From zinc homeostasis to disease progression: Unveiling the neurodegenerative puzzle.

Zinc is a crucial trace element in the human body, playing a role in various physiological processes such as oxidative stress, neurotransmission, protein synthesis, and DNA repair. The zinc transporters (ZnTs) family members are responsible for exporting intracellular zinc, while Zrt- and Irt-like proteins (ZIPs) are involved in importing extracellular zinc. These processes are essential for maintaining cellular zinc homeostasis. Imbalances in zinc metabolism have been linked to the development of neurodegenerative diseases. Disruptions in zinc levels can impact the survival and activity of neurons, thereby contributing to the progression of neurodegenerative diseases through mechanisms like cell apoptosis regulation, protein phase separation, ferroptosis, oxidative stress, and neuroinflammation. Therefore, conducting a systematic review of the regulatory network of zinc and investigating the relationship between zinc dysmetabolism and neurodegenerative diseases can enhance our understanding of the pathogenesis of these diseases. Additionally, it may offer new insights and approaches for the treatment of neurodegenerative diseases.

The metal ion hypothesis of Alzheimer's disease and the anti-neuroinflammatory effect of metal chelators.

Alzheimer's disease (AD), characterized by the ß-amyloid protein (Aß) deposition and tau hyperphosphorylation, is the most common dementia with uncertain etiology. The clinical trials of Aß monoclonal antibody drugs have almost failed, giving rise to great attention on the other etiologic hypothesis regarding AD such as metal ions dysmetabolism and chronic neuroinflammation. Mounting evidence revealed that the metal ions (iron, copper, and zinc) were dysregulated in the susceptible brain regions of AD patients, which was highly associated with Aß deposition, tau hyperphosphorylation, neuronal loss, as well as neuroinflammation. Further studies uncovered that iron, copper and zinc could not only enhance the production of Aß but also directly bind to Aß and tau to promote their aggregations. In addition, the accumulation of iron and copper could respectively promote ferroptosis and cuproptosis. Therefore, the metal ion chelators were recognized as promising agents for treating AD. This review comprehensively summarized the effects of metal ions on the Aß dynamics and tau phosphorylation in the progression of AD. Furthermore, taking chronic neuroinflammation contributes to the progression of AD, we also provided a summary of the mechanisms concerning metal ions on neuroinflammation and highlighted the metal ion chelators may be potential agents to alleviate neuroinflammation under the condition of AD. Nevertheless, more investigations regarding metal ions on neuroinflammation should be taken into practice, and the effects of metal ion chelators on neuroinflammation should gain more attention. Running title: Metal chelators against neuroinflammation.

Multicenter Study of the Utility of Convolutional Neural Network and Transformer Models for the Detection and Segmentation of Meningiomas.

PURPOSE: This study aimed to investigate the effectiveness and practicality of using models like convolutional neural network and transformer in detecting and precise segmenting meningioma from magnetic resonance images. METHODS: The retrospective study on T1-weighted and contrast-enhanced images of 523 meningioma patients from 3 centers between 2010 and 2020. A total of 373 cases split 8:2 for training and validation. Three independent test sets were built based on the remaining 150 cases. Six convolutional neural network detection models trained via transfer learning were evaluated using 4 metrics and receiver operating characteristic analysis. Detected images were used for segmentation. Three segmentation models were trained for meningioma segmentation and were evaluated via 4 metrics. In 3 test sets, intraclass consistency values were used to evaluate the consistency of detection and segmentation models with manually annotated results from 3 different levels of radiologists. RESULTS: The average accuracies of the detection model in the 3 test sets were 97.3%, 93.5%, and 96.0%, respectively. The model of segmentation showed mean Dice similarity coefficient values of 0.884, 0.834, and 0.892, respectively. Intraclass consistency values showed that the results of detection and segmentation models were highly consistent with those of intermediate and senior radiologists and lowly consistent with those of junior radiologists. CONCLUSIONS: The proposed deep learning system exhibits advanced performance comparable with intermediate and senior radiologists in meningioma detection and segmentation. This system could potentially significantly improve the efficiency of the detection and segmentation of meningiomas.

