Conformal and conductive biofilm-bridged artificial Z-scheme system for visible light-driven overall water splitting.

Semi-artificial Z-scheme systems offer promising potential toward efficient solar-to-chemical conversion, yet sustainable and stable designs are currently lacking. Here, we developed a sustainable hybrid Z-scheme system capable for visible light-driven overall water splitting by integrating the durability of inorganic photocatalysts with the interfacial adhesion and regenerative property of bacterial biofilms. The Z-scheme configuration is fabricated by drop casting a mixture of photocatalysts onto a glass plate, followed by the growth of biofilms for conformal conductive paste through oxidative polymerization of pyrrole molecules. Notably, the system exhibited scalability indicated by consistent catalytic efficiency across various sheet areas, resistance observed by remarkable maintaining of photocatalytic efficiency across a range of background pressures, and high stability as evidenced by minimal decay of photocatalytic efficiency after 100-hour reaction. Our work thus provides a promising avenue toward sustainable and high-efficiency artificial photosynthesis, contributing to the broader goal of sustainable energy solutions.

Application of HFO and scaling analysis of neuronal oscillations in the presurgical evaluation of focal epilepsy.

PURPOSE: To explore the utility of high frequency oscillations (HFO) and long-range temporal correlations (LRTCs) in preoperative assessment of epilepsy. METHODS: MEG ripples were detected in 59 drug-resistant epilepsy patients, comprising 5 with parietal lobe epilepsy (PLE), 21 with frontal lobe epilepsy (FLE), 14 with lateral temporal lobe epilepsy (LTLE), and 19 with mesial temporal lobe epilepsy (MTLE) to identify the epileptogenic zone (EZ). The results were compared with clinical MEG reports and resection area. Subsequently, LRTCs were quantified at the source-level by detrended fluctuation analysis (DFA) and life/waiting -time at 5 bands for 90 cerebral cortex regions. The brain regions with larger DFA exponents and standardized life-waiting biomarkers were compared with the resection results. RESULTS: Compared to MEG sensor-level data, ripple sources were more frequently localized within the resection area. Moreover, source-level analysis revealed a higher proportion of DFA exponents and life-waiting biomarkers with relatively higher rankings, primarily distributed within the resection area (p<0.01). Moreover, these two LRCT indices across five distinct frequency bands correlated with EZ. CONCLUSION: HFO and source-level LRTCs are correlated with EZ. Integrating HFO and LRTCs may be an effective approach for presurgical evaluation of epilepsy.

Glaucoma detection model by exploiting multi-region and multi-scan-pattern OCT images with dynamical region score.

Currently, deep learning-based methods have achieved success in glaucoma detection. However, most models focus on OCT images captured by a single scan pattern within a given region, holding the high risk of the omission of valuable features in the remaining regions or scan patterns. Therefore, we proposed a multi-region and multi-scan-pattern fusion model to address this issue. Our proposed model exploits comprehensive OCT images from three fundus anatomical regions (macular, middle, and optic nerve head regions) being captured by four scan patterns (radial, volume, single-line, and circular scan patterns). Moreover, to enhance the efficacy of integrating features across various scan patterns within a region and multiple regional features, we employed an attention multi-scan fusion module and an attention multi-region fusion module that auto-assign contribution to distinct scan-pattern features and region features adapting to characters of different samples, respectively. To alleviate the absence of available datasets, we have collected a specific dataset (MRMSG-OCT) comprising OCT images captured by four scan patterns from three regions. The experimental results and visualized feature maps both demonstrate that our proposed model achieves superior performance against the single scan-pattern models and single region-based models. Moreover, compared with the average fusion strategy, our proposed fusion modules yield superior performance, particularly reversing the performance degradation observed in some models relying on fixed weights, validating the efficacy of the proposed dynamic region scores adapted to different samples. Moreover, the derived region contribution scores enhance the interpretability of the model and offer an overview of the model's decision-making process, assisting ophthalmologists in prioritizing regions with heightened scores and increasing efficiency in clinical practice.

Accelerating the design of pili-enabled living materials using an integrative technological workflow.

