Novel benzothiazole-based fluorescent probe for efficient detection of Cu2+/S2- and Zn2+ and its applicability in cell imaging.

BACKGROUND: In recent years, environmental pollution has been increasing due to the excessive emission of toxic ions, which has caused serious harm to human health and ecological environment. There are various methods for detecting Cu2+, S2- and Zn2+, but the traditional ion detection methods have obvious disadvantages, such as poor selectivity and long detection time. Therefore, it is still crucial to develop simple, efficient and rapid detection methods. RESULTS: A fluorescent probe based on benzothiazole, (E)-N'-(3-(benzo[d]thiazol-2-yl)-2-hydroxy-5-methylbenzylidene)-3,4,5-tris(benzyloxy)benzohydrazide (BT), was designed and synthesized. It was characterized using ESI-MS, 1H NMR, and 13C NMR. BT can be used as a chemosensor to detect Cu2+, S2- and Zn2+ in CH3CN/H2O (7:3, v/v, pH = 7.4, HEPES buffer: 0.1 M), with detection limits of 0.301 µM, 0.017 µM, and 0.535 µM, respectively. At an excitation wavelength of 320 nm, BT exhibits an "on-off-on" response to Cu2+/S2- and enhanced fluorescence response to Zn2+, with a change in fluorescence color from orange to green. The coordination ratio of ions to the probe was determined to be 1:1 through Job's plot and hydrogen spectral titration. The recognition mechanism was discussed in conjunction with theoretical calculations. Furthermore, the probe has been successfully used in test strips and medical swabs colorimetry, as well as live cell imaging. SIGNIFICANCE: The probe BT lays the foundation for the design and synthesis of multifunctional fluorescent probes. As a portable detection method, probe BT was used to detect Cu2+, S2- and Zn2+ on strips. Furthermore, the probe was applied to biological cells to detect target ions with low cytotoxicity and excellent cell permeability. This indicating that it can be used as a potential candidate for tracking Cu2+ and S2- in clinical diagnostics and biological systems.

High-resolution flexible iontronic skins for both negative and positive pressure measurement in room temperature wind tunnel applications.

Flexible skins that can be laminated on curved surfaces are a desired form for wind pressure measurement. Sensors for such applications need to detect both negative and positive wind pressure with ultrahigh pressure resolution over a wide range up to 600 kPa, whereas existing flexible sensors are unsatisfactory to such demands. Here, we report a flexible skin containing two iontronic pressure sensors for negative pressure and positive pressure sensing, respectively, and show the potential of the skin for pressure measurement in room temperature wind tunnels. We control the contact state of iontronic interface of a sensor to introduce a pre-pressure that enables negative pressure sensing, exhibiting a pressure resolution of -20 Pa (0.025%) within a broad pressure regime (from -100 kPa to -10 Pa). The other sensor for positive pressure sensing exhibits a pressure-resolution of 100 Pa (0.025%) over 600 kPa, in addition to a wide frequency bandwidth up to 400 Hz. The skin with the two types of sensors can be attached on curved surfaces of wing surface for pressure measurement at various free stream velocities and angles of attack. This study provides a promising technology of using flexible skins for pressure measurement in aviation applications.

Data-driven inverse design of flexible pressure sensors.

Artificial skins or flexible pressure sensors that mimic human cutaneous mechanoreceptors transduce tactile stimuli to quantitative electrical signals. Conventional trial-and-error designs for such devices follow a forward structure-to-property routine, which is usually time-consuming and determines one possible solution in one run. Data-driven inverse design can precisely target desired functions while showing far higher productivity, however, it is still absent for flexible pressure sensors because of the difficulties in acquiring a large amount of data. Here, we report a property-to-structure inverse design of flexible pressure sensors, exhibiting a significantly greater efficiency than the conventional routine. We use a reduced-order model that analytically constrains the design scope and an iterative "jumping-selection" method together with a surrogate model that enhances data screening. As an exemplary scenario, hundreds of solutions that overcome the intrinsic signal saturation have been predicted by the inverse method, validating for a variety of material systems. The success in property design on multiple indicators demonstrates that the proposed inverse design is an efficient and powerful tool to target multifarious applications of flexible pressure sensors, which can potentially advance the fields of intelligent robots, advanced healthcare, and human-machine interfaces.

