A5012225901- Radiomics and Machine Learning in Medical Imaging
- Optical Imaging and Spectroscopy Techniques
- Photoacoustic and Ultrasonic Imaging
- AI in cancer detection
- Non-Invasive Vital Sign Monitoring
- Heart Rate Variability and Autonomic Control
- Lung Cancer Diagnosis and Treatment
- Advanced X-ray and CT Imaging
- Advanced Neuroimaging Techniques and Applications
- Optical Coherence Tomography Applications
- Advanced Neural Network Applications
- Digital Imaging for Blood Diseases
- ECG Monitoring and Analysis
- COVID-19 diagnosis using AI
- Functional Brain Connectivity Studies
- 3D IC and TSV technologies
- Medical Imaging and Analysis
- Insect-Plant Interactions and Control
- Genomics and Phylogenetic Studies
- Advanced MRI Techniques and Applications
- Adversarial Robustness in Machine Learning
- Electronic Packaging and Soldering Technologies
- Transcranial Magnetic Stimulation Studies
- Lung Cancer Treatments and Mutations
- Cardiac electrophysiology and arrhythmias
Beihang University
2004-2025
China Agricultural University
2022-2025
Shaanxi Normal University
2025
Beijing Advanced Sciences and Innovation Center
2021
Sun Yat-sen University
2020
Sun Yat-sen University Cancer Center
2020
State Key Laboratory of Oncology in South China
2020
Guangdong University of Technology
2020
Fudan University
2012-2013
Huashan Hospital
2013
While the potential of patient-derived organoids (PDOs) to predict patients' responses anti-cancer treatments has been well recognized, lengthy time and low efficiency in establishing PDOs hamper implementation PDO-based drug sensitivity tests clinics. We first adapt a mechanical sample processing method generate lung cancer (LCOs) from surgically resected biopsy tumor tissues. The LCOs recapitulate histological genetic features parental tumors have expand indefinitely. By employing an...
Ambulatory blood pressure (BP) monitoring plays a critical role in the early prevention and diagnosis of cardiovascular diseases. However, cuff-based inflatable devices cannot be used for continuous BP monitoring, while pulse transit time or multi-parameter-based methods require more bioelectrodes to acquire electrocardiogram signals. Thus, estimating waveforms only based on photoplethysmography (PPG) signals has essential clinical values. Nevertheless, extracting useful features from raw...
The accurate segmentation of multiple types lesions from adjacent tissues in medical images is significant clinical practice. Convolutional neural networks (CNNs) based on the coarse-to-fine strategy have been widely used this field. However, multi-lesion remains to be challenging due uncertainty size, contrast, and high interclass similarity tissues. In addition, commonly adopted cascaded rather demanding terms hardware, which limits potential deployment. To address problems above, we...
Fluorescence molecular tomography (FMT) is a promising and high sensitivity imaging modality that can reconstruct the three-dimensional (3D) distribution of interior fluorescent sources. However, spatial resolution FMT has encountered an insurmountable bottleneck cannot be substantially improved, due to simplified forward model severely ill-posed inverse problem. In this work, 3D fusion dual-sampling convolutional neural network, namely UHR-DeepFMT, was proposed achieve ultra-high...
Histopathological examinations heavily rely on hematoxylin and eosin (HE) immunohistochemistry (IHC) staining. IHC staining can offer more accurate diagnostic details but it brings significant financial time costs. Furthermore, either re-staining HE-stained slides or using adjacent for may compromise the accuracy of pathological diagnosis due to information loss. To address these challenges, we develop PST-Diff, a method generating virtual images from HE based diffusion models, which allows...
Background and Purpose Previous studies have noted changes in resting-state functional connectivity during motor recovery following stroke. However, these always uncover various patterns of recovery. Moreover, subgroups stroke patients with different outcomes hand function rarely been studied. Materials Methods We selected 24 who had a subcortical the left pathway displayed only deficits. The were divided into two subgroups: completely paralyzed hands (CPH) (12 patients) partially (PPH)...
Cerebral neuroplasticity after stroke has been elucidated by functional neuroimaging.However, little is known concerning how topological properties of the cortical motor-related network evolved following subcortical stroke.In present study, we investigated 24 patients with only left motor pathway damaged and matched healthy controls.A consisting 20 brain regions remote from primary lesion was constructed using resting-state MRI datasets.We subsequently used graph theoretical approaches to...
