Orientation of the NV centers are determined using the cylindrical vector beam array.

The determination of nitrogen-vacancy centers plays an important role in quantum information sensing. Efficiently and rapidly determining the orientation of multiple nitrogen-vacancy center s in a low-concentration diamond is challenging due to its size. Here, we solve this scientific problem by using an azimuthally polarized beam array as the incident beam. In this paper, the optical pen is used to modulate the position of beam array to excite distinctive fluorescence characterizing multiple and different orientations of nitrogen-vacancy centers. The important result is that in a low concentration diamond layer, the orientation of multiple NV centers can be judged except when they are too close within the diffraction limit. Hence, this efficient and rapid method has a good application prospect in quantum information sensing.

Multi-type feature fusion based on graph neural network for drug-drug interaction prediction.

BACKGROUND: Drug-Drug interactions (DDIs) are a challenging problem in drug research. Drug combination therapy is an effective solution to treat diseases, but it can also cause serious side effects. Therefore, DDIs prediction is critical in pharmacology. Recently, researchers have been using deep learning techniques to predict DDIs. However, these methods only consider single information of the drug and have shortcomings in robustness and scalability. RESULTS: In this paper, we propose a multi-type feature fusion based on graph neural network model (MFFGNN) for DDI prediction, which can effectively fuse the topological information in molecular graphs, the interaction information between drugs and the local chemical context in SMILES sequences. In MFFGNN, to fully learn the topological information of drugs, we propose a novel feature extraction module to capture the global features for the molecular graph and the local features for each atom of the molecular graph. In addition, in the multi-type feature fusion module, we use the gating mechanism in each graph convolution layer to solve the over-smoothing problem during information delivery. We perform extensive experiments on multiple real datasets. The results show that MFFGNN outperforms some state-of-the-art models for DDI prediction. Moreover, the cross-dataset experiment results further show that MFFGNN has good generalization performance. CONCLUSIONS: Our proposed model can efficiently integrate the information from SMILES sequences, molecular graphs and drug-drug interaction networks. We find that a multi-type feature fusion model can accurately predict DDIs. It may contribute to discovering novel DDIs.

Simultaneous and ultra-sensitive SERS detection of SLPI and IL-18 for the assessment of donor kidney quality using black phosphorus/gold nanohybrids.

Due to the global challenge of donor kidney shortage, expanding the pool of deceased donors has been proposed to include expanded criteria donors. However, the lack of methods to precisely measure donor kidney injury and predict the outcome still leads to high discard rates and recipient complications. As such, evaluation of deceased donor kidney quality is critical prior to transplantation. Biomarkers from donor urine or serum provide potential advantages for the precise measure of kidney quality. Herein, simultaneous detection of secretory leukocyte peptidase inhibitor (SLPI) and interleukin 18 (IL-18), two important kidney injury biomarkers, has been achieved, for the first time, with an ultra-high sensitivity using surface enhanced Raman scattering (SERS). Specifically, black phosphorus/gold (BP/Au) nanohybrids synthesized by depositing Au nanoparticles (NPs) onto the BP nanosheets serve as SERS-active substrates, which offer a high-density of inherent and accessible hot-spots. Meanwhile, the nanohybrids possess biocompatible surfaces for the enrichment of target biomarkers through the affinity with BP nanosheets. Quantitative detection of SLPI and IL-18 were then achieved by characterizing SERS signals of these two biomarkers. The results indicate high sensitivity and excellent reproducibility of this method. The limits of detection reach down to 1.53×10-8 mg/mL for SLPI and 0.23×10-8 mg/mL for IL-18. The limits of quantification are 5.10×10-8 mg/mL and 7.67×10-9 mg/mL for SLPI and IL-18. In addition, simultaneous detection of these biomarkers in serum was investigated, which proves the feasibility in biologic environment. More importantly, this method is powerful for detecting multiple analytes inheriting from excellent multiplexing ability of SERS. Giving that the combined assessment of SLPI and IL-18 expression level serves as an indicator of donor kidney quality and can be rapidly and reproducibly conducted, this SERS-based method holds great prospective in clinical practice.

