Artificial intelligence (AI), particularly deep learning, has demonstrated remarkable performance in medical imaging across a variety of modalities, including X-ray, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound, positron emission tomography (PET), and pathological imaging. However, most existing state-of-the-art AI techniques are task-specific and focus on a limited range of imaging modalities. Compared to these task-specific models, emerging foundation models represent a significant milestone in AI development. These models can learn generalized representations of medical images and apply them to downstream tasks through zero-shot or few-shot fine-tuning. Foundation models have the potential to address the comprehensive and multifactorial challenges encountered in clinical practice. This article reviews the clinical applications of both task-specific and foundation models, highlighting their differences, complementarities, and clinical relevance. We also examine their future research directions and potential challenges. Unlike the replacement relationship seen between deep learning and traditional machine learning, task-specific and foundation models are complementary, despite inherent differences. While foundation models primarily focus on segmentation and classification, task-specific models are integrated into nearly all medical image analyses. However, with further advancements, foundation models could be applied to other clinical scenarios. In conclusion, all indications suggest that task-specific and foundation models, especially the latter, have the potential to drive breakthroughs in medical imaging, from image processing to clinical workflows.
Humans ; Artificial Intelligence ; Deep Learning ; Diagnostic Imaging/methods* ; Magnetic Resonance Imaging ; Tomography, X-Ray Computed ; Positron-Emission Tomography

MicroRNA(miRNA)serves as a novel regulatory factor in viral and bacterial co-infections,exhibiting dual regulatory roles during pathogen infections.On the one hand,it exerts protective effects by modulating viral replication cycles and host anti-infection immune responses.On the other hand,it can be exploited by pathogens to promote infection progression,including influencing viral replication,antiviral immune responses,and regulating bacterial adhesion and proliferation.Mixed viral and bacterial infections lead to high mortality rates.By binding to specific target gene mRNA,miRNA degrades or inhibits its translation,thereby regulating target gene expres-sion,which may ultimately serve as potential diagnostic and therapeutic targets.This paper reviews recent ad-vances in research on the role of miRNA in viral and bacterial infections,particularly its mechanisms in co-infec-tions,aiming to provide theoretical reference for further research on the infection mechanisms.

Gene regulation relies on the precise binding of transcription factors (TFs) at regulatory elements, but simultaneously detecting hundreds of TFs on chromatin is challenging. We developed cFOOT-seq, a cytosine deaminase-based TF footprinting assay, for high-resolution, quantitative genome-wide assessment of TF binding in both open and closed chromatin regions, even with small cell numbers. By utilizing the dsDNA deaminase SsdAtox, cFOOT-seq converts accessible cytosines to uracil while preserving genomic integrity, making it compatible with techniques like ATAC-seq for sensitive and cost-effective detection of TF occupancy at the single-molecule and single-cell level. Our approach enables the delineation of TF footprints, quantification of occupancy, and examination of chromatin influences on TF binding. Notably, cFOOT-seq, combined with FootTrack analysis, enables de novo prediction of TF binding sites and tracking of TF occupancy dynamics. We demonstrate its application in capturing cell type-specific TFs, analyzing TF dynamics during reprogramming, and revealing TF dependencies on chromatin remodelers. Overall, cFOOT-seq represents a robust approach for investigating the genome-wide dynamics of TF occupancy and elucidating the cis-regulatory architecture underlying gene regulation.
Transcription Factors/genetics* ; Humans ; Chromatin/genetics* ; Animals ; Binding Sites ; Mice ; DNA Footprinting/methods*

To explore the regulatory mechanism of drought stress on the synthesis of chlorogenic acid and luteoloside in Lonicera japonica, we designed five drought gradients (soil water contents of 30%, 24%, 17%, 14%, and 10%) and screened and verified the differentially expressed genes (DEGs) by RNA sequencing (RNA-seq) and reverse transcription quantitative polymerase chain reaction (RT-qPCR). Furthermore, we employed HPLC to systematically measure the content changes of chlorogenic acid and luteoloside. The results revealed that drought significantly reduced the accumulation of secondary metabolites, and severe drought led to more obvious reductions. Under extreme drought (soil water content of 10%), the content of chlorogenic acid and luteoloside decreased significantly to 25.73 mg/g and 11.33 mg/g (with the decrease rates of 37.85% and 9.58%, respectively). A total of 77 454 genes were identified via transcriptome analysis, among which the number of DEGs reached 1 128 under the extraordinary drought. Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) enrichment analyses revealed that the DEGs were mainly involved in flavonoid synthesis, secondary metabolite biosynthesis, plant hormone signal transduction and the plant-pathogen interaction pathways, and the expression of key genes regulating the synthesis of chlorogenic acid and luteoloside was significantly downregulated. RT-qPCR verified the accuracy of the RNA-seq data. This study revealed that drought stress reduced the content of chlorogenic acid and luteoloside, the main secondary metabolites, by inhibiting the expression of key genes in the secondary metabolism pathways. The findings provide candidate gene resources for molecular breeding of drought-tolerant Lonicera japonica.
Lonicera/physiology* ; Chlorogenic Acid/metabolism* ; Droughts ; Stress, Physiological ; Gene Expression Regulation, Plant ; Glucosides/metabolism* ; Luteolin

To promote the standardization and normalization of perioperative care for follicular unit extraction(FUE) hair transplantation, ensure treatment efficacy, and align with advancements in the specialty, the Nursing Branch of the Chinese Association of Plastic and Aesthetics organized a panel of domestic experts. By integrating evidence-based medicine with clinical practice experience, and following thorough discussions, these experts developed the Clinical Practice Expert Consensus on Perioperative Nursing Care for Follicular Unit Extraction (2025). This consensus provides a comprehensive overview of hair transplantation, covering preoperative preparation, surgical procedures, intraoperative coordination, postoperative management, and complication prevention and treatment. It serves as a key reference and guide for implementing standardized perioperative nursing care in FUE hair transplantation.

