Explore the vast range of titles available.

Waardenburg syndrome (WS), also known as auditory-pigmentary syndrome, is the most common cause of syndromic hearing loss (HL), which accounts for approximately 2–5% of all patients with congenital hearing loss. WS is classified into four subtypes depending on the clinical phenotypes. Currently, pathogenic mutations of PAX3, MITF, SOX10, EDN3, EDNRB or SNAI2 are associated with different subtypes of WS. Although supportive techniques like hearing aids, cochlear implants, or other assistive listening devices can alleviate the HL symptom, there is no cure for WS to date. Recently major progress has been achieved in preclinical studies of genetic HL in animal models, including gene delivery and stem cell replacement therapies. This review focuses on the current understandings of pathogenic mechanisms and potential biological therapeutic approaches for HL in WS, providing strategies and directions for implementing WS biological therapies, as well as possible problems to be faced, in the future.

Atom-by-atom engineering of nanomaterials requires atomic-level knowledge of the size evolution mechanism of nanoparticles, which remains one of the greatest mysteries in nanochemistry. Here we reveal atomic-level dynamics of size evolution reaction of molecular-like nanoparticles, i.e., nanoclusters (NCs) by delicate mass spectrometry (MS) analyses. The model size-conversion reaction is [Au 23 (SR) 16 ] − → [Au 25 (SR) 18 ] − (SR = thiolate ligand). We demonstrate that such isoelectronic (valence electron count is 8 in both NCs) size-conversion occurs by a surface-motif-exchange-induced symmetry-breaking core structure transformation mechanism, surfacing as a definitive reaction of [Au 23 (SR) 16 ] −  + 2 [Au 2 (SR) 3 ] − → [Au 25 (SR) 18 ] −  + 2 [Au(SR) 2 ] − . The detailed tandem MS analyses further suggest the bond susceptibility hierarchies in feed and final Au NCs, shedding mechanistic light on cluster reaction dynamics at atomic level. The MS-based mechanistic approach developed in this study also opens a complementary avenue to X-ray crystallography to reveal size evolution kinetics and dynamics. How metal nanoclusters evolve in size is poorly understood, particularly at the atomic level. Here, the authors use mass spectrometry to study the size conversion dynamics between two isoelectronic gold nanoclusters with atomic resolution, revealing that the growth reaction proceeds through a distinct balanced equation.

Circular RNAs (circRNAs) are a class of single‐stranded closed RNAs that are produced by the back splicing of precursor mRNAs. The formation of circRNAs mainly involves intron‐pairing‐driven circularization, RNA‐binding protein (RBP)‐driven circularization, and lariat‐driven circularization. The vast majority of circRNAs are found in the cytoplasm, and some intron‐containing circRNAs are localized in the nucleus. CircRNAs have been found to function as microRNA (miRNA) sponges, interact with RBPs and translate proteins, and play an important regulatory role in the development and progression of cancer. CircRNAs exhibit tissue‐ and developmental stage–specific expression and are stable, with longer half‐lives than linear RNAs. CircRNAs have great potential as biomarkers for cancer diagnosis and prognosis, which is highlighted by their detectability in tissues, especially in fluid biopsy samples such as plasma, saliva, and urine. Here, we review the current studies on the properties and functions of circRNAs and their clinical application value. This study demonstrates that circular RNAs (circRNAs) have great potential as biomarkers for cancer diagnosis and prognosis, which is highlighted by their detectability in tissues and fluid biopsy samples. CircRNAs may act as biomarkers in cancers including glioma, bladder cancer, breast cancer, gastric cancer, hepatocellular carcinoma, colorectal cancer, lung cancer, kidney cancer, pancreatic cancer, and leukemia.

Three-color electrophoretic displays (EPDs) have the characteristics of colorful display, reflection display, low power consumption, and flexible display. However, due to the addition of red particles, response time of three-color EPDs is increased. In this paper, we proposed a new driving waveform based on high-frequency voltage optimization and electrophoresis theory, which was used to shorten the response time. The proposed driving waveform was composed of an activation stage, a new red driving stage, and a black or white driving stage. The response time of particles was effectively reduced by removing an erasing stage. In the design process, the velocity of particles in non-polar solvents was analyzed by Newton’s second law and Stokes law. Next, an optimal duration and an optimal frequency of the activation stage were obtained to reduce ghost images and improve particle activity. Then, an optimal voltage which can effectively drive red particles was tested to reduce the response time of red particles. Experimental results showed that compared with a traditional driving waveform, the proposed driving waveform had a better performance. Response times of black particles, white particles and red particles were shortened by 40%, 47.8% and 44.9%, respectively.