Classifying the subtypes of non-small cell lung cancer (NSCLC) is essential for clinically adopting optimal treatment strategies and improving clinical outcomes, but histological are confirmed by invasive biopsy or post-operative examination at present. Based on multi-center data, this study aimed to analyze importance extracted CT radiomics features develop model with good generalization performance precisely distinguishing major NSCLC subtypes: adenocarcinoma (ADC) squamous carcinoma...
The control region (CR) of the mitochondrial genome (mitogenome) represents a major noncoding fragment with several special structural features that are thought to be responsible for initiation mitogenome transcription and replication. However, few studies have revealed evolutionary patterns CR in phylogenetic context. Here, we explain characteristics evolution Tortricidae, inferred from mitogenome-based phylogeny. first complete mitogenomes genera Meiligma Matsumuraeses were sequenced. Both...
Addressing heavy metal contamination in water bodies is a critical concern for environmental scientists. Traditional detection methods are often complex and costly. Recent advancements artificial intelligence (AI), particularly machine learning (ML) deep (DL), have shown significant potential analytical chemistry. However, these AI models require extensive spectral data, which traditional struggle to provide quickly. To overcome this challenge, we developed new digital imaging system rapidly...
The mitochondrial genomes of three species Elmidae were sequenced. sizes 16,309 bp (C. jaechi), 16,291 (G. longiusculus), and 15,480 (S. punctulata). Each genome includes 13 protein-coding genes (PCGs), 22 transfer RNA (tRNAs), two ribosomal (rRNAs), a control region (CR). All mitogenomes show AT bias. Except for trnS1, lacking the dihydrouridine (DHU) arm, all tRNA had typical cloverleaf structure. codon usage preferences showed high similarity. arrangement in was consistent among them but...
Background Advances in high-throughput sequencing technology have led to a rapid increase the number of sequenced mitochondrial genomes (mitogenomes), ensuring emergence phylogenomics, as powerful tool for understanding evolutionary history various animal groups. Methods In this study, we utilized assemble and annotate mitogenomes Letana rubescens (Stål) Isopsera denticulata Ebner. We described characteristics genes these two species. Utilizing 13 PCGs 2 rRNA genes, reconstructed...
Coccinellidae (ladybird beetles) comprises around 6900 described species with a worldwide distribution and exhibits broad trophic diversity. Complete mitochondrial genomes (mitogenomes) are valuable resources in many research fields, such as genomics, population genetics, molecular evolution, phylogenetics. Here we sequenced report the complete mitogenome of Calvia chinensis, Micraspis discolor, Harmonia eucharis, Oenopia kirbyi. By comparing 36 mitogenomes published GenBank, found that long...
Abstract Fluorescence molecular imaging (FMI), a promising in vivo non‐invasive optical technology, has high sensitivity, specificity, and spatio‐temporal resolution. Owing to its further advantages of low cost no radiation, FMI been extensively employed the diagnosis treatment tumors, vascular system, guided delivery drugs. Nonetheless, affected by strong photon absorption scattering, faces two major challenges: limited tissue penetration depth complicated tomography reconstruction. This...
Pancreatic cancer is one of the most malignant cancers with high mortality. The rapid on-site evaluation (ROSE) technique can significantly accelerate diagnostic workflow pancreatic by immediately analyzing fast-stained cytopathological images pathologists. However, broader expansion ROSE diagnosis has been hindered shortage experienced Deep learning great potential for automatic classification in diagnosis. But it challenging to model complicated local and global image features. traditional...
Abstract Mosquito transmit numbers of parasites and pathogens resulting in fatal diseases. Species identification is a prerequisite for effective mosquito control. Existing morphological molecular classification methods have evitable disadvantages. Here we introduced Deep learning techniques species identification. A balanced, high-definition dataset with 9900 original images covering 17 was constructed. After three rounds screening adjustment-testing (first round among 3 convolutional...
Bioluminescence tomography (BLT) is a promising pre-clinical imaging technique for wide variety of biomedical applications, which can non-invasively reveal functional activities inside living animal bodies through the detection visible or near-infrared light produced by bioluminescent reactions. Recently, reconstruction approaches based on deep learning have shown great potential in optical modalities. However, these reports only generate data with stationary patterns constant target number,...