Enhanced hyaline cartilage formation and continuous osteochondral regeneration via 3D-Printed heterogeneous hydrogel with multi-crosslinking inks.

The unique gradient structure and complex composition of osteochondral tissue pose significant challenges in defect regeneration. Restoration of tissue heterogeneity while maintaining hyaline cartilage components has been a difficulty of an osteochondral tissue graft. A novel class of multi-crosslinked polysaccharide-based three-dimensional (3D) printing inks, including decellularized natural cartilage (dNC) and nano-hydroxyapatite, was designed to create a gradient scaffold with a robust interface-binding force. Herein, we report combining a dual-nozzle cross-printing technology and a gradient crosslinking method to create the scaffolds, demonstrating stable mechanical properties and heterogeneous bilayer structures. Biofunctional assessments revealed the remarkable regenerative effects of the scaffold, manifesting three orders of magnitude of mRNA upregulation during chondrogenesis and the formation of pure hyaline cartilage. Transcriptomics of the regeneration site in vivo and scaffold cell interaction tests in vitro showed that printed porous multilayer scaffolds could form the correct tissue structure for cell migration. More importantly, polysaccharides with dNC provided a hydrophilic microenvironment. The microenvironment is crucial in osteochondral regeneration because it could guide the regenerated cartilage to ensure the hyaline phenotype.

Displacement-pressure biparametrically regulated softness sensory system for intraocular pressure monitoring.

High intraocular pressure (IOP) is one of the high-risk pathogenic factors of glaucoma. Existing methods of IOP measurement are based on the direct interaction with the cornea. Commercial ophthalmic tonometers based on snapshot measurements are expensive, bulky, and their operation requires trained personnel. Theranostic contact lenses are easy to use, but they may block vision and cause infection. Here, we report a sensory system for IOP assessment that uses a soft indentor with two asymmetrically deployed iontronic flexible pressure sensors to interact with the eyelid-eyeball in an eye-closed situation. Inspired by human fingertip assessment of softness, the sensory system extracts displacement-pressure information for soft evaluation, achieving high accuracy IOP monitoring (>96%). We further design and custom-make a portable and wearable ophthalmic tonometer based on the sensory system and demonstrate its high efficacy in IOP screening. This sensory system paves a way towards cost-effective, robust, and reliable IOP monitoring.

Dual cross-linked COL1/HAp bionic gradient scaffolds containing human amniotic mesenchymal stem cells promote rotator cuff tendon-bone interface healing.

The tendon-bone interface heals through scar tissue, while the lack of a natural interface gradient structure and collagen fibre alignment leads to the occurrence of retearing. Therefore, the promotion of tendon healing has become the focus of regenerative medicine. The purpose of this study was to develop a gradient COL1/ hydroxyapatite (HAp) biomaterial loaded with human amniotic mesenchymal stem cells (hAMSCs). The performance of common cross-linking agents, Genipin, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide/N-hydroxysuccinimide (EDC/NHS), and dual cross-linked materials were compared to select the best cross-linking mechanism to optimize the biological and mechanical properties of the scaffold. The optimal COL1/HAp-loaded with hAMSCs were implanted into the tendon-bone rotator cuff interfaces in rats and the effect on the tendon-bone healing was assessed by micro-CT, histological analysis, and biomechanical properties. The results showed that Genipin and EDC/NHS dual cross-linked COL1/HAp had good biological activity and mechanical properties and promoted the proliferation and differentiation of hAMSCs. Animal experiments showed that the group using a scaffold loaded with hAMSCs had excellent continuity and orientation of collagen fibers, increased fibrocartilage and bone formation, and significantly higher biomechanical functions than the control group at the interface at 12 weeks post operation. This study demonstrated that dual cross-linked gradient COL1/HAp-loaded hAMSCs could promote interface healing, thereby providing a feasible strategy for tendon-bone interface regeneration.