Bacteria can be programmed to create engineered living materials (ELMs) with self-healing and evolvable functionalities. However, further development of ELMs is greatly hampered by the lack of engineerable nonpathogenic chassis and corresponding programmable endogenous biopolymers. Here, we describe a technological workflow for facilitating ELMs design by rationally integrating bioinformatics, structural biology and synthetic biology technologies. We first develop bioinformatics software, termed Bacteria Biopolymer Sniffer (BBSniffer), that allows fast mining of biopolymers and biopolymer-producing bacteria of interest. As a proof-of-principle study, using existing pathogenic pilus as input, we identify the covalently linked pili (CLP) biosynthetic gene cluster in the industrial workhorse Corynebacterium glutamicum. Genetic manipulation and structural characterization reveal the molecular mechanism of the CLP assembly, ultimately enabling a type of programmable pili for ELM design. Finally, engineering of the CLP-enabled living materials transforms cellulosic biomass into lycopene by coupling the extracellular and intracellular bioconversion ability.

Semi-supervised point consistency network for retinal artery/vein classification.

Recent deep learning methods with convolutional neural networks (CNNs) have boosted advance prosperity of medical image analysis and expedited the automatic retinal artery/vein (A/V) classification. However, it is challenging for these CNN-based approaches in two aspects: (1) specific tubular structures and subtle variations in appearance, contrast, and geometry, which tend to be ignored in CNNs with network layer increasing; (2) limited well-labeled data for supervised segmentation of retinal vessels, which may hinder the effectiveness of deep learning methods. To address these issues, we propose a novel semi-supervised point consistency network (SPC-Net) for retinal A/V classification. SPC-Net consists of an A/V classification (AVC) module and a multi-class point consistency (MPC) module. The AVC module adopts an encoder-decoder segmentation network to generate the prediction probability map of A/V for supervised learning. The MPC module introduces point set representations to adaptively generate point set classification maps of the arteriovenous skeleton, which enjoys its prediction flexibility and consistency (i.e. point consistency) to effectively alleviate arteriovenous confusion. In addition, we propose a consistency regularization between the predicted A/V classification probability maps and point set representations maps for unlabeled data to explore the inherent segmentation perturbation of the point consistency, reducing the need for annotated data. We validate our method on two typical public datasets (DRIVE, HRF) and a private dataset (TR280) with different resolutions. Extensive qualitative and quantitative experimental results demonstrate the effectiveness of our proposed method for supervised and semi-supervised learning.

Learning with limited annotations: A survey on deep semi-supervised learning for medical image segmentation.

Medical image segmentation is a fundamental and critical step in many image-guided clinical approaches. Recent success of deep learning-based segmentation methods usually relies on a large amount of labeled data, which is particularly difficult and costly to obtain, especially in the medical imaging domain where only experts can provide reliable and accurate annotations. Semi-supervised learning has emerged as an appealing strategy and been widely applied to medical image segmentation tasks to train deep models with limited annotations. In this paper, we present a comprehensive review of recently proposed semi-supervised learning methods for medical image segmentation and summarize both the technical novelties and empirical results. Furthermore, we analyze and discuss the limitations and several unsolved problems of existing approaches. We hope this review can inspire the research community to explore solutions to this challenge and further advance the field of medical image segmentation.

Rational design of functional amyloid fibrillar assemblies.

Amyloid fibrillar assemblies, originally identified as pathological entities in neurodegenerative diseases, have been widely adopted by various proteins to fulfill diverse biological functions in living organisms. Due to their unique features, such as hierarchical assembly, exceptional mechanical properties, environmental stability, and self-healing properties, amyloid fibrillar assemblies have been employed as functional materials in numerous applications. Recently, with the rapid advancement in synthetic biology and structural biology tools, new trends in the functional design of amyloid fibrillar assemblies have begun to emerge. In this review, we provide a comprehensive overview of the design principles for functional amyloid fibrillar assemblies from an engineering perspective, as well as through the lens of structural insights. Initially, we introduce the fundamental structural configurations of amyloid assemblies and highlight the functions of representative examples. We then focus on the underlying design principles of two prevalent strategies for the design of functional amyloid fibrillar assemblies: (1) introducing new functions via protein modular design and/or hybridization, with typical applications encompassing catalysis, virus disinfection, biomimetic mineralization, bio-imaging, and biotherapy; and (2) dynamically regulating living amyloid fibrillar assemblies using synthetic gene circuits, with typical applications in pattern formation, leakage repair, and pressure sensing. Next, we summarize how breakthroughs in characterization techniques have contributed to unveiling the structural polymorphism of amyloid fibrils at the atomic level, and further clarifying the highly diverse regulation mechanisms of amyloid fibrillar assembly and disassembly fine-tuned by various factors. The structural knowledge may significantly aid in the structure-guided design of amyloid fibrillar assemblies with diverse bio-activities and adjustable regulatory properties. Finally, we envision that a new trend in functional amyloid design may emerge by integrating structural tunability, synthetic biology and artificial intelligence.

