Search

Open set recognition (OSR) for medical images is vital to ensure practical safeguards in clinical applications. It demands establishing model awareness of rare and unknown conditions, rejecting what the model does not know, and thus preventing these conditions from being dangerously misclassified into any false known classes. Meanwhile, it often further requires training the OSR model based on a mixture of severely class-imbalanced and partially labeled data due to the prevalence of long-tailed data distribution and the massive cost of medical image annotations. Unfortunately, most existing OSR methods perform poorly under such complex conditions, largely because they neglect to thoroughly exploit the knowledge from the imbalanced and unlabeled data to help better distinguish the known and unknown lesion patterns. This paper proposes a novel Semi-supervised and Class-Imbalanced Open Set medical image Recognition (SCI-OSR) algorithm as a joint solution to address these issues. The major contributions are as follows. Firstly, we propose a modified open mixup approach boasting selective fine-grained lesion foreground fusion based on the learned weighted attentive masks to augment the unlabeled, small-sized, and even pseudo-unknown class data, where the generated attentive masks serve as effective candidate lesion area estimators. Secondly, our approach can more effectively extract the disentangled and discriminative feature representations based on our tailored consistency and contrastive losses, which helps more properly exploit the augmented data and aids in pushing the learned features of known and unknown classes to be well separated by large margins. Finally, a prototype-based open set classifier is established based on the extracted discriminative features of the original and augmented data. It favors producing equally low confidence scores over all non-positive classes and thus more reliably rejects the unknown query data. Through comparative experiments and ablation studies conducted on several popular medical benchmarks, we demonstrate that our proposed SCI-OSR can successfully boost OSR performances, leading to improved known class classification and unknown class rejection over other state-of-the-art methods.

Globally, age-related macular degeneration (AMD) is the leading cause of irreversible visual impairment. Up to 80% of severe vision loss is caused by AMD, which is characterized by the development of choroidal neovascularization (CNV). Uncertainty exists regarding the precise pathophysiological mechanisms of CNV. It has been suggested that the interleukin (IL) IL-6/IL-6R signaling pathway is crucial in the progression of CNV. Tocilizumab (TCZ), a monoclonal antibody, binds to soluble and membrane-bound IL-6R and competitively inhibits IL-6 downstream signaling. Previous research has demonstrated that TCZ promotes several roles related to inflammation and neovascularization. However, the effects of TCZ on CNV and the underlying mechanism are still unknown. This study found that TCZ administration decreased the area and leakage of CNV lesions in the mice model of laser-induced CNV. Additionally, results demonstrated that TCZ promotes the expression of iNOS, CCL-3, CCL-5, TNF-α and inhibits the expression of Arg-1, IL-10, YM-1 and CD206. Furthermore, TCZ treatment inhibited the signal transducer and activator of transcription (STAT) STAT3/vascular endothelial growth factor (VEGF) pathway, which was activated after CNV formation. Colivelin, a STAT3 agonist, reversed the inhibitory effects of TCZ on CNV formation and macrophage polarization. In a mouse model of laser-induced CNV, our findings demonstrated that TCZ attenuated CNV formation and inhibited the leakage of CNV lesions by regulating macrophage polarization via inhibiting the STAT3/VEGF axis. TCZ is the potential therapeutic strategy for CNV.

Exhaled ammonia (NH3) is an essential noninvasive biomarker for disease diagnosis. In this study, an acetone-modifier positive photoionization ion mobility spectrometry (AM-PIMS) method was developed for accurate qualitative and quantitative analysis of exhaled NH3 with high selectivity and sensitivity. Acetone was introduced into the drift tube along with the drift gas as a modifier, and the characteristic NH3 product ion peak of (C3H6O)4NH4+ (K0 = 1.45 cm2/V·s) was obtained through the ion-molecule reaction with acetone reactant ions (C3H6O)2H+ (K0 = 1.87 cm2/V·s), which significantly increased the peak-to-peak resolution and improved the accuracy of exhaled NH3 qualitative identification. Moreover, the interference of high humidity and the memory effect of NH3 molecules were significantly reduced via online dilution and purging sampling, thus realizing breath-by-breath measurement. As a result, a wide quantitative range of 5.87–140.92 μmol/L with a response time of 40 ms was achieved, and the exhaled NH3 profile could be synchronized with the concentration curve of exhaled CO2. Finally, the analytical capacity of AM-PIMS was demonstrated by measuring the exhaled NH3 of healthy subjects, demonstrating its great potential for clinical disease diagnosis.

The pandemic has made wearing masks commonplace, prompting researchers to investigate their effects on interpersonal perception. Findings indicate masks obstruct face identification and expression recognition, with lower face cues being most affected. When judging attractiveness, masks can enhance the appeal of less attractive faces, but reduce the appeal of more attractive faces. Trust and speech perception outcomes are inconclusive. Future studies could focus on individual differences in how masks influence our perception of others.