Type II collagen scaffolds repair critical-sized osteochondral defects under induced conditions of osteoarthritis in rat knee joints via inhibiting TGF-ß-Smad1/5/8 signaling pathway.

The bidirectional relationship between osteochondral defects (OCD) and osteoarthritis (OA), with each condition exacerbating the other, makes OCD regeneration in the presence of OA challenging. Type II collagen (Col2) is important in OCD regeneration and the management of OA, but its potential applications in cartilage tissue engineering are significantly limited. This study investigated the regeneration capacity of Col2 scaffolds in critical-sized OCDs under surgically induced OA conditions and explored the underlying mechanisms that promoted OCD regeneration. Furthermore, the repair potential of Col2 scaffolds was validated in over critical-sized OCD models. After 90 days or 150 days since scaffold implantation, complete healing was observed histologically in critical-sized OCD, evidenced by the excellent integration with surrounding native tissues. The newly formed tissue biochemically resembled adjacent natural tissue and exhibited comparable biomechanical properties. The regenerated OA tissue demonstrated lower expression of genes associated with cartilage degradation than native OA tissue but comparable expression of genes related to osteochondral anabolism compared with normal tissue. Additionally, transcriptome and proteome analysis revealed the hindrance of TGF-ß-Smad1/5/8 in regenerated OA tissue. In conclusion, the engrafting of Col2 scaffolds led to the successful regeneration of critical-sized OCDs under surgically induced OA conditions by inhibiting the TGF-ß-Smad1/5/8 signaling pathway.

Construction of a Decellularized Multicomponent Extracellular Matrix Interpenetrating Network Scaffold by Gelatin Microporous Hydrogel 3D Cell Culture System.

Interface tissue repair requires the construction of biomaterials with integrated structures of multiple protein types. Hydrogels that modulate internal porous structures provide a 3D microenvironment for encapsulated cells, making them promise for interface tissue repair. Currently, reduction of intrinsic immunogenicity and increase of bioactive extracellular matrix (ECM) secretion are issues to be considered in these materials. In this study, gelatin methacrylate (GelMA) hydrogel is used to encapsulate chondrocytes and construct a phase transition 3D cell culture system (PTCC) by utilizing the thermosensitivity of gelatin microspheres to create micropores within the hydrogel. The types of bioactive extracellular matrix protein formation by chondrocytes encapsulated in hydrogels are investigated in vitro. After 28 days of culture, GelMA PTCC forms an extracellular matrix predominantly composed of collagen type II, collagen type I, and fibronectin. After decellularization, the protein types and mechanical properties are well preserved, fabricating a decellularized tissue-engineered extracellular matrix and GelMA hydrogel interpenetrating network hydrogel (dECM-GelMA IPN) consisting of GelMA hydrogel as the first-level network and the ECM secreted by chondrocytes as the second-level network. This material has the potential to mediate the repair and regeneration of tendon-bone interface tissues with multiple protein types.

Dietary flavonoids-microbiota crosstalk in intestinal inflammation and carcinogenesis.

