A5090055283- Radiomics and Machine Learning in Medical Imaging
- Hepatocellular Carcinoma Treatment and Prognosis
- MRI in cancer diagnosis
- Advanced MRI Techniques and Applications
- Music and Audio Processing
- Advanced Neuroimaging Techniques and Applications
- COVID-19 Clinical Research Studies
- Liver Disease Diagnosis and Treatment
- Speech Recognition and Synthesis
- Speech and Audio Processing
- Music Technology and Sound Studies
- Long-Term Effects of COVID-19
- COVID-19 diagnosis using AI
- Functional Brain Connectivity Studies
- Liver Disease and Transplantation
- Renal cell carcinoma treatment
- Cardiovascular Function and Risk Factors
- Advanced X-ray and CT Imaging
- Speech and dialogue systems
- Cardiac Imaging and Diagnostics
- Natural Language Processing Techniques
- SARS-CoV-2 and COVID-19 Research
- COVID-19 and healthcare impacts
- Chronic Kidney Disease and Diabetes
- Respiratory Support and Mechanisms
Zhongda Hospital Southeast University
2016-2025
Southwest University
2020-2025
Nanjing University of Information Science and Technology
2024
Chinese University of Hong Kong, Shenzhen
2022-2024
Chinese Academy of Forestry
2022-2024
Soochow University
2024
Research Institute of Forestry
2022-2024
Jiangsu Normal University
2024
Second Affiliated Hospital of Soochow University
2024
Wuhan Institute of Technology
2022-2024
Background: The coronavirus disease 2019 (COVID-19) has become a global challenge since the December 2019. hospital stay is one of prognostic indicators, and its predicting model based on CT radiomics features important for assessing patients' clinical outcome. study aimed to develop test machine learning-based models in patients with COVID-19 pneumonia. Methods: This retrospective, multicenter enrolled laboratory-confirmed SARS-CoV-2 infection their initial images from 5 designated...
Background Prediction of microvascular invasion (MVI) may help determine treatment strategies for hepatocellular carcinoma (HCC). Purpose To develop a radiomics approach predicting MVI status based on preoperative multiphase CT images and to identify MVI-associated differentially expressed genes. Materials Methods Patients with pathologically proven HCC from May 2012 September 2020 were retrospectively included four medical centers. Radiomics features extracted tumors peritumor regions...
Abstract Objectives To develop and test machine learning-based CT radiomics models for predicting hospital stay in patients with pneumonia associated SARS-CoV-2 infection. Design Cross-sectional Setting Multicenter Participants A total of 52 laboratory-confirmed infection their initial images were enrolled from 5 designated hospitals Ankang, Lishui, Zhenjiang, Lanzhou, Linxia between January 23, 2020 February 8, 2020. As 20, remained or non-findings excluded. Therefore, 31 72 lesion segments...
Rationale: Chest computed tomography (CT) has been used for the coronavirus disease 2019 (COVID-19) monitoring. However, imaging risk factors poor clinical outcomes remain unclear. In this study, we aimed to assess characteristics and associated with adverse composite endpoints in patients COVID-19 pneumonia. Methods: This retrospective cohort study enrolled laboratory-confirmed from 24 designated hospitals Jiangsu province, China, between 10 January 18 February 2020. Clinical initial CT...
While recent large-scale text-to-speech (TTS) models have achieved significant progress, they still fall short in speech quality, similarity, and prosody. Considering intricately encompasses various attributes (e.g., content, prosody, timbre, acoustic details) that pose challenges for generation, a natural idea is to factorize into individual subspaces representing different generate them individually. Motivated by it, we propose NaturalSpeech 3, TTS system with novel factorized diffusion...
To determine the patterns of chest computed tomography (CT) evolution according to disease severity in a large coronavirus 2019 (COVID-19) cohort Jiangsu Province, China.This retrospective study was conducted from January 10, 2020, February 18, 2020. All patients diagnosed with COVID-19 Province were included, retrospectively. Quantitative CT measurements pulmonary opacities including volume, density, and location extracted by deep learning algorithm. Dynamic these investigated symptom onset...
Abstract Background Coronavirus Disease-2019 (COVID-19) pandemic has become a major health event that endangers people throughout China and the world. Understanding factors associated with COVID-19 disease severity could support early identification of patients high risk for progression, inform prevention control activities, potentially reduce mortality. This study aims to describe characteristics severe or critically ill presentation in Jiangsu province, China. Methods Multicentre...
The main challenge of single image super resolution (SISR) is the recovery high frequency details such as tiny textures. However, most state-of-the-art methods lack specific modules to identify areas, causing output be blurred. We propose an attention-based approach give a discrimination between texture areas and smooth areas. After positions are located, compensation carried out. This can incorporate with previously proposed SISR networks. By providing enhancement, better performance visual...