Dual-targeting SERS-encoded graphene oxide nanocarrier for intracellular co-delivery of doxorubicin and 9-aminoacridine with enhanced combination therapy.

A graphene oxide (GO)-based nanocarrier that imparts tumor-selective delivery of dual-drug with enhanced therapeutic index, is introduced. GO is conjugated with Au@Ag and Fe3O4 nanoparticles, which facilitates it with SERS tracking and magnetic targeting abilities, followed by the covalent binding of the anti-HER2 antibody, thus allowing it to both actively and passively target SKBR3 cells, human breast cancer cells expressed with HER2. Intracellular drug delivery behaviors are probed using SERS spectroscopy in a spatiotemporal manner, which demonstrates that nanocarriers are internalized into the lysosomes and release the drug in response to the acidic microenvironment. The nanocarriers loaded with dual-drug possess increased cancer cytotoxicity in comparison to those loaded with a single drug. Attractively, the enhanced cytotoxicity against cancer cells is achieved with relatively low concentrations of the drug, which is demonstrated to be involved in the drug adsorption status. These results may give us the new prospects to design GO-based delivery systems with rational drug dosages, thus achieving optimal therapeutic response of the multi-drug with increased tumor selectivity and reduced side effects.

In situ synthesis of graphene oxide/gold nanocomposites as ultrasensitive surface-enhanced Raman scattering substrates for clenbuterol detection.

A highly sensitive approach to detect trace amount of clenbuterol (CB) based on graphene oxide/gold nanoparticles (GO/Au NPs) by surface-enhanced Raman spectroscopy (SERS) was presented. To be specific, the GO/Au nanocomposites were formed by depositing Au NPs onto the surface of GO through an in situ reduction process, where a high density of inherent hot spots was created between Au NPs. By optimizing the depositing density of Au NPs, the strongest electromagnetic coupling effect originating from highly dense hot spots was obtained. The optimized GO/Au was demonstrated to enhance the Raman signals of CB by 4.8 times more than that of CB enhanced by Au NPs. Moreover, GO/Au nanocomposites exhibit good biocompatibility and accessible surface for high adsorption of target molecules through the pi-pi stacking with graphene oxide. Hence, the proposed GO/Au nanocomposites were utilized to capture aromatic molecules like CB and served as excellent sensitive SERS-active substrates for sensing of it, which exhibited an excellent linear performance in the range of 5 × 10-8 to 1 × 10-6 mol/L with a limit of detection (LOD) of 3.34 × 10-8 mol/L (S/N = 3). Due to high-density hot spots with easy operation, this proposed GO/Au nanocomposite-based SERS technique holds great potential in the application of food safety analysis and biomedical science.

Deep learning-based automated detection of glaucomatous optic neuropathy on color fundus photographs.

PURPOSE: To develop a deep learning approach based on deep residual neural network (ResNet101) for the automated detection of glaucomatous optic neuropathy (GON) using color fundus images, understand the process by which the model makes predictions, and explore the effect of the integration of fundus images and the medical history data from patients. METHODS: A total of 34,279 fundus images and the corresponding medical history data were retrospectively collected from cohorts of 2371 adult patients, and these images were labeled by 8 glaucoma experts, in which 26,585 fundus images (12,618 images with GON-confirmed eyes, 1114 images with GON-suspected eyes, and 12,853 NORMAL eye images) were included. We adopted 10-fold cross-validation strategy to train and optimize our model. This model was tested in an independent testing dataset consisting of 3481 images (1524 images from NORMAL eyes, 1442 images from GON-confirmed eyes, and 515 images from GON-suspected eyes) from 249 patients. Moreover, the performance of the best model was compared with results obtained by two experts. Accuracy, sensitivity, specificity, kappa value, and area under receiver operating characteristic (AUC) were calculated. Further, we performed qualitative evaluation of model predictions and occlusion testing. Finally, we assessed the effect of integrating medical history data in the final classification. RESULTS: In a multiclass comparison between GON-confirmed eyes, GON-suspected eyes and NORMAL eyes, our model achieved 0.941 (95% confidence interval [CI], 0.936-0.946) accuracy, 0.957 (95% CI, 0.953-0.961) sensitivity, and 0.929 (95% CI, 0.923-0.935) specificity. The AUC distinguishing referrals (GON-confirmed and GON-suspected eyes) from observation was 0.992 (95% CI, 0.991-0.993). Our best model had a kappa value of 0.927, while the two experts' kappa values were 0.928 and 0.925 independently. The best 2 binary classifiers distinguishing GON-confirmed/GON-suspected eyes from NORMAL eyes obtained 0.955, 0.965 accuracy, 0.977, 0.998 sensitivity, and 0.929, 0.954 specificity, while the AUC was 0.992, 0.999 respectively. Additionally, the occlusion testing showed that our model identified the neuroretinal rim region, retinal nerve fiber layer (RNFL) defect areas (superior or inferior) as the most important parts for the discrimination of GON, which evaluated fundus images in a way similar to clinicians. Finally, the results of integration of fundus images with medical history data showed a slight improvement in sensitivity and specificity with similar AUCs. CONCLUSIONS: This approach could discriminate GON with high accuracy, sensitivity, specificity, and AUC using color fundus photographs. It may provide a second opinion on the diagnosis of glaucoma to the specialist quickly, efficiently and at low cost, and assist doctors and the public in large-scale screening for glaucoma.