Background Kidney cancer originates from the urinary tubule epithelial system of the renal parenchyma, accounting for 20% of all urinary system tumors. Approximately 70% of cases are localized at diagnosis, and 30% are metastatic. Most localized kidney cancers can be cured by surgery, but most metastatic patients relapse after surgery and eventually die of kidney cancer. Therefore, accurately predicting patient survival and identifying high‐risk metastatic patients will effectively guide interventions and improve prognosis. Methods This study used the data of 12,394 kidney cancer patients from the surveillance, epidemiology, and end results database to construct a research cohort related to kidney cancer survival and metastasis. Eight machine learning models (including support vector machines, logistic regression, decision tree, random forest, XGBoost, AdaBoost, K‐nearest neighbors, and multilayer perceptron) were developed to predict the survival and metastasis of kidney cancer and six evaluation indicators (accuracy, precision, sensitivity, specificity, F1 score, and area under the receiver operating characteristic [AUROC]) were used to verify, evaluate, and optimize the models. Results Among the eight machine learning models, Logistic Regression has the highest AUROC in both prediction scenarios. For 3‐year survival prediction, the Logistic Regression model had an accuracy of 0.684, a sensitivity of 0.702, a specificity of 0.670, a precision of 0.686, an F1 score of 0.683, and an AUROC of 0.741. For tumor metastasis prediction, the Logistic Regression model had an accuracy of 0.800, a sensitivity of 0.540, a specificity of 0.830, a precision of 0.769, an F1 score of 0.772, and an AUROC of 0.804. Conclusion In this study, we selected appropriate variables from both statistical and clinical significance and developed and compared eight machine learning models for predicting 3‐year survival and metastasis of kidney cancer. The prediction results and evaluation results demonstrated that our model could provide decision support for early intervention for kidney cancer patients. We used the data of 12,394 kidney cancer patients in the SEER (surveillance, epidemiology, and final results) database to construct a research cohort, combine with statistical relevance and clinical experience to screen for factors related to kidney cancer survival and prognosis. Eight machine learning models (Support Vector Machines, Logistic Regression, Decision Trees, Random Forests, XGBoost, AdaBoost, K‐Nearest Neighbors, and Multilayer Perceptrons) were developed to predict kidney cancer survival and tumor metastasis, using six indicators (Accuracy, Precision, Sensitivity, Specificity, F1 Score and AUROC) to validate, evaluate and optimize models. This study can provide decision support for early intervention in renal cancer patients.

Background: Digital breast tomosynthesis (DBT) has demonstrated improved cancer detection and reduced false-positive and recall rates compared with digital mammography (DM) alone in trials of women aged over 50. However, its benefits for older women remain uncertain. We hypothesized that among women aged 67+, breast cancer detected via DBT are more likely to be invasive lobular carcinoma (ILC) and present at a smaller size at diagnosis, and this association persists among women aged 75+. Additionally, we hypothesized that DBT detects ILC at a smaller tumor size.Methods: We conducted a retrospective cohort study of women aged 67+ with screen-detected ER+/HER2- breast cancer from 2015-2019 using SEER-Medicare data. The exposure was screening modality, dichotomized into DBT versus DM for cancer detection. The primary outcome was tumor histology, dichotomized into invasive lobular carcinoma (ILC) vs. other histology, which included invasive ductal carcinoma (IDC) or mixed ductal lobular histology (MDLC). The secondary outcome was tumor size at diagnosis, categorized as an ordinal variable with bins defined as 0–10 mm, 11–20 mm, 21–30 mm, 31–40 mm, 41–50 mm, and 51 mm or greater. Multivariable logistic regression was used to assess the association between screening modality and tumor histology and ordinal logistic regression was used to examine the association between screening modality and tumor size. Models were adjusted for sociodemographic, health care utilization, and other covariates associated with the exposure and the outcomes in bivariate analyses. To better account for differences in women’s characteristics associated with screening modality at detection, inverse probability weighting (IPW) derived from the propensity score for receiving DBT vs DM was incorporated into all adjusted logistic regressions described above.Results: Among the 12,582 included women, nearly half of them (49.2%) received DBT at breast cancer detection. The majority were non-Hispanic White (82.4%) and not dual eligible for Medicare and Medicaid (89.5%), and over half were aged 67 to 75 (55.8%). Among the overall cohort, 15.1% of the women had ILC and 80.2% had tumors smaller than 20 mm at diagnosis. In multivariable analyses, DBT at breast cancer detection was associated with 24% higher odds of being diagnosed with ILC compared to DM (aOR: 1.25, 95% CI: 1.12–1.39), an association that persisted among women over 75 (aOR: 1.22, CI: 1.10–1.35). DBT was associated with 9% higher odds of being diagnosed with smaller tumors (aOR: 1.09, CI: 1.01–1.17) compared to DM, but this association was not observed in women aged 75+ (aOR: 1.01, CI: 0.91–1.13). No significant interactions were observed between screening modality and tumor histology on tumor size.Conclusions: In this cohort study of women with screen-detected ER+/HER2- breast cancer, DBT at breast cancer detection was associated increased odds of being diagnosed with ILC and a smaller tumor compared to DM. While DBT remained associated with an increased diagnosis of ILC among women aged 75 and older, it was not associated with the diagnosis of smaller tumors in this age group. Our findings support the use of DBT as a routine screening modality to enhance the detection of ILC. However, more evidence is needed to determine whether DBT facilitates earlier cancer detection compared to DM.