Colorectal cancer (CRC) is currently the third leading cancer and commonly develops from chronic intestinal inflammation. A strong association was found between gut microbiota and intestinal inflammation and carcinogenic risk. Flavonoids, which are abundant in vegetables and fruits, can inhibit inflammation, regulate gut microbiota, protect gut barrier integrity, and modulate immune cell function, thereby attenuating colitis and preventing carcinogenesis. Upon digestion, about 90% of flavonoids are transported to the colon without being absorbed in the small intestine. This phenomenon increases the abundance of beneficial bacteria and enhances the production of short-chain fatty acids. The gut microbe further metabolizes these flavonoids. Interestingly, some metabolites of flavonoids play crucial roles in anti-inflammation and anti-tumor effects. This review summarizes the modulatory effect of flavonoids on gut microbiota and their metabolism by intestinal microbe under disease conditions, including inflammatory bowel disease, colitis-associated cancer (CAC), and CRC. We focus on dietary flavonoids and microbial interactions in intestinal mucosal barriers as well as intestinal immune cells. Results provide novel insights to better understand the crosstalk between dietary flavonoids and gut microbiota and support the standpoint that dietary flavonoids prevent intestinal inflammation and carcinogenesis.

A robotic sensory system with high spatiotemporal resolution for texture recognition.

Humans can gently slide a finger on the surface of an object and identify it by capturing both static pressure and high-frequency vibrations. Although modern robots integrated with flexible sensors can precisely detect pressure, shear force, and strain, they still perform insufficiently or require multi-sensors to respond to both static and high-frequency physical stimuli during the interaction. Here, we report a real-time artificial sensory system for high-accuracy texture recognition based on a single iontronic slip-sensor, and propose a criterion-spatiotemporal resolution, to corelate the sensing performance with recognition capability. The sensor can respond to both static and dynamic stimuli (0-400 Hz) with a high spatial resolution of 15 µm in spacing and 6 µm in height, together with a high-frequency resolution of 0.02 Hz at 400 Hz, enabling high-precision discrimination of fine surface features. The sensory system integrated on a prosthetic fingertip can identify 20 different commercial textiles with a 100.0% accuracy at a fixed sliding rate and a 98.9% accuracy at random sliding rates. The sensory system is expected to help achieve subtle tactile sensation for robotics and prosthetics, and further be applied to haptic-based virtual reality and beyond.

Determination of tobacco-specific nitrosamines in tobacco and mainstream cigarette smoke using one-step clean-up coupled with liquid chromatography- tandem mass spectrometry.

A method for determining tobacco-specific nitrosamines (TSNAs) in tobacco and cigarette smoke using liquid chromatography-tandem mass spectrometry was established. The established method amended the deficiencies that exist in current mainstream methods. In this method, TSNAs in tobacco and cigarette smoke were extracted by water. The aqueous extract was then extracted by dichloromethane, and the extract could be analyzed by liquid chromatography-tandem mass spectrometry after a solvent replacement. This method was used to analyze flue-cured tobacco samples, and the response of the target compounds was about 10 times higher than that of the ammonium acetate extraction method. When analyzing cigarette smoke samples, the response strength and chromatographic peak purity of the target compounds were also significantly improved. The proposed method exhibited good linearities for both tobacco and cigarette smoke samples (r2 > 0.99). The limits of detection (LODs) for tobacco and cigarette smoke samples were 0.2-1.0 ng/g and 0.1-0.3 ng/cigarette, respectively. Additionally, this method exhibited desirable accuracy and precision. The TSNAs recovery values from tobacco and cigarette smoke samples ranged from 95.7 % to 107.7 % with inter- and intra-day relative standard deviations (RSDs) of less than 7.4 %. This method is simple, effective, and has wide adaptability. It is a useful upgrade to the existing methods for analyzing TSNAs in tobacco and cigarette smoke.

Ethylene inhibits ABA-induced stomatal closure via regulating NtMYB184-mediated flavonol biosynthesis in tobacco.