Three-dimensional detection and quantification of defects in SiC by optical coherence tomography.

Silicon carbide (SiC) is widely used in high power electronic devices. However, defects on the SiC significantly reduce the yield and decrease the performance of SiC. Accurate detection of the defects is essential in the process control. We demonstrated a noninvasive three-dimensional (3D) defect detection method for SiC using optical coherence tomography (OCT). Defects including the triangular defects, hexagonal voids, grain boundaries, and carrot defects were inspected and analyzed on SiC wafers. The 3D images of defects acquired with OCT provided detailed information on the 3D structures and dimensions of defects, and the locations and orientations of the defects inside the wafers. This technique was not only useful for rapid defect screening in the process control, it was also extremely helpful in understanding the formation mechanism of these defects in SiC.

Colorimetric and SERS dual-mode detection of lead Ions based on Au-Ag core-shell nanospheres: featuring quick screening with ultra-high sensitivity.

A novel strategy for colorimetric and surface-enhanced Raman scattering (SERS) dual-mode sensing of the lead ions (Pb2+) was established based on gluconate ion (Gluc) modified and 2-Naphthalenethiol (2-NT) tagged Au-Ag core-shell nanoparticles (NPs). Due to the complex formation between adsorbed Gluc and Pb2+, the addition of Pb2+ can induce the aggregation of Gluc/2-NT@Au@Ag NPs. Correspondingly, the aggregated Gluc/2-NT@Au@Ag NPs caused a significant difference in the color and SERS intensity. As a result, such Gluc/2-NT@Au@Ag NPs can achieve the sensing of Pb2+ using both colorimetric and SERS signals as the indicator, which features with wide response range from 10-11 to 10-5 M, rapid screening and high sensitivity (with a limit of detection (LOD) of 0.185 pM). Furthermore, such dual-mode sensor was demonstrated not to be responsive to other cations, and facilitate the sensing of real samples in practical environment. With rapid screening ability and outstanding sensitivity, we anticipate that this method would holding great potential for the applications in environmental monitoring.

Fully automated detection of retinal disorders by image-based deep learning.