Abstract Background Coronary atherosclerosis (CA) is the most common type of atherosclerosis. However, the inherent pathogenesis and mechanisms of CA are unclear, and the relationship with ferroptosis-related genes (FRGs) has not been reported. The purpose of this study was to use bioinformatics techniques to evaluate potential therapeutic targets for CA.Please provide the given name for author “Dingshun”.Please provide the given name for author “Dingshun”. Methods First, the GSE132651 dataset was acquired from the Gene Expression Omnibus database. Gene Ontology enrichment analysis, Kyoto Encyclopedia of Genes and Genomes enrichment analysis, and Protein–Protein interaction network were successively conducted. Next, overlapping genes between hub genes and CA genes were found. FRGs were found when comparing the CA group with the normal group. The correlation between overlapping genes and FRGs was further analyzed. At last, we performed Elisa to validate the expression of these genes in human blood specimens. Mice aortic tissues were used for western blot to detect the expression of proteins. Results Based on the GSE132651 dataset, 102 differentially expressed genes were identified. Five overlapping genes between hub genes and CA genes were found (CCNA2, RRM2, PBK, PCNA, CDK1). TFRC and GPX4 were found to be FRGs. TFRC was positively correlated with CCNA2, PBK, PCNA, CDK1, RRM2, with CDK1 being the strongest correlation. GPX4 was negatively correlated with these genes, among which CCNA2 was the strongest correlation. The ELISA results showed that CCNA2, CDK1, and TFRC expression were markedly increased in serum of the CA samples compared with controls, while GPX4 expression was markedly decreased in the CA samples. The western blot results show that GPX4 expression was lower in the model group, TFRC, CDK1, and CCNA2 protein expression were high in the model group. Conclusions Ferroptosis-related genes GPX4 and TFRC were closely correlated with the identified overlapping genes CCNA2 and CDK1, which may serve as targeted therapies for the treatment of CA.

Fusarium heading blight (FHB) is a devastating disease in wheat, primarily caused by field invasion of Fusarium graminearum. Due to the scarcity of resistant wheat varieties, the agricultural sector resorts to chemical fungicides to control FHB incidence. On the other hand, biocontrol represents a promising, eco-friendly approach aligned with sustainable and green agriculture concepts. In the present study, a bacterial endophyte, Pseudescherichia sp. (GSE25), was isolated from wheat seeds and identified through complete genome sequencing and phylogenetic analysis. In vitro testing of this endophytic strain demonstrated strong antifungal activity against F. graminearum PH-1 by inhibiting spore germination, suppressing germ tube growth, and causing cell membrane damage. Under field conditions, the strain GSE25 significantly reduced the FHB incidence and the associated deoxynivalenol mycotoxin accumulation by over 60% and 80%, respectively. These findings highlight the potential of the isolated bacterial endophyte Pseudescherichia sp. GSE25 strain as a biocontrol agent in protecting wheat from FHB-caused F. graminearum. This is the first report showing a biocontrol effect of Pseudescherichia sp. a strain against phytopathogens.

To confirm whether Chinese mitten crabs (Eriocheir sinensis), commonly known as hairy crabs or river crabs, in non-Yangcheng Lake areas undergo morphological convergence with the original crabs in the Yangcheng Lake purse seine and high-standard modified aquaculture ponds after being “introduced” or “bathed”-cultured, we employed a geometric morphometrics approach. This approach allowed us to compare and analyze the dynamic changes in the carapace morphology of both the original and “introduced” crabs in Yangcheng Lake and high-standard ponds in Kunshan City at 0, 7, 14, and 30 days after “bathing” culture. The geometric morphological analysis of the carapace was conducted using a system of 35 established landmarks. The stepwise discriminant analysis of the relative distortion score revealed morphological differences between “introduced” and resident original crabs in the Yangcheng Lake area and in high-standard modified ponds after 7, 14, and 30 days of bathing culture. The accuracy of the discriminant analysis was 100%. The results of the geometric morphological visualization demonstrated that the carapaces of the bathed crabs underwent adaptive changes in the water. However, even after one month of bathing culture, the “introduced” crabs in lakes and ponds could not reach the morphological characteristics of the original crabs, i.e., their carapaces did not exhibit the characteristics of the original crabs.

Electric potential produced in deformed piezoelectric nanostructures is of significance for both fundamental study and practical applications. To reveal the piezoelectric property of ZnO nanohelices, the piezoelectric potential in single-crystal nanohelices was simulated by finite element method calculations. For a nanohelix with a length of 1200 nm, a mean coil radius of 150 nm, five active coils, and a hexagonal coiled wire with a side length 100 nm, a compressing force of 100 nN results in a potential of 1.85 V. This potential is significantly higher than the potential produced in a straight nanowire with the same length and applied force. Maintaining the length and increasing the number of coils or mean coil radius leads to higher piezoelectric potential in the nanohelix. Appling a force along the axial direction produces higher piezoelectric potential than in other directions. Adding lateral forces to an existing axial force can change the piezoelectric potential distribution in the nanohelix, while the maximum piezoelectric potential remains largely unchanged in some cases. This research demonstrates the promising potential of ZnO nanohelices for applications in sensors, micro-electromechanical systems (MEMS) devices, nanorobotics, and energy sciences.