positive readings caused by interference and ambiguous signals, and has significant implications for use in precise physiological and pathological diagnosis.

Superwettable nanodendritic gold substrates for direct miRNA SERS detection.

By combining a superwettable interface with a nanodendritic gold structure, we have fabricated a superwettable nanodendritic gold substrate for direct SERS detection of multiple concentrations of miRNAs. The nanodendritic gold substrate provides numerous hotspots for Raman signal enhancement, and the superwettable interface ensures the immobilization of droplets in superhydrophilic microwells, which hold great potentials for applications in disease diagnostics.