PURPOSE: With the aging population and the global diabetes epidemic, the prevalence of age-related macular degeneration (AMD) and diabetic macular edema (DME) diseases which are the leading causes of blindness is further increasing. Intravitreal injections with anti-vascular endothelial growth factor (anti-VEGF) medications are the standard of care for their indications. Optical coherence tomography (OCT), as a noninvasive imaging modality, plays a major part in guiding the administration of anti-VEGF therapy by providing detailed cross-sectional scans of the retina pathology. Fully automating OCT image detection can significantly decrease the tedious clinician labor and obtain a faithful pre-diagnosis from the analysis of the structural elements of the retina. Thereby, we explore the use of deep transfer learning method based on the visual geometry group 16 (VGG-16) network for classifying AMD and DME in OCT images accurately and automatically. METHOD: A total of 207,130 retinal OCT images between 2013 and 2017 were selected from retrospective cohorts of 5319 adult patients from the Shiley Eye Institute of the University of California San Diego, the California Retinal Research Foundation, Medical Center Ophthalmology Associates, the Shanghai First People's Hospital, and the Beijing Tongren Eye Center, with 109,312 images (37,456 with choroidal neovascularization, 11,599 with diabetic macular edema, 8867 with drusen, and 51,390 normal) for the experiment. After images preprocessing, 1000 images (250 images from each category) from 633 patients were selected as validation dataset while the rest images from another 4686 patients were used as training dataset. We used deep transfer learning method to fine-tune the VGG-16 network pre-trained on the ImageNet dataset, and evaluated its performance on the validation dataset. Then, prediction accuracy, sensitivity, specificity, and receiver-operating characteristic (ROC) were calculated. RESULTS: Experimental results proved that the proposed approach had manifested superior performance in retinal OCT images detection, which achieved a prediction accuracy of 98.6%, with a sensitivity of 97.8%, a specificity of 99.4%, and introduced an area under the ROC curve of 100%. CONCLUSION: Deep transfer learning method based on the VGG-16 network shows significant effectiveness on classification of retinal OCT images with a relatively small dataset, which can provide assistant support for medical decision-making. Moreover, the performance of the proposed approach is comparable to that of human experts with significant clinical experience. Thereby, it will find promising applications in an automatic diagnosis and classification of common retinal diseases.

Comparison of Choroidal Thickness in Preeclamptic, Healthy Pregnant, and Nonpregnant Women: A Systematic Review and Meta-Analysis.

PURPOSE: To compare the choroidal thickness (CT) measured by enhanced depth imaging optical coherence tomography (EDI-OCT) in preeclamptic, healthy pregnant, and healthy nonpregnant women. METHODS: Studies that focused on the CT evaluation of pregnant women were retrieved by searching PubMed, Embase, Ovid, Cochrane, and Web of Science. We used Stata 14.0 SE for the meta-analysis and presented the results as the weighted mean difference (WMD) with a corresponding 95% CI. RESULTS: A total of 14 studies with 1,227 participants were included in our meta-analysis. The CT of the healthy pregnancies (µm, WMD = 34.19, 95% CI: 20.63-47.76) was significantly higher than that of the nonpregnancies (Test of WMD = 0: z = 4.94, p = 0.000), but the CT of the preeclampsia (µm, WMD = 54.30, 95% CI: -13.40 to 122.01) was not significantly different from the nonpregnancies (Test of WMD = 0: z = 1.57, p = 0.116). In the preeclampsia versus healthy pregnancy group, 3 studies found that the choroid was thinner with preeclampsia, only one study found the CT increased. CONCLUSIONS: This meta-analysis suggested that the CT of the healthy pregnant women was significantly higher than that of the nonpregnant women. The presence of preeclampsia might complicate this situation. Most studies found that the CT decreased in the preeclamptic patients because of the increases in the systemic vasospasm and the blood pressure, which led to no significant difference compared with the nonpregnant women.

Photonic Crystal Fiber Plasmonic Sensor Based on Dual Optofluidic Channel.

A hexagonal photonic crystal fiber (PCF) sensor with a dual optofluidic channel based on surface plasmon resonance (SPR) effect is proposed. The sensor characteristic is numerically explored by software integrated with the finite element method (FEM). The numerical results show that, when the analyte refractive index (RI) varies from 1.32 to 1.38, high linearity between resonance wavelength and analyte RI is obtained and the value of adjusted R2 is up to 0.9993. Simultaneously, the proposed sensor has maximum wavelength sensitivity (WS) of 5500 nm/RIU and maximum amplitude sensitivity (AS) of 150 RIU-1, with an RI resolution of 1.82 × 10-5 RIU. Besides, owing to a simple structure and good tolerance of the proposed sensor, it can be easily fabricated by means of existing technology. The proposed sensor suggests promising applications in oil detection, temperature measurement, water quality monitoring, bio-sensing, and food safety.