Retinal vessel caliber and tortuosity and prediction of 5-year incidence of hypertension.

PURPOSE: With arterial hypertension as a global risk factor for cerebrovascular and cardiovascular diseases, we examined whether retinal blood vessel caliber and tortuosity assessed by a vessel-constraint network model can predict the incidence of hypertension. METHODS: The community-based prospective study included 9230 individuals who were followed for 5 years. Ocular fundus photographs taken at baseline were analyzed by a vessel-constraint network model. RESULTS: Within the 5-year follow-up, 1279 (18.8%) and 474 (7.0%) participants out of 6813 individuals free of hypertension at baseline developed hypertension and severe hypertension, respectively. In multivariable analysis, a higher incidence of hypertension was related to a narrower retinal arteriolar diameter ( P  < 0.001), wider venular diameter ( P  = 0.005), and a smaller arteriole-to-venule diameter ratio ( P  < 0.001) at baseline. Individuals with the 5% narrowest arteriole or the 5% widest venule diameter had a 17.1-fold [95% confidence interval (CI):7.9, 37.2] or 2.3-fold (95% CI: 1.4, 3.7) increased risk for developing hypertension, as compared with those with the 5% widest arteriole or the 5% narrowest venule. The area under the receiver operator characteristic curve for predicting the 5-year incidence of hypertension and severe hypertension was 0.791 (95% CI: 0.778, 0.804) and 0.839 (95% CI: 0.821, 0.856), respectively. Although the venular tortuosity was positively associated with the presence of hypertension at baseline ( P  = 0.01), neither arteriolar tortuosity nor venular tortuosity was associated with incident hypertension (both P  ≥ 0.10). CONCLUSION AND RELEVANCE: Narrower retinal arterioles and wider venules indicate an increased risk for incident hypertension within 5 years, while tortuous retinal venules are associated with the presence rather than the incidence of hypertension. The automatic assessment of retinal vessel features performed well in identifying individuals at risk of developing hypertension.

Uncertainty-guided mutual consistency learning for semi-supervised medical image segmentation.

Medical image segmentation is a fundamental and critical step in many clinical approaches. Semi-supervised learning has been widely applied to medical image segmentation tasks since it alleviates the heavy burden of acquiring expert-examined annotations and takes the advantage of unlabeled data which is much easier to acquire. Although consistency learning has been proven to be an effective approach by enforcing an invariance of predictions under different distributions, existing approaches cannot make full use of region-level shape constraint and boundary-level distance information from unlabeled data. In this paper, we propose a novel uncertainty-guided mutual consistency learning framework to effectively exploit unlabeled data by integrating intra-task consistency learning from up-to-date predictions for self-ensembling and cross-task consistency learning from task-level regularization to exploit geometric shape information. The framework is guided by the estimated segmentation uncertainty of models to select out relatively certain predictions for consistency learning, so as to effectively exploit more reliable information from unlabeled data. Experiments on two publicly available benchmark datasets showed that: (1) Our proposed method can achieve significant performance improvement by leveraging unlabeled data, with up to 4.13% and 9.82% in Dice coefficient compared to supervised baseline on left atrium segmentation and brain tumor segmentation, respectively. (2) Compared with other semi-supervised segmentation methods, our proposed method achieve better segmentation performance under the same backbone network and task settings on both datasets, demonstrating the effectiveness and robustness of our method and potential transferability for other medical image segmentation tasks.