Relationships between X-ray Diffraction Peaks, Molecular Components, and Heat Properties of C-Type Starches from Different Sweet Potato Varieties.

C-type starches with different proportions of A- and B-type crystallinities have different intensities and crystallinities of X-ray diffraction peaks. In this study, the intensities and crystallinities of X-ray diffraction peaks, molecular components and heat properties of C-type starches were investigated in seven sweet potato varieties, and their relationships were analyzed. The intensity and crystallinity of a diffraction peak at 5.6° were significantly positively correlated to the DP6-12 branch-chains of amylopectin and significantly negatively correlated to the true amylose content (TAC) determined by concanavalin A precipitation, gelatinization temperature, gelatinization enthalpy, water solubility at 95 °C, and pasting temperature. The intensity of diffraction peaks at 15° and 23° were significantly positively correlated to the gelatinization temperature and pasting temperature and significantly negatively correlated to the pasting peak viscosity. The significantly positive relationships were detected between the crystallinity of a diffraction peak at 15° and the DP13-24 branch-chains of amylopectin, gelatinization conclusion temperature and water solubility, between the crystallinity of diffraction peak at 17-18° and the TAC, gelatinization onset temperature, water solubility and pasting temperature, between the crystallinity of a diffraction peak at 23° and the gelatinization conclusion temperature and pasting peak time, and between the total crystallinity and the TAC, gelatinization conclusion temperature, water solubility and pasting temperature. The score plot of principle component analysis showed that the molecular components and heat property parameters could differentiate the C-type starches and agreed with their characteristics of X-ray diffraction peaks. This study provides some references for the utilizations of C-type starches.

Sizes, Components, Crystalline Structure, and Thermal Properties of Starches from Sweet Potato Varieties Originating from Different Countries.

Sweet potato is a root tuber crop and an important starch source. There are hundreds of sweet potato varieties planted widely in the world. Starches from varieties with different genotype types and originating from different countries have not been compared for their physicochemical properties. In the research, starches from 44 sweet potato varieties originating from 15 countries but planted in the same growing conditions were investigated for their physicochemical properties to reveal the similarities and differences in varieties. The results showed that the 44 starches had granule size (D[4,3]) from 8.01 to 15.30 µm. Starches had different iodine absorption properties with OD680 from 0.259 to 0.382 and OD620/550 from 1.142 to 1.237. The 44 starches had apparent amylose content from 19.2% to 29.2% and true amylose content from 14.2% to 20.2%. The starches exhibited A-, CA-, CC-, or CB-type X-ray diffraction patterns. The thermograms of 44 starches exhibited one-, two-, or three-peak curves, leading to a significantly different gelatinization temperature range from 13.1 to 29.2 °C. The significantly different starch properties divide the 44 sweet potato varieties into different groups due to their different genotype backgrounds. The research offers references for the utilization of sweet potato germplasm.

Comparative Metabolomic and Transcriptomic Analyses of Phytochemicals in Two Elite Sweet Potato Cultivars for Table Use.

To elucidate nutritional components in sweet potato cultivars for table use and to compare the phytochemicals of cultivars from different countries, 'Kokei No. 14' and 'Xinxiang' were selected. The physiological parameters and metabolites were determined using the colorimetric method and widely targeted metabolomics, respectively. Transcriptomic analysis was performed to explain the mechanism that resulted in phytochemical differences. 'Xinxiang' showed higher flavonoid and carotenoid contents. Metabolomics showed five upregulated flavonoids. Two essential amino acids (EAAs) and one conditionally essential amino acid (CEAA) were upregulated, whereas four EAAs and two CEAAs were downregulated. Unlike lipids, in which only one of thirty-nine was upregulated, nine of twenty-seven differentially accumulated phenolic acids were upregulated. Three of the eleven different alkaloids were upregulated. Similarly, eight organic acids were downregulated, with two upregulated. In addition, three of the seventeen different saccharides and alcohols were upregulated. In 'other metabolites,' unlike vitamin C, 6'-O-Glucosylaucubin and pantetheine were downregulated. The differentially accumulated metabolites were enriched to pathways of the biosynthesis of secondary metabolites, ABC transporters, and tyrosine metabolism, whereas the differentially expressed genes were mainly concentrated in the metabolic pathway, secondary metabolite biosynthesis, and transmembrane transport functions. These results will optimize the sweet potato market structure and enable a healthier diet for East Asian residents.