Stomatal movement can be regulated by ABA signaling through synthesis of reactive oxygen species (ROS) in guard cells. By contrast, ethylene triggers the biosynthesis of antioxidant flavonols to suppress ROS accumulation and prevent ABA-induced stomatal closure; however, the underlying mechanism remains largely unknown. In this study, we isolated and characterized the tobacco (Nicotiana tabacum) R2R3-MYB transcription factor NtMYB184, which belongs to the flavonol-specific SG7 subgroup. RNAi suppression and CRISPR/Cas9 mutation (myb184) of NtMYB184 in tobacco caused down-regulation of flavonol biosynthetic genes and decreased the concentration of flavonols in the leaves. Yeast one-hybrid assays, transactivation assays, EMSAs, and ChIP-qPCR demonstrated that NtMYB184 specifically binds to the promoters of flavonol biosynthetic genes via MYBPLANT motifs. NtMYB184 regulated flavonol biosynthesis in guard cells to modulate ROS homeostasis and stomatal aperture. ABA-induced ROS production was accompanied by the suppression of NtMYB184 and flavonol biosynthesis, which may accelerate ABA-induced stomatal closure. Furthermore, ethylene stimulated NtMYB184 expression and flavonol biosynthesis to suppress ROS accumulation and curb ABA-induced stomatal closure. In myb184, however, neither the flavonol and ROS concentrations nor the stomatal aperture varied between the ABA and ABA+ethylene treatments, indicating that NtMYB184 was indispensable for the antagonism between ethylene and ABA via regulating flavonol and ROS concentrations in the guard cells.

Embedment of sensing elements for robust, highly sensitive, and cross-talk-free iontronic skins for robotics applications.

Iontronic pressure sensors are promising in robot haptics because they can achieve high sensing performance using nanoscale electric double layers (EDLs) for capacitive signal output. However, it is challenging to achieve both high sensitivity and high mechanical stability in these devices. Iontronic sensors need microstructures that offer subtly changeable EDL interfaces to boost sensitivity, while the microstructured interfaces are mechanically weak. Here, we embed isolated microstructured ionic gel (IMIG) in a hole array (28 × 28) of elastomeric matrix and cross-link the IMIGs laterally to achieve enhanced interfacial robustness without sacrificing sensitivity. The embedded configuration toughens and strengthens the skin by pinning cracks and by the elastic dissipation of the interhole structures. Furthermore, cross-talk between the sensing elements is suppressed by isolating the ionic materials and by designing a circuit with a compensation algorithm. We have demonstrated that the skin is potentially useful for robotic manipulation tasks and object recognition.

Structure Elucidation and Quantitation of 11 N'-n-Acylnornicotines in Cherry-Red Tobacco.

Eleven consecutive N'-n-acylnornicotines from cherry-red tobacco were structurally elucidated and quantitively analyzed using chromatography and mass spectrometry. All of these N'-n-acylnornicotines are first reported in cherry-red tobacco, whereas N'-propionylnornicotine, N'-n-valerylnornicotine, N'-n-nonanoylnornicotine and N'-n-undecanoylnornicotine are first reported in natural products. The concentration distribution of the identified N'-n-acylnornicotines was studied and it was found that N'-n-octanoylnornicotine and N'-n-hexanoylnornicotine showed the highest concentration, accounting for 94% of the detected N'-n-acylnornicotines. The cherry-red color density of the related tobacco leaves was found to be positively correlated with the concentration of the N'-n-acylnornicotines, whereas the ultraviolet-visible spectra of the N'-n-acylnornicotines showed no absorption larger than 300 nm, indicating the discovered compounds are not responsible for the cherry-red color appearance.

A portable smartphone-based detection of glyphosate based on inhibiting peroxidase-like activity of heptanoic acid/Prussian blue decorated Fe3O4 nanoparticles.