The association of magnetoencephalography high-frequency oscillations with epilepsy types and a ripple-based method with source-level connectivity for mapping epilepsy sources.

OBJECTIVE: To explore the association between high-frequency oscillations (HFOs) and epilepsy types and to improve the accuracy of source localization. METHODS: Magnetoencephalography (MEG) ripples of 63 drug-resistant epilepsy patients were detected. Ripple rates, distribution, spatial complexity, and the clustering coefficient of ripple channels were used for the preliminary classification of lateral temporal lobe epilepsy (LTLE), mesial temporal lobe epilepsy (MTLE), and nontemporal lobe epilepsy (NTLE), mainly frontal lobe epilepsy (FLE). Furthermore, the seizure site identification was improved using the Tucker LCMV method and source-level betweenness centrality. RESULTS: Ripple rates were significantly higher in MTLE than in LTLE and NTLE (p < 0.05). The LTLE and MTLE were mainly distributed in the temporal lobe, followed by the parietal lobe, occipital lobe, and frontal lobe, whereas MTLE ripples were mainly distributed in the frontal lobe, then parietal lobe and occipital lobe. Nevertheless, the NTLE ripples were primarily in the frontal lobe and partially in the occipital lobe (p < 0.05). Meanwhile, the spatial complexity of NTLE was significantly higher than that of LTLE and MTLE and was lowest in MTLE (p < 0.01). However, an opposite trend was observed for the standardized clustering coefficient compared with spatial complexity (p < 0.01). Finally, the tucker algorithm showed a higher percentage of ripples at the surgical site when the betweenness centrality was added (p < 0.01). CONCLUSION: This study demonstrated that HFO rates, distribution, spatial complexity, and clustering coefficient of ripple channels varied considerably among the three epilepsy types. Additionally, tucker MEG estimation combined with ripple rates based on the source-level functional connectivity is a promising approach for presurgical epilepsy evaluation.

Cost-efficient and glaucoma-specifical model by exploiting normal OCT images with knowledge transfer learning.

Monitoring the progression of glaucoma is crucial for preventing further vision loss. However, deep learning-based models emphasize early glaucoma detection, resulting in a significant performance gap to glaucoma-confirmed subjects. Moreover, developing a fully-supervised model is suffering from insufficient annotated glaucoma datasets. Currently, sufficient and low-cost normal OCT images with pixel-level annotations can serve as valuable resources, but effectively transferring shared knowledge from normal datasets is a challenge. To alleviate the issue, we propose a knowledge transfer learning model for exploiting shared knowledge from low-cost and sufficient annotated normal OCT images by explicitly establishing the relationship between the normal domain and the glaucoma domain. Specifically, we directly introduce glaucoma domain information to the training stage through a three-step adversarial-based strategy. Additionally, our proposed model exploits different level shared features in both output space and encoding space with a suitable output size by a multi-level strategy. We have collected and collated a dataset called the TongRen OCT glaucoma dataset, including pixel-level annotated glaucoma OCT images and diagnostic information. The results on the dataset demonstrate our proposed model outperforms the un-supervised model and the mixed training strategy, achieving an increase of 5.28% and 5.77% on mIoU, respectively. Moreover, our proposed model narrows performance gap to the fully-supervised model decreased by only 1.01% on mIoU. Therefore, our proposed model can serve as a valuable tool for extracting glaucoma-related features, facilitating the tracking progression of glaucoma.

HRU-Net: A Transfer Learning Method for Carotid Artery Plaque Segmentation in Ultrasound Images.