The Kernel Size-Related Quantitative Trait Locus qKW9 Encodes a Pentatricopeptide Repeat Protein That Aaffects Photosynthesis and Grain Filling.

In maize (Zea mays), kernel weight is an important component of yield that has been selected during domestication. Many genes associated with kernel weight have been identified through mutant analysis. Most are involved in the biogenesis and functional maintenance of organelles or other fundamental cellular activities. However, few quantitative trait loci (QTLs) underlying quantitative variation in kernel weight have been cloned. Here, we characterize a QTL, qKW9, associated with maize kernel weight. This QTL encodes a DYW motif pentatricopeptide repeat protein involved in C-to-U editing of ndhB, a subunit of the chloroplast NADH dehydrogenase-like complex. In a null qkw9 background, C-to-U editing of ndhB was abolished, and photosynthesis was reduced, resulting in less maternal photosynthate available for grain filling. Characterization of qKW9 highlights the importance of optimizing photosynthesis for maize grain yield production.

A Novel Mutation of OsPPDKB, Encoding Pyruvate Orthophosphate Dikinase, Affects Metabolism and Structure of Starch in the Rice Endosperm.

Starch, as a main energy storage substance, plays an important role in plant growth and human life. Despite the fact that several enzymes and regulators involved in starch biosynthesis have been identified, the regulating mechanism of starch synthesis is still unclear. In this study, we isolated a rice floury endosperm mutant M14 from a mutant pool induced by 60Co. Both total starch content and amylose content in M14 seeds significantly decreased, and starch thermal and pasting properties changed. Compound starch granules were defected in the floury endosperm of M14 seeds. Map-based cloning and a complementation test showed that the floury endosperm phenotype was determined by a gene of OsPPDKB, which encodes pyruvate orthophosphate dikinase (PPDK, EC 2.7.9.1). Subcellular localization analysis demonstrated that PPDK was localized in chloroplast and cytoplasm, the chOsPPDKB highly expressed in leaf and leaf sheath, and the cyOsPPDKB constitutively expressed with a high expression in developing endosperm. Moreover, the expression of starch synthesis-related genes was also obviously altered in M14 developing endosperm. The above results indicated that PPDK played an important role in starch metabolism and structure in rice endosperm.

[Prediction of seizures in sleep based on power spectrum].

Seizures during sleep increase the probability of complication and sudden death. Effective prediction of seizures in sleep allows doctors and patients to take timely treatments to reduce the aforementioned probability. Most of the existing methods make use of electroencephalogram (EEG) to predict seizures, which are not specific developed for the sleep. However, EEG during sleep has its characteristics compared with EEG during other states. Therefore, in order to improve the sensitivity and reduce the false alarm rate, this paper utilized the characteristics of EEG to predict seizures during sleep. We firstly constructed the feature vector including the absolute power spectrum, the relative power spectrum and the power spectrum ratio in different frequencies. Secondly, the separation criterion and branch-and-bound method were applied to select features. Finally, support vector machine classifier were trained, which is then employed for online prediction. Compared with the existing method that do not consider the characteristics of sleeping EEG (sensitivity 91.67%, false alarm rate 9.19%), the proposed method was superior in terms of sensitivity (100%) and false alarm rate (2.11%). This method can improve the existing epilepsy prediction methods and has important clinical value.

Unified Phase Diagram for Iron-Based Superconductors.