The rapid and onsite detection of glyphosate in tobacco products is still a great challenge. In this study, a novel smartphone-assisted sensing platform for the detection of glyphosate has been successfully proposed through the peroxidase-like activity of Fe3O4-based nanozyme. Heptanoic acid/Prussian blue (PB) decorated Fe3O4 nanoparticles (Fe3O4@C7/PB) could catalyze and oxidize 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS, colorless) into a steel blue colored product in the presence of hydrogen peroxide. Glyphosate could specifically inhibit the peroxidase-like activity of Fe3O4@C7/PB by occupying the active site, thereby the glyphosate detection could be accomplished within 10 min by monitoring the color change of ABTS. This study has developed a smartphone-based portable detection platform for online analysis of glyphosate with a detection limit of 0.1 µg mL-1. The absorbance response curve of glyphosate showed good linearity in the concentration range of 0.125-15 µg mL-1 at 415, 647, and 730 nm. Moreover, by employing a co-precipitation technology and inhibiting the peroxidase-like activity, the glyphosate analysis would be less affected by the tobacco sample matrix. The nanosensor possesses excellent selectivity and anti-interference ability, which has application value in actual samples for onsite screening.

Superlarge living hyaline cartilage graft contributed by the scale-changed porous 3D culture system for joint defect repair.

It is known that an excellent hyaline cartilage phenotype, an internal microstructure with safe crosslinking and available size flexibility are the key factors of cartilage grafts that allow for clinical application. Living hyaline cartilage grafts (LhCGs) constructed by phase-transfer hydrogel (PTCC) systems were reported to have a hyaline phenotype and bionic microstructure. By employing chondrocytes to secrete matrix in the hydrogel and then removing the material to obtain material-free tissuein vitro, LhCG technology exhibited superior performance in cartilage repair. However, PTCC systems could only produce small-sized LhCGs because of medium delivery limitations, which hinders the clinical application of LhCGs. In this study, we prepared three different noncrosslinked gelatin microspheres with diameters from 200 µm to 500 µm, which replaced the original pore-forming agent. The new PTCC system with the mixed and gradient porous structure was used for the preparation of superlarge LhCGs with a continuous structure and hyaline phenotype. Compared to the original technique, the porous gradient structure promoted nutrient delivery and cartilage matrix secretion. The small size of the microporous structure promoted the rapid formation of matrix junctions. The experimental group with a mixed gradient increased cartilage matrix secretion significantly by more than 50% compared to the that of the control. The LhCG final area reached 7 cm2without obvious matrix stratification in the mixed gradient group. The design of the scale-changed porous PTCC system will make LhCGs more promising for clinical application.

Transcriptome and Metabonomics Analysis Revealed the Molecular Mechanism of Differential Metabolite Production of Dendrobium nobile Under Different Epiphytic Patterns.

The cultivation medium of Dendrobium nobile has an effect on the contents of its main medicinal components, but the specific mechanism is still unclear. In this study, the callus, seedlings, rhizomes, and leaves of D. nobile were sequenced for the PacBio SMRT. The 2-year-old stems were selected for the Illumina sequencing and metabolome sequencing to analyze the genetic mechanism of metabolic differences under different epiphytic patterns. As a result, a total of 387 differential genes were obtained, corresponding to 66 differential metabolites. Different epiphytic patterns can induce a series of metabolic changes at the metabolome and transcriptome levels of D. nobile, including flavonoid metabolism, purine metabolism, terpenoid backbone biosynthesis, amino acid metabolism, and alpha-linolenic acid metabolic, and related regulatory genes include ALDH2B7, ADC, EPSPS-1, SHKA, DHAPS-1, GES, ACS1, SAHH, ACS2, CHLP, LOX2, LOX2.3, and CYP74B2. The results showed that the genetic mechanism of D. nobile under various epiphytic patterns was different. In theory, the content of metabolites under the epiphytic patterns of Danxia stone is higher, which is more suitable for field cultivation.

ß-Cell function and body mass index are predictors of exercise response in elderly patients with prediabetes.