Carotid artery stenotic plaque segmentation in ultrasound images is a crucial means for the analysis of plaque components and vulnerability. However, segmentation of severe stenotic plaques remains a challenging task because of the heterogeneities of inter-plaques and intra-plaques, and obscure boundaries of plaques. In this paper, we propose an automated HRU-Net transfer learning method for segmenting carotid plaques, using the limited images. The HRU-Net is based on the U-Net encoder−decoder paradigm, and cross-domain knowledge is transferred for plaque segmentation by fine-tuning the pretrained ResNet-50. Moreover, a cropped-blood-vessel image augmentation is customized for the plaque position constraint during training only. Moreover, hybrid atrous convolutions (HACs) are designed to derive diverse long-range dependences for refined plaque segmentation that are used on high-level semantic layers to exploit the implicit discrimination features. The experiments are performed on 115 images; Firstly, the 10-fold cross-validation, using 40 images with severe stenosis plaques, shows that the proposed method outperforms some of the state-of-the-art CNN-based methods on Dice, IoU, Acc, and modified Hausdorff distance (MHD) metrics; the improvements on metrics of Dice and MHD are statistically significant (p < 0.05). Furthermore, our HRU-Net transfer learning method shows fine generalization performance on 75 new images with varying degrees of plaque stenosis, and it may be used as an alternative for automatic noisy plaque segmentation in carotid ultrasound images clinically.

Bridging 2D and 3D segmentation networks for computation-efficient volumetric medical image segmentation: An empirical study of 2.5D solutions.

Recently, deep convolutional neural networks have achieved great success for medical image segmentation. However, unlike segmentation of natural images, most medical images such as MRI and CT are volumetric data. In order to make full use of volumetric information, 3D CNNs are widely used. However, 3D CNNs suffer from higher inference time and computation cost, which hinders their further clinical applications. Additionally, with the increased number of parameters, the risk of overfitting is higher, especially for medical images where data and annotations are expensive to acquire. To issue this problem, many 2.5D segmentation methods have been proposed to make use of volumetric spatial information with less computation cost. Despite these works lead to improvements on a variety of segmentation tasks, to the best of our knowledge, there has not previously been a large-scale empirical comparison of these methods. In this paper, we aim to present a review of the latest developments of 2.5D methods for volumetric medical image segmentation. Additionally, to compare the performance and effectiveness of these methods, we provide an empirical study of these methods on three representative segmentation tasks involving different modalities and targets. Our experimental results highlight that 3D CNNs may not always be the best choice. Despite all these 2.5D methods can bring performance gains to 2D baseline, not all the methods hold the benefits on different datasets. We hope the results and conclusions of our study will prove useful for the community on exploring and developing efficient volumetric medical image segmentation methods.

Extensible and self-recoverable proteinaceous materials derived from scallop byssal thread.

Biologically derived and biologically inspired fibers with outstanding mechanical properties have found attractive technical applications across diverse fields. Despite recent advances, few fibers can simultaneously possess high-extensibility and self-recovery properties especially under wet conditions. Here, we report protein-based fibers made from recombinant scallop byssal proteins with outstanding extensibility and self-recovery properties. We initially investigated the mechanical properties of the native byssal thread taken from scallop Chlamys farreri and reveal its high extensibility (327 ± 32%) that outperforms most natural biological fibers. Combining transcriptome and proteomics, we select the most abundant scallop byssal protein type 5-2 (Sbp5-2) in the thread region, and produce a recombinant protein consisting of 7 tandem repeat motifs (rTRM7) of the Sbp5-2 protein. Applying an organic solvent-enabled drawing process, we produce bio-inspired extensible rTRM7 fiber with high-extensibility (234 ± 35%) and self-recovery capability in wet condition, recapitulating the hierarchical structure and mechanical properties of the native scallop byssal thread. We further show that the mechanical properties of rTRM7 fiber are highly regulated by hydrogen bonding and intermolecular crosslinking formed through disulfide bond and metal-carboxyl coordination. With its outstanding mechanical properties, rTRM7 fiber can also be seamlessly integrated with graphene to create motion sensors and electrophysiological signal transmission electrode.

Photocatalyst-mineralized biofilms as living bio-abiotic interfaces for single enzyme to whole-cell photocatalytic applications.