High-temperature superconductivity is closely adjacent to a long-range antiferromagnet, which is called a parent compound. In cuprates, all parent compounds are alike and carrier doping leads to superconductivity, so a unified phase diagram can be drawn. However, the properties of parent compounds for iron-based superconductors show significant diversity and both carrier and isovalent dopings can cause superconductivity, which casts doubt on the idea that there exists a unified phase diagram for them. Here we show that the ordered moments in a variety of iron pnictides are inversely proportional to the effective Curie constants of their nematic susceptibility. This unexpected scaling behavior suggests that the magnetic ground states of iron pnictides can be achieved by tuning the strength of nematic fluctuations. Therefore, a unified phase diagram can be established where superconductivity emerges from a hypothetical parent compound with a large ordered moment but weak nematic fluctuations, which suggests that iron-based superconductors are strongly correlated electron systems.

Association of Metal Exposure with Novel Immunoinflammatory Indicators.

Objective: We aimed to investigate the relationship between metal exposure and novel immunoinflammatory indicators. Methods: Data on adults participating in the National Health and Nutrition Examination Survey (NHANES) from 2009 to 2018 were analyzed. Various statistical models were employed to assess the association between metal exposure and novel immune-inflammation-related indicators. Additionally, the impact of metal exposure on inflammation in different gender populations was explored. Results: This study included 4482 participants, of whom 51.1% were male. Significant correlations were observed among various metals. Both elastic net (ENET) and linear regression models revealed robust associations between cadmium (Cd), cobalt (Co), arsenic (As), mercury (Hg), and immunoinflammatory indicators. Weighted quantile sum (WQS) and Quantile g-computation (Q-gcomp) models demonstrated strong associations between barium (Ba), Co, and Hg and immunoinflammatory indicators. Bayesian kernel machine regression (BKMR) analysis indicated an overall positive correlation between in vivo urinary metal levels and systemic inflammation response index (SIRI) and aggregate index of systemic inflammation (AISI). Furthermore, Co, As, and Hg emerged as key metals contributing to changes in novel immunoinflammatory indicators. Conclusions: Metals exhibit associations with emerging immunoinflammatory indicators, and concurrent exposure to mixed metals may exacerbate the inflammatory response. Furthermore, this relationship varies across gender populations.

Meningioma consistency assessment based on the fusion of deep learning features and radiomics features.

PURPOSE: This study aims to combine deep learning features with radiomics features for the computer-assisted preoperative assessment of meningioma consistency. METHODS: 202 patients with surgery and pathological diagnosis of meningiomas at our institution between December 2016 and December 2018 were retrospectively included in the study. The T2-fluid attenuated inversion recovery (T2-Flair) images were evaluated to classify meningioma as soft or hard by professional neurosurgeons based on Zada's consistency grading system. All the patients were split randomly into a training cohort (n = 162) and a testing cohort (n = 40). A convolutional neural network (CNN) model was proposed to extract deep learning features. These deep learning features were combined with radiomics features. After multiple feature selections, selected features were used to construct classification models using four classifiers. AUC was used to evaluate the performance of each classifier. A signature was further constructed by using the least absolute shrinkage and selection operator (LASSO). A nomogram based on the signature was created for predicting meningioma consistency. RESULTS: The logistic regression classifier constructed using 17 radiomics features and 9 deep learning features provided the best performance with a precision of 0.855, a recall of 0.854, an F1-score of 0.852 and an AUC of 0.943 (95 % CI, 0.873-1.000) in the testing cohort. The C-index of the nomogram was 0.822 (95 % CI, 0.758-0.885) in the training cohort and 0.943 (95 % CI, 0.873-1.000) in the testing cohort with good calibration. Decision curve analysis further confirmed the clinical usefulness of the nomogram for predicting meningioma consistency. CONCLUSIONS: The proposed method for assessing meningioma consistency based on the fusion of deep learning features and radiomics features is potentially clinically valuable. It can be used to assist physicians in the preoperative determination of tumor consistency.

Potential Application Prospects of Biomolecule-Modified Two-Dimensional Chiral Nanomaterials in Biomedicine.