Objectives The identification of gene mutations enables more appropriate genetic counseling and proper medical management for EVA patients. The purpose of this study was to validate the accuracy and sensitivity of our method for comprehensive mutation detection in EVA, and summarize these data to explore a more accurate and convenient genetic diagnosis method. Methods A multiplex PCR sequencing panel was designed to capture the exons of three known EVA-associated genes ( SLC26A4, KCNJ10 , and FOXI1 ), and NGS was conducted in 17 Chinese families with EVA. Results A total of 16 SLC26A4 variants were found in 21 probands with bilateral EVA, including three novel variants (c.416G>A, c.823G>A and c.1027G>C), which were not reported in the dbSNP, gnomAD database, and ClinVar databases. One patient carried a FOXI1 variant (heterozygous, c.214C>A) and one patient carried a KCNJ10 variant (heterozygous, c.1054C>A), both of which were novel variants. Biallelic potential pathogenic variants were detected in 21/21patient samples, leading to a purported diagnostic rate of 100%. All results were verified by Sanger sequencing. Conclusion This result supplemented the mutation spectrum of EVA, and supports that combined multiple PCR-targeted enrichment, and NGS is a valuable molecular diagnostic tool for EVA, and is suitable for clinical application.

Pioneered by the foundational information theory by Claude Shannon and the visionary framework of machine intelligence by Alan Turing, the convergent evolution of information and communication technologies (IT/CT) has created an unbroken wave of connectivity and computation. This synergy has sparked a technological revolution, now reaching its peak with large artificial intelligence (AI) models that are reshaping industries and redefining human-machine collaboration. However, the realization of ubiquitous intelligence faces considerable challenges due to substantial resource consumption in large models and high communication bandwidth demands. To address these challenges, AI Flow has been introduced as a multidisciplinary framework that integrates cutting-edge IT and CT advancements, with a particular emphasis on the following three key points. First, device-edge-cloud framework serves as the foundation, which integrates end devices, edge servers, and cloud clusters to optimize scalability and efficiency for low-latency model inference. Second, we introduce the concept of familial models , which refers to a series of different-sized models with aligned hidden features, enabling effective collaboration and the flexibility to adapt to varying resource constraints and dynamic scenarios. Third, connectivity- and interaction-based intelligence emergence is a novel paradigm of AI Flow. By leveraging communication networks to enhance connectivity, the collaboration among AI models across heterogeneous nodes achieves emergent intelligence that surpasses the capability of any single model. The innovations of AI Flow provide enhanced intelligence, timely responsiveness, and ubiquitous accessibility to AI services, paving the way for the tighter fusion of AI techniques and communication systems. These advancements are crucial to numerous application scenarios, including but not limited to embodied AI, wearable devices, and smart cities.

Large models have achieved remarkable performance across a range of reasoning and understanding tasks. Prior work often utilizes model ensembles or multi-agent systems to collaboratively generate responses, effectively operating in a server-to-server paradigm. However, such approaches do not align well with practical deployment settings, where a limited number of server-side models are shared by many clients under modern internet architectures. In this paper, we introduce CoLM (Collaboration in Large-Models), a novel framework for collaborative reasoning that redefines cooperation among large models from a client-server perspective. Unlike traditional ensemble methods that rely on simultaneous inference from multiple models to produce a single output, CoLM allows the outputs of multiple models to be aggregated or shared, enabling each client model to independently refine and update its own generation based on these high-quality outputs. This design enables collaborative benefits by fully leveraging both client-side and shared server-side models. We further extend CoLM to vision-language models (VLMs), demonstrating its applicability beyond language tasks. Experimental results across multiple benchmarks show that CoLM consistently improves model performance on previously failed queries, highlighting the effectiveness of collaborative guidance in enhancing single-model capabilities.