Fast and accurate auto-focusing algorithm based on the combination of depth from focus and improved depth from defocus.

An auto-focus method for digital imaging systems is proposed that combines depth from focus (DFF) and improved depth from defocus (DFD). The traditional DFD method is improved to become more rapid, which achieves a fast initial focus. The defocus distance is first calculated by the improved DFD method. The result is then used as a search step in the searching stage of the DFF method. A dynamic focusing scheme is designed for the control software, which is able to eliminate environmental disturbances and other noises so that a fast and accurate focus can be achieved. An experiment is designed to verify the proposed focusing method and the results show that the method's efficiency is at least 3-5 times higher than that of the traditional DFF method.

CoVi-Net: A hybrid convolutional and vision transformer neural network for retinal vessel segmentation.

Retinal vessel segmentation plays a crucial role in the diagnosis and treatment of ocular pathologies. Current methods have limitations in feature fusion and face challenges in simultaneously capturing global and local features from fundus images. To address these issues, this study introduces a hybrid network named CoVi-Net, which combines convolutional neural networks and vision transformer. In our proposed model, we have integrated a novel module for local and global feature aggregation (LGFA). This module facilitates remote information interaction while retaining the capability to effectively gather local information. In addition, we introduce a bidirectional weighted feature fusion module (BWF). Recognizing the variations in semantic information across layers, we allocate adjustable weights to different feature layers for adaptive feature fusion. BWF employs a bidirectional fusion strategy to mitigate the decay of effective information. We also incorporate horizontal and vertical connections to enhance feature fusion and utilization across various scales, thereby improving the segmentation of multiscale vessel images. Furthermore, we introduce an adaptive lateral feature fusion (ALFF) module that refines the final vessel segmentation map by enriching it with more semantic information from the network. In the evaluation of our model, we employed three well-established retinal image databases (DRIVE, CHASEDB1, and STARE). Our experimental results demonstrate that CoVi-Net outperforms other state-of-the-art techniques, achieving a global accuracy of 0.9698, 0.9756, and 0.9761 and an area under the curve of 0.9880, 0.9903, and 0.9915 on DRIVE, CHASEDB1, and STARE, respectively. We conducted ablation studies to assess the individual effectiveness of the three modules. In addition, we examined the adaptability of our CoVi-Net model for segmenting lesion images. Our experiments indicate that our proposed model holds promise in aiding the diagnosis of retinal vascular disorders.

Novel non-invasive method for urine mapping: Deep-learning-enabled SERS spectroscopy for the rapid differential detection of kidney allograft injury.

The kidney allograft has been under continuous attack from diverse injuries since the very beginning of organ procurement, leading to a gradual decline in function, chronic fibrosis, and allograft loss. It is vital to routinely and precisely monitor the risk of injuries after renal transplantation, which is difficult to achieve because the traditional laboratory tests lack sensitivity and specificity, and graft biopsies are invasive with the risk of many complications and time-consuming. Herein, a novel method for the diagnosis of graft injury is demonstrated, using deep learning-assisted surface-enhanced Raman spectroscopy (SERS) of the urine analysis. Specifically, we developed a hybrid SERS substrate composed of gold and silver with high sensitivity to the urine composition under test, eliminating the need for labels, which makes measurements easy to perform and meanwhile results in extremely abundant and complex Raman vibrational bands. Deep learning algorithms were then developed to improve the interpretation of the SERS spectral fingerprints. The deep learning model was trained with SERS signals of urine samples of recipients with different injury types including delayed graft function (DGF), calcineurin-inhibitor toxicity (CNIT), T cell-mediated rejection (TCMR), antibody-mediated rejection (AMR), and BK virus nephropathy (BKVN), which explored the features of these types and achieved the injury differentiation with an overall accuracy of 93.03%. The results highlight the potential of combining label-free SERS spectroscopy with deep learning as a method for liquid biopsy of kidney allograft injuries, which can provide great potential to diagnose and evaluate allograft injuries, and thus extend the life of kidney allografts.

Endlessly single-polarization single-mode holey fibers with low confinement loss.