Compared to traditional machine learning algorithms,which often suffer from low feature extraction efficiency,insufficient nonlinear pattern recognition capabilities and slow training speeds,the broad learning system(BLS)enhances the learning ability and efficiency by horizontally expanding the network structure.BLS offers advantages such as a simple structure,fast training speed,and strong generalization capabilities.While BLS has demonstrated potential in various fields,but its application in identification of complex samples has not been fully explored.This research investigated the feasibility of using BLS algorithm for identification of complex samples based on physicochemical indicators.Using the iris,wine,and breast cancer datasets,the length and width of petals and sepals of iris flowers,the physicochemical properties of wine,and the nuclear characteristics of breast cancer cells were used as input variables to establish BLS models for identifying iris species,wine varieties,and benign versus malignant nucleus.The model performance was evaluated by confusion matrices,accuracy,and runtime.Compared with partial least squares-discriminant analysis(PLS-DA),soft independent modeling of class analogies(SIMCA),and artificial neural networks(ANN),the results indicated that BLS demonstrated significant advantages in computational efficiency and recognition accuracy.Thus,BLS was an efficient and reliable method for identification of complex samples.

Objective:To evaluate the clinical efficacy of the V-shaped bone tunnel technique at the base of the distal phalanx for treating tendinous mallet finger.Methods:A retrospective analysis was conducted on clinical data from patients with tendinous mallet finger treated in the Department of Hand and Foot Surgery at Linyi People’s Hospital between August 2022 and November 2023. Two oblique bone tunnels were created in a V-shaped configuration at the dorsal base of the distal phalanx, adjacent to the extensor tendon insertion, using a 0.8 mm Kirschner wire. A 4-0 double-needle monofilament tendon suture was passed through the tunnels to secure the ruptured extensor tendon to the base of the distal phalanx, followed by fixation of the distal interphalangeal (DIP) joint with a 1.0 mm Kirschner wire. The Kirschner wire was removed at 4 weeks postoperatively to initiate functional exercises. Regular follow-up was conducted to monitor wound healing and functional recovery of the DIP joint. At the final follow-up, the range of flexion and extension of the DIP joint was measured, and treatment outcomes were evaluated using Crawford’s mallet finger evaluation criteria, which classified results into four grades: excellent, good, fair, and poor.Results:Fifteen patients (16 fingers) were included, comprising 11 males and 4 females, with a mean age of 44.5 years (range: 17-65 years). The injured fingers included 2 index, 4 middle, 5 ring, and 5 little fingers, all presenting with DIP joint flexion deformity and limited active extension. Postoperative follow-up ranged from 6 to 28 months (mean: 17 months). All wounds healed primarily without complications such as infection or skin necrosis, and no cases of tendon re-rupture occurred. At the final follow-up, the measurement results of flexion and extension range of motion of the affected fingers at the DIP joint were as follows: the maximum flexion angle of all 16 fingers was 45°, among which 7 fingers had an extension angle of 0°, 8 fingers had limited extension ranging from 1° to 10°, and 1 finger had limited extension of 15°. Among the 16 fingers, 7 fingers were rated as excellent, 8 fingers as good and 1 finger as fair.Conclusion:The V-shaped bone tunnel technique for tendon-to-bone reattachment of the extensor tendon insertion is a simple and effective method for treating tendinous mallet finger. It provides satisfactory functional recovery, improves finger appearance, and is associated with minimal complications.

This study provides a comprehensive review of the development and achievements in the field of oral genetic disorders research in China. Since the 1950s, Chinese scholars have progressed from epidemiological surveys to the elucidation of molecular mechanisms and functional genomics, marking a transition from macro-level observations to micro-level analyses. The research focuses on three major categories: cleft lip and palate, dental developmental anomalies, and rare hereditary oral diseases. Several critical pathogenic genes have been identified, and their molecular mechanisms and phenotypic characteristics elucidated. This paper summarizes representative findings, analyzes current challenges, and outlines future research directions, aiming to provide theoretical support for advancing both basic research and clinical translation in the field of oral genetics in China.

The integration of traditional Chinese and Western medicine is a crucial contemporary medical model widely utilized today.Modern medicine is intricately linked to"data"and"calculation".The key to the high-quality development of integrated Chinese and Western medicine is to realise the in-depth integration of information technology and modern medicine,including tackling the three major scientific problems(algorithm establishment,queue creation and data quality),introducing laws and policies,fostering cross-innovation talents in integrated Chinese and Western medicine,and regulating standards.At the same time,adhering to the principle of integrity and innovation,and being firm in cultural self-confidence are important factors in promoting the healthy,rapid and sustainable development of the cause of integrative Chinese and Western medicine and traditional Chinese medicine.