AIMS/INTRODUCTION: To explore the predicting factors of exercise response (whether the participants converted to diabetes) in elderly patients with prediabetes. MATERIALS AND METHODS: This is a retrospective subgroup analysis of the registered clinical trial with previous publication of the same cohort. A total of 248 participants with prediabetes were randomized to the aerobic training (n = 83) group, resistance training (n = 82) group and control group (n = 83). The patients who finished the 2-year exercise intervention were included in this analysis to explore the factors impacting exercise response. RESULTS: A total of 113 patients with prediabetes completed 2 years of exercise, with 56 participants in the aerobic exercise group and 57 in the resistance exercise group. Patients who reversed to normal glucose tolerance, remained in prediabetes and developed diabetes were 18 (15.90%), 70 (62.00%) and 25 (22.10%), respectively. Logistic regression showed that baseline, homeostatic model 2 assessment of ß-cell function (ß = -0.143, P = 0.039), hemoglobin A1c (ß = 3.301, P = 0.007) and body mass index (ß = 0.402, P = 0.012) were related to exercise response, whereas the waist-to-hip ratio (ß = -3.277, P = 0.693) and types of exercise (ß = 1.192, P = 0.093) were not significantly related to exercise response. CONCLUSIONS: Baseline homeostatic model 2 assessment of ß-cell function, hemoglobin A1c and body mass index were the predictors for the response to exercise in elderly patients with prediabetes.

High-Porosity Foam-Based Iontronic Pressure Sensor with Superhigh Sensitivity of 9280 kPa-1.

Flexible pressure sensors with high sensitivity are desired in the fields of electronic skins, human-machine interfaces, and health monitoring. Employing ionic soft materials with microstructured architectures in the functional layer is an effective way that can enhance the amplitude of capacitance signal due to generated electron double layer and thus improve the sensitivity of capacitive-type pressure sensors. However, the requirement of specific apparatus and the complex fabrication process to build such microstructures lead to high cost and low productivity. Here, we report a simple strategy that uses open-cell polyurethane foams with high porosity as a continuous three-dimensional network skeleton to load with ionic liquid in a one-step soak process, serving as the ionic layer in iontronic pressure sensors. The high porosity (95.4%) of PU-IL composite foam shows a pretty low Young's modulus of 3.4 kPa and good compressibility. A superhigh maximum sensitivity of 9,280 kPa-1 in the pressure regime and a high pressure resolution of 0.125% are observed in this foam-based pressure sensor. The device also exhibits remarkable mechanical stability over 5,000 compression-release or bending-release cycles. Such high porosity of composite structure provides a simple, cost-effective and scalable way to fabricate super sensitive pressure sensor, which has prominent capability in applications of water wave detection, underwater vibration sensing, and mechanical fault monitoring.

Automated delineation of orbital abscess depicted on CT scan using deep learning.

OBJECTIVES: To develop and validate a deep learning algorithm to automatically detect and segment an orbital abscess depicted on computed tomography (CT). METHODS: We retrospectively collected orbital CT scans acquired on 67 pediatric subjects with a confirmed orbital abscess in the setting of infectious orbital cellulitis. A context-aware convolutional neural network (CA-CNN) was developed and trained to automatically segment orbital abscess. To reduce the requirement for a large dataset, transfer learning was used by leveraging a pre-trained model for CT-based lung segmentation. An ophthalmologist manually delineated orbital abscesses depicted on the CT images. The classical U-Net and the CA-CNN models with and without transfer learning were trained and tested on the collected dataset using the 10-fold cross-validation method. Dice coefficient, Jaccard index, and Hausdorff distance were used as performance metrics to assess the agreement between the computerized and manual segmentations. RESULTS: The context-aware U-Net with transfer learning achieved an average Dice coefficient and Jaccard index of 0.78 ± 0.12 and 0.65 ± 0.13, which were consistently higher than the classical U-Net or the context-aware U-Net without transfer learning (P < 0.01). The average differences of the abscess between the computerized results and the experts in terms of volume and Hausdorff distance were 0.10 ± 0.11 mL and 1.94 ± 1.21 mm, respectively. The context-aware U-Net detected all orbital abscess without false positives. CONCLUSIONS: The deep learning solution demonstrated promising performance in detecting and segmenting orbital abscesses on CT images in strong agreement with a human observer.