Chirality, one of the most fundamental properties of natural molecules, plays a significant role in biochemical reactions. Nanomaterials with chiral characteristics have superior properties, such as catalytic properties, optoelectronic properties, and photothermal properties, which have significant potential for specific applications in nanomedicine. Biomolecular modifications such as nucleic acids, peptides, proteins, and polysaccharides are sources of chirality for nanomaterials with great potential for application in addition to intrinsic chirality, artificial macromolecules, and metals. Two-dimensional (2D) nanomaterials, as opposed to other dimensions, due to proper surface area, extensive modification sites, drug loading potential, and simplicity of preparation, are prepared and utilized in diagnostic applications, drug delivery research, and tumor therapy. Current advanced studies on 2D chiral nanomaterials for biomedicine are focused on novel chiral development, structural control, and materials sustainability applications. However, despite the advances in biomedical research, chiral 2D nanomaterials still confront challenges such as the difficulty of synthesis, quality control, batch preparation, chiral stability, and chiral recognition and selectivity. This review aims to provide a comprehensive overview of the origins, synthesis, applications, and challenges of 2D chiral nanomaterials with biomolecules as cargo and chiral modifications and highlight their potential roles in biomedicine.

Construction of high-density genetic map based on SLAF-seq and QTL analysis of major traits in sweetpotato [Ipomoea batatas (L.) Lam.].

Sweetpotato, Ipomoea batatas (L.) Lam., is an important worldwide crop used as feed, food, and fuel. However, its polyploidy, high heterozygosity and self-incompatibility makes it difficult to study its genetics and genomics. Longest vine length (LVL), yield per plant (YPP), dry matter content (DMC), starch content (SC), soluble sugar content (SSC), and carotenoid content (CC) are some of the major agronomic traits being used to evaluate sweetpotato. However limited research has actually examined how these traits are inherited. Therefore, after selecting 212 F1 from a Xin24 × Yushu10 crossing as the mapping population, this study applied specific-locus amplified fragment sequencing (SLAF-seq), at an average sequencing depth of 26.73 × (parents) and 52.25 × (progeny), to detect single nucleotide polymorphisms (SNPs). This approach generated an integrated genetic map of length 2441.56 cM and a mean distance of 0.51 cM between adjacent markers, encompassing 15 linkage groups (LGs). Based on the linkage map, 26 quantitative trait loci (QTLs), comprising six QTLs for LVL, six QTLs for YPP, ten QTLs for DMC, one QTL for SC, one QTL for SSC, and two QTLs for CC, were identified. Each of these QTLs explained 6.3-10% of the phenotypic variation. It is expected that the findings will be of benefit for marker-assisted breeding and gene cloning of sweetpotato.

Astrocyte-derived lactoferrin reduces ß-amyloid burden by promoting the interaction between p38 kinase and PP2A phosphatase in male APP/PS1 transgenic mice.

BACKGROUND AND PURPOSE: Overexpression of astrocytic lactoferrin (Lf) was observed in the brain of Alzheimer's disease (AD) patients, whereas the role of astrocytic Lf in AD progression remains unexplored. In this study, we aimed to evaluate the effects of astrocytic Lf on AD progression. EXPERIMENTAL APPROACH: Male APP/PS1 mice with astrocytes overexpressing human Lf were developed to evaluate the effects of astrocytic Lf on AD progression. N2a-sw cells also were employed to further uncover the mechanism of astrocytic Lf on ß-amyloid (Aß) production. KEY RESULTS: Astrocytic Lf overexpression increased protein phosphatase 2A (PP2A) activity and reduced amyloid precursor protein (APP) phosphorylation, Aß burden and tau hyperphosphorylation in APP/PS1 mice. Mechanistically, astrocytic Lf overexpression promoted the uptake of astrocytic Lf into neurons in APP/PS1 mice, and conditional medium from astrocytes overexpressing Lf inhibited p-APP (Thr668) expression in N2a-sw cells. Furthermore, recombinant human Lf (hLf) significantly enhanced PP2A activity and inhibited p-APP expression, whereas inhibition of p38 or PP2A activities abrogated the hLf-induced p-APP down-regulation in N2a-sw cells. Additionally, hLf promoted the interaction of p38 and PP2A via p38 activation, thereby enhancing PP2A activity, and low-density lipoprotein receptor-related protein 1 (LRP1) knockdown significantly reversed the hLf-induced p38 activation and p-APP down-regulation. CONCLUSIONS AND IMPLICATIONS: Our data suggested that astrocytic Lf promoted neuronal p38 activation, via targeting to LRP1, subsequently promoting p38 binding to PP2A to enhance PP2A enzyme activity, which finally inhibited Aß production via APP dephosphorylation. In conclusion, promoting astrocytic Lf expression may be a potential strategy against AD. LINKED ARTICLES: This article is part of a themed issue From Alzheimer's Disease to Vascular Dementia: Different Roads Leading to Cognitive Decline. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v181.6/issuetoc.