We present a novel structure for holey fibers (HFs) with endlessly single-polarization single-mode characteristics, which is realized by introducing four elliptical airholes arranged in a hexagonal matrix in the core region. The validation of the design is done by use of a full-vectorial finite element method. We exhibit one typical design that can deliver a single-polarization single-mode region of more than 2400 nm with a confinement loss level lower than 0.01 dB/km. We have also shown that the persevered polarization state possesses a wide wavelength band of flat dispersion behavior. As a consequence, such HFs are useful in high-speed communication systems or optical-fiber sensors since they are free of polarization mode dispersion and simultaneously immune to cross-talk effect.

CCS-UNet: a cross-channel spatial attention model for accurate retinal vessel segmentation.

Precise segmentation of retinal vessels plays an important role in computer-assisted diagnosis. Deep learning models have been applied to retinal vessel segmentation, but the efficacy is limited by the significant scale variation of vascular structures and the intricate background of retinal images. This paper supposes a cross-channel spatial attention U-Net (CCS-UNet) for accurate retinal vessel segmentation. In comparison to other models based on U-Net, our model employes a ResNeSt block for the encoder-decoder architecture. The block has a multi-branch structure that enables the model to extract more diverse vascular features. It facilitates weight distribution across channels through the incorporation of soft attention, which effectively aggregates contextual information in vascular images. Furthermore, we suppose an attention mechanism within the skip connection. This mechanism serves to enhance feature integration across various layers, thereby mitigating the degradation of effective information. It helps acquire cross-channel information and enhance the localization of regions of interest, ultimately leading to improved recognition of vascular structures. In addition, the feature fusion module (FFM) module is used to provide semantic information for a more refined vascular segmentation map. We evaluated CCS-UNet based on five benchmark retinal image datasets, DRIVE, CHASEDB1, STARE, IOSTAR and HRF. Our proposed method exhibits superior segmentation efficacy compared to other state-of-the-art techniques with a global accuracy of 0.9617/0.9806/0.9766/0.9786/0.9834 and AUC of 0.9863/0.9894/0.9938/0.9902/0.9855 on DRIVE, CHASEDB1, STARE, IOSTAR and HRF respectively. Ablation studies are also performed to evaluate the the relative contributions of different architectural components. Our proposed model is potential for diagnostic aid of retinal diseases.

Design and Simulation of a Ratiometric SPR Sensor Based on a 2D van der Waals Heterojunction for Refractive Index Measurement.

Surface plasmon resonance (SPR) sensors have been widely applied in many fields because of their advantages of working in real time and high sensitivity. However, because the spectrum of an SPR sensor is easily affected by the smoothness of the metal surface, this type of sensor has obvious disadvantages in the application of quantitative detection. We designed an SPR refractive index sensor for molecular detection that has the advantage of quantifiability. A ratio spectral quantitative analysis method was established based on the two coherent dips of the SPR spectrum formed by the strong coupling effect between the surface plasmon polaritons and the excitons of the J-aggregate molecule 5,6-dichloro-2-[3-[5,6-dichloro-1-ethyl-3-(4-sulfobutyl)-2-benzimidazoline subunit] propenyl]-3-ethyl-1-(4-sulfobutyl) benzimidazole hydroxide inner salt (TDBC). The introduced MoS2/graphene van der Waals heterojunction produced an effective charge transfer to the Ag film, resulting in significant electric field enhancement at the sensing interface and further improving the detection sensitivity of the sensor. The simulation results showed that for 43 nm Ag film, for example, the ratiometric SPR sensor with the Ag film structure can obtain 16.12 RIU-1 sensing sensitivity, applied to the detection of gas molecules, while the SPR sensor with single-layer graphene and three layers of MoS2 heterostructures can obtain 50.68 RIU-1 sensing sensitivity. The addition of van der Waals heterostructures can significantly improve sensing performance by 215%.

Self-assembly of Au nanocubes for ultrasensitive detection of Alzheimer's disease biomarkers by SERS.