There is an increasing trend of combining living cells with inorganic semiconductors to construct semi-artificial photosynthesis systems. Creating a robust and benign bio-abiotic interface is key to the success of such solar-to-chemical conversions but often faces a variety of challenges, including biocompatibility and the susceptibility of cell membrane to high-energy damage arising from direct interfacial contact. Here, we report living mineralized biofilms as an ultrastable and biocompatible bio-abiotic interface to implement single enzyme to whole-cell photocatalytic applications. These photocatalyst-mineralized biofilms exhibited efficient photoelectrical responses and were further exploited for diverse photocatalytic reaction systems including a whole-cell photocatalytic CO2 reduction system enabled by the same biofilm-producing strain. Segregated from the extracellularly mineralized semiconductors, the bacteria remained alive even after 5 cycles of photocatalytic NADH regeneration reactions, and the biofilms could be easily regenerated. Our work thus demonstrates the construction of biocompatible interfaces using biofilm matrices and establishes proof of concept for future sustainable photocatalytic applications.

Detection of Schizophrenia Cases From Healthy Controls With Combination of Neurocognitive and Electrophysiological Features.

Background: The search for a method that utilizes biomarkers to identify patients with schizophrenia from healthy individuals has occupied researchers for decades. However, no single indicator can be employed to achieve the good in clinical practice. We aim to develop a comprehensive machine learning pipeline based on neurocognitive and electrophysiological combined features for distinguishing schizophrenia patients from healthy people. Methods: In the present study, 69 patients with schizophrenia and 50 healthy controls participated. Neurocognitive (contains seven specific domains of cognition) and electrophysiological [prepulse inhibition, electroencephalography (EEG) power spectrum, detrended fluctuation analysis, and fractal dimension (FD)] features were collected, all these features were taken together to generate the identification models of schizophrenia by applying logistics, random forest, and extreme gradient boosting algorithm. The classification capabilities of these models were also evaluated. Results: Both the neurocognitive and electrophysiological feature sets showed a good classification effect with the highest accuracy greater than 85% and AUC greater than 90%. Specifically, the performances of the combined neurocognitive and electrophysiological feature sets achieved the highest accuracy of 93.28% and AUC of 97.91%. The extreme gradient boosting algorithm as a whole presented more stably and precisely in classification efficiency. Conclusion: The highest classification accuracy of 93.28% by combination of neurocognitive and electrophysiological features shows that both measurements are appropriate indicators to be used in discriminating schizophrenia patients and healthy individuals. Also, among three algorithms, extreme gradient boosting had better classified performances than logistics and random forest algorithms.

Multi-Scale Interactive Network With Artery/Vein Discriminator for Retinal Vessel Classification.

Automatic classification of retinal arteries and veins plays an important role in assisting clinicians to diagnosis cardiovascular and eye-related diseases. However, due to the high degree of anatomical variation across the population, and the presence of inconsistent labels by the subjective judgment of annotators in available training data, most of existing methods generally suffer from blood vessel discontinuity and arteriovenous confusion, the artery/vein (A/V) classification task still faces great challenges. In this work, we propose a multi-scale interactive network with A/V discriminator for retinal artery and vein recognition, which can reduce the arteriovenous confusion and alleviate the disturbance of noisy label. A multi-scale interaction (MI) module is designed in encoder for realizing the cross-space multi-scale features interaction of fundus images, effectively integrate high-level and low-level context information. In particular, we also design an ingenious A/V discriminator (AVD) that utilizes the independent and shared information between arteries and veins, and combine with topology loss, to further strengthen the learning ability of model to resolve the arteriovenous confusion. In addition, we adopt a sample re-weighting (SW) strategy to effectively alleviate the disturbance from data labeling errors. The proposed model is verified on three publicly available fundus image datasets (AV-DRIVE, HRF, LES-AV) and a private dataset. We achieve the accuracy of 97.47%, 96.91%, 97.79%, and 98.18% respectively on these four datasets. Extensive experimental results demonstrate that our method achieves competitive performance compared with state-of-the-art methods for A/V classification. To address the problem of training data scarcity, we publicly release 100 fundus images with A/V annotations to promote relevant research in the community.

Dynamic brain functional network based on EEG microstate during sensory gating in schizophrenia.