Disorder-Broadened Phase Boundary with Enhanced Amorphous Superconductivity in Pressurized In2Te5.

As an empirical tool in materials science and engineering, the iconic phase diagram owes its robustness and practicality to the topological characteristics rooted in the celebrated Gibbs phase law free variables (F) = components (C) - phases (P) + 2. When crossing the phase diagram boundary, the structure transition occurs abruptly, bringing about an instantaneous change in physical properties and limited controllability on the boundaries (F = 1). Here, the sharp phase boundary is expanded to an amorphous transition region (F = 2) by partially disrupting the long-range translational symmetry, leading to a sequential crystalline-amorphous-crystalline (CAC) transition in a pressurized In2Te5 single crystal. Through detailed in situ synchrotron diffraction, it is elucidated that the phase transition stems from the rotation of immobile blocks [In2Te2]2+, linked by hinge-like [Te3]2- trimers. Remarkably, within the amorphous region, the amorphous phase demonstrates a notable 25% increase of the superconducting transition temperature (Tc), while the carrier concentration remains relatively constant. Furthermore, a theoretical framework is proposed revealing that the unconventional boost in amorphous superconductivity might be attributed to an intensified electron correlation, triggered by a disorder-augmented multifractal behavior. These findings underscore the potential of disorder and prompt further exploration of unforeseen phenomena on the phase boundaries.

On Using Classification Datasets to Evaluate Graph Outlier Detection: Peculiar Observations and New Insights.

It is common practice of the outlier mining community to repurpose classification datasets toward evaluating various detection models. To that end, often a binary classification dataset is used, where samples from one of the classes are designated as the "inlier" samples, and the other class is substantially down-sampled to create the (ground-truth) "outlier" samples. Graph-level outlier detection (GLOD) is rarely studied but has many potentially influential real-world applications. In this study, we identify an intriguing issue with repurposing graph classification datasets for GLOD. We find that ROC-AUC performance of the models changes significantly ("flips" from high to very low, even worse than random) depending on which class is down-sampled. Interestingly, ROC-AUCs on these two variants approximately sum to 1 and their performance gap is amplified with increasing propagations for a certain family of propagation-based outlier detection models. We carefully study the graph embedding space produced by propagation-based models and find two driving factors: (1) disparity between within-class densities, which is amplified by propagation, and (2) overlapping support (mixing of embeddings) across classes. We also study other graph embedding methods and downstream outlier detectors, and we find that the intriguing "performance flip" issue still widely exists but which version of the down-sample achieves higher performance may vary. Thoughtful analysis over comprehensive results further deepens our understanding of the established issue. With this study, we aim at drawing attention to this (to our knowledge) previously unnoticed issue for the rarely studied GLOD problem, and specifically to the following questions: (1) Given the performance flip issue we identified, where one version of the down-sample often yields worse-than-random performance, is it appropriate to evaluate GLOD by average performance across all down-sampled versions when repurposing graph classification datasets? (2) Considering consistently observed performance flip issue across different graph embedding methods we studied, is it possible to design better graph embedding methods to overcome the issue? We conclude the article with our insights to these questions.

Fast and Accurate Illumination Estimation Using LDR Panoramic Images for Realistic Rendering.