Since presently Alzheimer's disease (AD) is incurable, early diagnosis of AD is crucial. Aß 1-42 and tau-441 proteins are promising core biomarkers for early diagnosis and early therapeutic intervention in AD. Here we constructed a surface-enhanced Raman spectroscopy (SERS) biosensor for highly sensitive quantitative detection of Aß 1-42 and tau proteins by preparing gold nanocube (AuNC) superlattices through evaporation self-assembly. The results showed that the method has a wide response range (0.1-10 000 ng mL-1 and 0.01-1000 ng mL-1, respectively) and high sensitivity. The detection limits of Aß1-42 and tau protein were 0.0416 ng mL-1 and 0.0087 ng mL-1, respectively. In addition, the method was able to rapidly and simultaneously detect the two biomarkers in serum, which showed the feasibility of the method in complex biological environments. The detection of Aß 1-42 and tau protein has great potential for the accurate prediction and early diagnosis of Alzheimer's disease.

Joint optic disk and cup segmentation for glaucoma screening using a region-based deep learning network.

OBJECTIVES: To develop and validate an end-to-end region-based deep convolutional neural network (R-DCNN) to jointly segment the optic disc (OD) and optic cup (OC) in retinal fundus images for precise cup-to-disc ratio (CDR) measurement and glaucoma screening. METHODS: In total, 2440 retinal fundus images were retrospectively obtained from 2033 participants. An R-DCNN was presented for joint OD and OC segmentation, where the OD and OC segmentation problems were formulated into object detection problems. We compared R-DCNN's segmentation performance on our in-house dataset with that of four ophthalmologists while performing quantitative, qualitative and generalization analyses on the publicly available both DRISHIT-GS and RIM-ONE v3 datasets. The Dice similarity coefficient (DC), Jaccard coefficient (JC), overlapping error (E), sensitivity (SE), specificity (SP) and area under the curve (AUC) were measured. RESULTS: On our in-house dataset, the proposed model achieved a 98.51% DC and a 97.07% JC for OD segmentation, and a 97.63% DC and a 95.39% JC for OC segmentation, achieving a performance level comparable to that of the ophthalmologists. On the DRISHTI-GS dataset, our approach achieved 97.23% and 94.17% results in DC and JC results for OD segmentation, respectively, while it achieved a 94.56% DC and an 89.92% JC for OC segmentation. Additionally, on the RIM-ONE v3 dataset, our model generated DC and JC values of 96.89% and 91.32% on the OD segmentation task, respectively, whereas the DC and JC values acquired for OC segmentation were 88.94% and 78.21%, respectively. CONCLUSION: The proposed approach achieved very encouraging performance on the OD and OC segmentation tasks, as well as in glaucoma screening. It has the potential to serve as a useful tool for computer-assisted glaucoma screening.

CNN-assisted SERS enables ultra-sensitive and simultaneous detection of Scr and BUN for rapid kidney function assessment.

Kidney disease is highly prevalent and may result in severe clinical outcomes. Serum creatinine (Scr) and blood urea nitrogen (BUN) are the most widely used biomarkers for kidney function assessment, yet when measured alone, the result can be affected by a variety of parameters such as age, gender, protein consumption, etc. Measuring Scr and BUN simultaneously can eliminate most of the external influences and greatly improve the assessment of kidney function. In this study, a real-time kidney function assessment system based on dual biomarker detection was proposed. Scr and BUN were determined using surface-enhanced Raman scattering (SERS) within the concentration range of 10-1 to 10-6 M and 0.28 to 100 mg dl-1, respectively. A one-dimensional convolutional neural network (1D-CNN) model was employed to quantitatively analyze the concentration of biomarkers from the SERS spectral measurements. Moreover, we simulated a variety of kidney health conditions with 16 groups of mixed Scr and BUN in serum. The proposed CNN-assisted SERS method was used to quantify both biomarkers and provide diagnostic results. The Au core-Ag shell nanoprobes provided ultra-sensitive SERS detection and the CNN model achieved excellent regression results with an R2 of 0.9871 in the testing dataset. The system demonstrated a rapid and robust evaluation for the assessment of kidney function, providing a promising idea for medical diagnosis with the help of spectroscopy and deep learning methods.