To determine the age-specific clinical presentations and incidence of adverse outcomes among patients with COVID-19 in Jiangsu, China.Retrospective, multicentre cohort study performed at 24 hospitals China.625 enrolled between 10 January 15 March 2020.Of 625 (median age, 46 years; 329 (52.6%) men), 37 (5.9%) were children (18 years or younger), 261 (41.8%) young adults (19-44 years), 248 (39.7%) middle-aged (45-64 years) 79 (12.6%) elderly (65 older). The hypertension, coronary heart...
Patients with HCC receiving TACE have various clinical outcomes. Several prognostic models been proposed to predict outcomes for patients hepatocellular carcinomas (HCC) undergoing transarterial chemoembolization (TACE), but establishing an accurate model remains necessary. We aimed develop a radiomics signature from pretreatment CT establish combined radiomics-clinic (CRC) survival these patients. compared this CRC the existing in predicting patient survival. This retrospective study...
Background Computed tomography (CT) and magnetic resonance imaging (MRI) are both capable of predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC). However, which modality is better unknown. Purpose To intraindividually compare CT MRI for MVI solitary HCC investigate the added value radiomics analyses. Study Type Retrospective. Subjects Included were 402 consecutive patients with (training set:validation set = 300:102). Field Strength/Sequence T2‐weighted,...
Polycyclic aromatic hydrocarbons (PAHs) are well known persistent organic pollutants that have carcinogenic, teratogenic, and mutagenic effects on humans animals. Arbuscular mycorrhizal fungi (AMF) can infest plant hosts form symbioses may help plants to enhance potential rhizosphere effects, thus contributing the rhizodegradation of PAH-contaminated soils. The present study aimed assess effectiveness AMF enhancing Salix viminalis-mediated phytoremediation PAH-polluted soil clarify enzymatic...
Background A better understanding of the association between liver MRI proton density fat fraction (PDFF) and diseases might support clinical implementation PDFF. Purpose To quantify genetically predicted causal effect PDFF on disease risk. Materials Methods This population-based prospective observational study used summary-level data mainly from UK Biobank FinnGen. Mendelian randomization analysis was conducted using inverse variance–weighted method to explore risk with Bonferroni...
With the proposal of "biological-psychological-social" model, clinical decision-makers and researchers have paid more attention to bidirectional interactive effects between psychological factors diseases. The brain-gut-microbiota axis, as an important pathway for communication brain gut, plays role in occurrence development inflammatory bowel disease. This article reviews mechanism by which disorders mediate disease affecting axis. Research progress on causing "comorbidities mind body"...
LI-RADS Treatment Response Algorithm (TRA) version 2024 (v2024) introduced separate algorithms for detecting hepatocellular carcinoma (HCC) viability after radiation and nonradiation locoregional therapies (LRT). The algorithm incorporated MRI-based ancillary features to optionally upgrade lesions from LR-TR Equivocal Viable.
We introduce AnyEnhance, a unified generative model for voice enhancement that processes both speech and singing voices. Based on masked model, AnyEnhance is capable of handling voices, supporting wide range tasks including denoising, dereverberation, declipping, super-resolution, target speaker extraction, all simultaneously without fine-tuning. introduces prompt-guidance mechanism in-context learning, which allows the to natively accept reference speaker's timbre. In this way, it could...
Amphion is an open-source toolkit for Audio, Music, and Speech Generation, designed to lower the entry barrier junior researchers engineers in these fields. It provides a versatile framework that supports variety of generation tasks models. In this report, we introduce v0.2, second major release developed 2024. This features 100K-hour multilingual dataset, robust data preparation pipeline, novel models such as text-to-speech, audio coding, voice conversion. Furthermore, report includes...
Recent advancements in speech generation have been driven by the large-scale training datasets. However, current models fall short of capturing spontaneity and variability inherent real-world human speech, due to their reliance on audiobook datasets limited formal read-aloud styles. To bridge this gap, we introduce Emilia-Pipe, an open-source preprocessing pipeline extract high-quality data from valuable yet underexplored in-the-wild that capture spontaneous contexts. By leveraging construct...
To develop and validate a predictive model for predicting six-month outcome by integrating pretreatment MRI features one-month treatment response after TACE. A total of 108 patients with 160 hCCs from single-arm, multicenter clinical trial (NCT03113955) were analyzed served as the training cohort. An external dataset (ChiCTR2100046020) consisting 63 99 test dataset. Radiomics was constructed based on selected MR images. Univariate multivariate logistic regression analysis radiological...
We introduce Metis, a foundation model for unified speech generation. Unlike previous task-specific or multi-task models, Metis follows pre-training and fine-tuning paradigm. It is pre-trained on large-scale unlabeled data using masked generative modeling then fine-tuned to adapt diverse generation tasks. Specifically, 1) utilizes two discrete representations: SSL tokens derived from self-supervised learning (SSL) features, acoustic directly quantized waveforms. 2) performs tokens, utilizing...