A high dynamic range (HDR) image is commonly used to reveal stereo illumination, which is crucial for generating high-quality realistic rendering effects. Compared to the high-cost HDR imaging technique, low dynamic range (LDR) imaging provides a low-cost alternative and is preferable for interactive graphics applications. However, the limited LDR pixel bit depth significantly bothers accurate illumination estimation using LDR images. The conflict between the realism and promptness of illumination estimation for realistic rendering is yet to be resolved. In this paper, an efficient method that accurately infers illuminations of real-world scenes using LDR panoramic images is proposed. It estimates multiple lighting parameters, including locations, types and intensities of light sources. In our approach, a new algorithm that extracts illuminant characteristics during the exposure attenuation process is developed to locate light sources and outline their boundaries. To better predict realistic illuminations, a new deep learning model is designed to efficiently parse complex LDR panoramas and classify detected light sources. Finally, realistic illumination intensities are calculated by recovering the inverse camera response function and extending the dynamic range of pixel values based on previously estimated parameters of light sources. The reconstructed radiance map can be used to compute high-quality image-based lighting of virtual models. Experimental results demonstrate that the proposed method is capable of efficiently and accurately computing comprehensive illuminations using LDR images. Our method can be used to produce better realistic rendering results than existing approaches.

Detection of Image Artifacts Using Improved Cascade Region-Based CNN for Quality Assessment of Endoscopic Images.

Endoscopy is a commonly used clinical method for gastrointestinal disorders. However, the complexity of the gastrointestinal environment can lead to artifacts. Consequently, the artifacts affect the visual perception of images captured during endoscopic examinations. Existing methods to assess image quality with no reference display limitations: some are artifact-specific, while others are poorly interpretable. This study presents an improved cascade region-based convolutional neural network (CNN) for detecting gastrointestinal artifacts to quantitatively assess the quality of endoscopic images. This method detects eight artifacts in endoscopic images and provides their localization, classification, and confidence scores; these scores represent image quality assessment results. The artifact detection component of this method enhances the feature pyramid structure, incorporates the channel attention mechanism into the feature extraction process, and combines shallow and deep features to improve the utilization of spatial information. The detection results are further used for image quality assessment. Experimental results using white light imaging, narrow-band imaging, and iodine-stained images demonstrate that the proposed artifact detection method achieved the highest average precision (62.4% at a 50% IOU threshold). Compared to the typical networks, the accuracy of this algorithm is improved. Furthermore, three clinicians validated that the proposed image quality assessment method based on the object detection of endoscopy artifacts achieves a correlation coefficient of 60.71%.

AI-assisted identification of intrapapillary capillary loops in magnification endoscopy for diagnosing early-stage esophageal squamous cell carcinoma: a preliminary study.

Esophageal squamous cell carcinoma (ESCC) is one of the most common histological types of esophageal cancers. It can seriously affect public health, particularly in Eastern Asia. Early diagnosis and effective therapy of ESCC can significantly help improve patient prognoses. The visualization of intrapapillary capillary loops (IPCLs) under magnification endoscopy (ME) can greatly support the identification of ESCC occurrences by endoscopists. This paper proposes an artificial-intelligence-assisted endoscopic diagnosis approach using deep learning for localizing and identifying IPCLs to diagnose early-stage ESCC. An improved Faster region-based convolutional network (R-CNN) with a polarized self-attention (PSA)-HRNetV2p backbone was employed to automatically detect IPCLs in ME images. In our study, 2887 ME with blue laser imaging (ME-BLI) images of 246 patients and 493 ME with narrow-band imaging (ME-NBI) images of 81 patients were collected from multiple hospitals and used to train and test our detection model. The ME-NBI images were used as the external testing set to verify the generalizability of the model. The experimental evaluation revealed that the proposed method achieved a recall of 79.25%, precision of 75.54%, F1-score of 0.764 and mean average precision (mAP) of 74.95%. Our method outperformed other existing approaches in our evaluation. It can effectively improve the accuracy of ESCC detection and provide a useful adjunct to the assessment of early-stage ESCC for endoscopists.