Objective.Cognitive impairment is one of the core symptoms of schizophrenia, with an emphasis on dysfunctional information processing. Sensory gating deficits have consistently been reported in schizophrenia, but the underlying physiological mechanism is not well-understood. We report the discovery and characterization of P50 dynamic brain connections based on microstate analysis.Approach.We identify five main microstates associated with the P50 response and the difference between the first and second click presentation (S1-S2-P50) in first-episode schizophrenia (FESZ) patients, ultra-high-risk individuals (UHR) and healthy controls (HCs). We used the signal segments composed of consecutive time points with the same microstate label to construct brain functional networks.Main results.The microstate with a prefrontal extreme location during the response to the S1 of P50 are statistically different in duration, occurrence and coverage among the FESZ, UHR and HC groups. In addition, a microstate with anterior-posterior orientation was found to be associated with S1-S2-P50 and its coverage was found to differ among the FESZ, UHR and HC groups. Source location of microstates showed that activated brain regions were mainly concentrated in the right temporal lobe. Furthermore, the connectivities between brain regions involved in P50 processing of HC were widely different from those of FESZ and UHR.Significance.Our results indicate that P50 suppression deficits in schizophrenia may be due to both aberrant baseline sensory perception and adaptation to repeated stimulus. Our findings provide new insight into the mechanisms of P50 suppression in the early stage of schizophrenia.

CSM-Net: Automatic joint segmentation of intima-media complex and lumen in carotid artery ultrasound images.

The intima-media thickness (IMT) is an effective biomarker for atherosclerosis, which is commonly measured by ultrasound technique. However, the intima-media complex (IMC) segmentation for the IMT is challenging due to confused IMC boundaries and various noises. In this paper, we propose a flexible method CSM-Net for the joint segmentation of IMC and Lumen in carotid ultrasound images. Firstly, the cascaded dilated convolutions combined with the squeeze-excitation module are introduced for exploiting more contextual features on the highest-level layer of the encoder. Furthermore, a triple spatial attention module is utilized for emphasizing serviceable features on each decoder layer. Besides, a multi-scale weighted hybrid loss function is employed to resolve the class-imbalance issues. The experiments are conducted on a private dataset of 100 images for IMC and Lumen segmentation, as well as on two public datasets of 1600 images for IMC segmentation. For the private dataset, our method obtain the IMC Dice, Lumen Dice, Precision, Recall, and F1 score of 0.814 ± 0.061, 0.941 ± 0.024, 0.911 ± 0.044, 0.916 ± 0.039, and 0.913 ± 0.027, respectively. For the public datasets, we obtain the IMC Dice, Precision, Recall, and F1 score of 0.885 ± 0.067, 0.885 ± 0.070, 0.894 ± 0.089, and 0.885 ± 0.067, respectively. The results demonstrate that the proposed method precedes some cutting-edge methods, and the ablation experiments show the validity of each module. The proposed method may be useful for the IMC segmentation of carotid ultrasound images in the clinic. Our code is publicly available at https://github.com/yuanyc798/US-IMC-code.

Joint Learning of Multi-Level Tasks for Diabetic Retinopathy Grading on Low-Resolution Fundus Images.

Diabetic retinopathy (DR) is a leading cause of permanent blindness among the working-age people. Automatic DR grading can help ophthalmologists make timely treatment for patients. However, the existing grading methods are usually trained with high resolution (HR) fundus images, such that the grading performance decreases a lot given low resolution (LR) images, which are common in clinic. In this paper, we mainly focus on DR grading with LR fundus images. According to our analysis on the DR task, we find that: 1) image super-resolution (ISR) can boost the performance of both DR grading and lesion segmentation; 2) the lesion segmentation regions of fundus images are highly consistent with pathological regions for DR grading. Based on our findings, we propose a convolutional neural network (CNN)-based method for joint learning of multi-level tasks for DR grading, called DeepMT-DR, which can simultaneously handle the low-level task of ISR, the mid-level task of lesion segmentation and the high-level task of disease severity classification on LR fundus images. Moreover, a novel task-aware loss is developed to encourage ISR to focus on the pathological regions for its subsequent tasks: lesion segmentation and DR grading. Extensive experimental results show that our DeepMT-DR method significantly outperforms other state-of-the-art methods for DR grading over three datasets. In addition, our method achieves comparable performance in two auxiliary tasks of ISR and lesion segmentation.