A5056078823- Lung Cancer Diagnosis and Treatment
- Radiomics and Machine Learning in Medical Imaging
- Topic Modeling
- Stock Market Forecasting Methods
- Recommender Systems and Techniques
- Text and Document Classification Technologies
- Advanced Text Analysis Techniques
- Sentiment Analysis and Opinion Mining
- Advanced X-ray and CT Imaging
- Medical Imaging Techniques and Applications
- Natural Language Processing Techniques
- EEG and Brain-Computer Interfaces
- Neural dynamics and brain function
- COVID-19 diagnosis using AI
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Blind Source Separation Techniques
- Financial Markets and Investment Strategies
- Forecasting Techniques and Applications
- Digital Marketing and Social Media
- Human Mobility and Location-Based Analysis
- Complex Network Analysis Techniques
- Domain Adaptation and Few-Shot Learning
- Time Series Analysis and Forecasting
- Authorship Attribution and Profiling
- Web Data Mining and Analysis
Shanghai Changzheng Hospital
2021-2025
Shanghai University of Finance and Economics
2017-2024
Second Military Medical University
2018-2024
Wuhan No.1 Hospital
2023
Xuzhou Medical College
2023
Philips (China)
2023
Weifang Medical University
2023
Chizhou University
2023
Guigang City People's Hospital
2023
United States Department of the Navy
2023
We present an attention-based bidirectional LSTM approach to improve the target-dependent sentiment classification. Our method learns alignment between target entities and most distinguishing features. conduct extensive experiments on a real-life dataset. The experimental results show that our model achieves state-of-the-art results.
To compare sensitivity in the detection of lung nodules between deep learning (DL) model and radiologists using various patient population scanning parameters to assess whether radiologists' performance could be enhanced when DL for assistance.A total 12 754 thin-section chest CT scans from January 2012 June 2017 were retrospectively collected training, validation, testing. Pulmonary these categorized into four types: solid, subsolid, calcified, pleural. The testing dataset was divided three...
The study aims to investigate the value of intratumoral and peritumoral radiomics clinical-radiological features for predicting spread through air spaces (STAS) in patients with clinical stage IA non-small cell lung cancer (NSCLC). A total 336 NSCLC from our hospital were randomly divided into training cohort (n = 236) internal validation 100) at a ratio 7:3, 69 other two external hospitals collected as cohort. Univariate multivariate analyses used select construct model. GTV, PTV5, PTV10,...
Abstract Background Computed tomography (CT) plays a great role in characterizing and quantifying changes lung structure function of chronic obstructive pulmonary disease (COPD). This study aimed to explore the performance CT-based whole radiomic discriminating COPD patients non-COPD patients. Methods retrospective was performed on 2785 who underwent examination 5 hospitals were divided into group group. The features volume extracted. Least absolute shrinkage selection operator (LASSO)...
This work takes the lead to study aspect-level sentiment classification <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in domain adaptation scenario</i> . Given a document of any domains, model needs figure out sentiments with respect fine-grained aspects in documents. Two main challenges exist this problem. One is build robust modeling across domains; other mine domain-specific and make use lexicon. In paper, we propose novel approach...
Open-domain dialog generation, which is a crucial component of artificial intelligence, an essential and challenging problem. In this article, we present personalized system, leverages the advantages multitask learning reinforcement for dialogue generation (MRPDG). Specifically, MRPDG consists two subtasks: 1) author profiling module that recognizes user characteristics from input sentence (auxiliary task) 2) system generates informative, grammatical, coherent responses with algorithms...
In lung cancer, preoperative prediction of visceral pleural invasion (VPI) is helpful for choosing the best treatment plan and improving prognosis patients. This study aimed to investigate usefulness computed tomography (CT) features in predicting VPI clinical stage IA peripheral adenocarcinoma (LUAD) with contact. divided type contact between tumor pleura into indirect direct contacts. retrospectively analyzed patients LUAD three hospitals enrolled 485 The CT lesions were predict VPI,...
In the screening of pulmonary subsolid nodules (SSNs), it is crucial to compare quantitative parameters under consistent computed tomography (CT) acquisition conditions, including same degree lung inflation. When non-end-inspiratory chest CT scan performed due poor breath holding, there a risk inaccurate measurement and erroneous assessment nodule growth. This study aims investigate effect respiratory phase on three-dimensional (3D) SSNs, further explore impact change judgment SSNs growth...
Recently, researchers found a new type of attacks, called time synchronization attack (TS attack), in cyber-physical systems. Instead modifying the measurements from system, this only changes stamps measurements. Studies show that these attacks are realistic and practical. However, existing detection techniques, e.g. bad data (BDD) machine learning methods, may not be able to catch attacks. In paper, we develop "first difference aware" (FDML) classifier detect attack. The key concept behind...
To develop and validate a model for predicting chronic obstructive pulmonary disease (COPD) in patients with lung cancer based on computed tomography (CT) radiomic signatures clinical imaging features.We retrospectively enrolled 443 who underwent function test as the primary cohort. They were randomly assigned to training (n = 311) or validation 132) set 7:3 ratio. Additionally, an independent external cohort of 54 was evaluated. The nodule signature constructed using least absolute...
Objective To develop and validate the model for predicting benign malignant ground-glass nodules (GGNs) based on whole-lung baseline CT features deriving from deep learning radiomics. Methods This retrospective study included 385 GGNs 3 hospitals, confirmed by pathology. We used 239 Hospital 1 as training internal validation set; 115 31 2 external test sets 2, respectively. An additional 32 stable with more than five years of follow-up were set 3. evaluated clinical morphological at chest...
Investor social media, such as StockTwist, are gaining increasing popularity. These sites allow users to post their investing opinions and suggestions in the form of microblogs. Given growth posted data, a significant challenging research problem is how utilize personal wisdom different viewpoints these help investment. Previous work aggregates sentiments related stocks generates buy or hold recommendations for obtaining favorable votes while suggesting sell short actions with negative...
Recently, cross-domain learning has become one of the most important research directions in data mining and machine learning. In multi-domain learning, problem is that classification patterns distributions are different among domains, which leads to knowledge (e.g. hyperplane) can not be directly transferred from domain another. This paper proposes a framework combine class-separate objectives (maximize separability classes) domain-merge (minimize domains) achieve representation Three...
Preoperative prediction of visceral pleural invasion (VPI) is important because it enables thoracic surgeons to choose appropriate surgical plans. This study aimed develop and validate a multivariate logistic regression model incorporating the maximum standardized uptake value (SUVmax) valuable computed tomography (CT) signs for non-invasive VPI status in subpleural clinical stage IA lung adenocarcinoma patients before surgery.A total 140 with peripheral were recruited divided into training...
Topic modeling of textual corpora is an important and challenging problem. In most previous work, the “bag-of-words” assumption usually made which ignores ordering words. This simplifies computation, but it unrealistically loses information semantic words in context. this paper, we present a Gaussian Mixture Neural Model (GMNTM) incorporates both meaning sentences into topic modeling. Specifically, represent each as cluster multi-dimensional vectors embed corpus collection generated by...
Min Yang, Wenting Tu, Ziyu Lu, Wenpeng Yin, Kam-Pui Chow. Proceedings of the 2015 Conference North American Chapter Association for Computational Linguistics: Human Language Technologies. 2015.