Tingting Zhai

ORCID: 0000-0002-4660-2125
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Text and Document Classification Technologies
  • Acute Ischemic Stroke Management
  • Data Stream Mining Techniques
  • Spam and Phishing Detection
  • Gout, Hyperuricemia, Uric Acid
  • Venous Thromboembolism Diagnosis and Management
  • Cerebrovascular and Carotid Artery Diseases
  • Ideological and Political Education
  • Russian Literature and Bakhtin Studies
  • Anomaly Detection Techniques and Applications
  • Alcohol Consumption and Health Effects
  • Educational Reforms and Innovations
  • Machine Learning and Algorithms
  • Stroke Rehabilitation and Recovery
  • Intracranial Aneurysms: Treatment and Complications
  • Imbalanced Data Classification Techniques
  • Advanced Bandit Algorithms Research
  • Face and Expression Recognition
  • Intracerebral and Subarachnoid Hemorrhage Research
  • Web Data Mining and Analysis
  • Traumatic Brain Injury and Neurovascular Disturbances
  • Machine Learning and Data Classification
  • Digital Transformation in Industry
  • Educational Technology and Assessment
  • Chronic Kidney Disease and Diabetes

First Affiliated Hospital of Zhengzhou University
2024

Yangzhou University
2018-2024

First People's Hospital of Jingzhou
2018-2024

Novartis (United States)
2023

Nanjing University
2017-2019

Qingdao University
2016-2018

Affiliated Hospital of Qingdao University
2016-2018

Anhui Medical University
2015

Jinan University
2013

Fuzhou University
2013

In this paper, we investigate the problem of sparse online linear classification in changing environments. We first analyze tracking performance standard classifiers, which use gradient descent for minimizing regularized hinge loss. The derived shifting bounds highlight importance choosing appropriate step sizes presence concept drifts. Notably, show that a better adaptability to drifts can be achieved using constant rather than state-of-the-art decreasing sizes. Based on these observations,...

10.1109/tnnls.2018.2877433 article EN IEEE Transactions on Neural Networks and Learning Systems 2018-11-13

In the RECO study, we investigated impact of operator's choice stent retriever size on patients with internal carotid artery (ICA) occlusion.

10.1016/j.heliyon.2024.e28873 article EN cc-by-nc-nd Heliyon 2024-03-30

To determine whether items of the Chinese version Montreal Cognitive Assessment Basic (MoCA-BC) could discriminate among cognitively normal controls (NC), and those with mild cognitive impairment (MCI), Alzheimer's disease (AD), moderate-severe as well their sensitivity specificity.MCI (n = 456), AD 502) 102) patients were recruited from memory clinic, Huashan Hospital, Shanghai, China. NC 329) health checkup outpatients. Five MoCA-BC item scores collected in interviews.The orientation test...

10.1186/s12883-019-1513-1 article EN cc-by BMC Neurology 2019-11-04

Abstract Background: The clinical use of tirofiban remains controversial for patients with acute ischemic stroke (AIS), we aimed to conduct a meta- analysis cohort studies assess the efficacy and safety AIS patients. Methods: All apparently unconfounded randomized controlled trials (RCTs) case-controlled studies, or without blinding, in individuals will be included this review. We literature search 2 databases Pubmed Embase, using indexing terms related cerebral infarctions include articles...

10.1097/md.0000000000014673 article EN cc-by-nc Medicine 2019-02-28

Most existing multilabel classification methods are batch learning methods, which may suffer from expensive retraining costs when dealing with new incoming data. In order to overcome the drawbacks of learning, we develop a family online algorithms, can update model instantly and efficiently, make timely prediction data arrive. Our algorithms all take closed-form update, is obtained by solving constrained optimization problem in each round learning. Label correlation explicitly modeled our...

10.1109/tnnls.2022.3164906 article EN IEEE Transactions on Neural Networks and Learning Systems 2022-04-18

Abstract The RECO is a novel endovascular treatment (EVT) device that adjusts the distance between two mesh segments to axially hold thrombus. We organized this postmarket study assess safety and performance of in acute ischaemic stroke (AIS) patients with large vessel occlusion (LVO). This was single-arm prospective multicentre enrolled as first-line treated at 9 centres. primary outcome measures included functional independence 90 days (mRS 0–2), symptomatic intracranial haemorrhage...

10.1038/s41598-024-52207-z article EN cc-by Scientific Reports 2024-01-25

10.1504/ijiids.2024.10065440 article EN International Journal of Intelligent Information and Database Systems 2024-01-01

10.1504/ijcse.2024.141338 article EN International Journal of Computational Science and Engineering 2024-01-01

10.1504/ijiids.2024.141766 article EN International Journal of Intelligent Information and Database Systems 2024-01-01

To evaluate the effect of dl-3-N-butylphthalide (NBP) on new cerebral microbleeds (CMBs) in patients with acute ischemic stroke (AIS).We will prospectively enroll AIS admitted to center Jingjiang People's Hospital. Qualified participants be randomly assigned either NBP group (NBP injection) or control injection placebo) a ratio 1:1. Patients complete brain magnetic resonance imaging within 48 hours and 14 days after onset observe CMBs through susceptibility weighted imaging, whether use...

10.1097/md.0000000000021594 article EN cc-by-nc Medicine 2020-08-04

Background: There may be a delay in or poor outcome of endovascular treatment (EVT) among acute ischemic stroke (AIS) patients with large-vessel occlusion (LVO) during off-hours. By using prospective, nationwide registry, we compared the workflow intervals and radiological/clinical outcomes between LVO treated EVT presenting off- on-hours. Methods: We analyzed prospectively collected Endovascular Treatment Key Technique Emergency Work Flow Improvement Acute Ischemic Stroke (ANGEL-ACT) data....

10.3389/fneur.2021.771803 article EN cc-by Frontiers in Neurology 2021-12-21

Existing online multi-label classification works cannot well handle the label thresholding problem and lack regret analysis for their algorithms. This paper proposes a novel framework of adaptive algorithms classification, with aim to overcome drawbacks existing methods. The key feature our is that both scoring models are included as important components classifier incorporated into one optimization problem. Further, in order establish relationship between models, loss function derived,...

10.48550/arxiv.2112.02301 preprint EN other-oa arXiv (Cornell University) 2021-01-01

Online active learning can both effectively reduce the labeling cost and process large-scale or streaming data, thus it has become an important research area in machine learning. However, there are few online studies regarding multi-label classification tasks. And existing algorithms either ignore label correlation cardinality inconsistency, only be applied to a particular application field. In this paper, we propose novel algorithm, termed MSGDA overcome drawbacks of algorithms. Our hybrid...

10.1109/mlbdbi54094.2021.00094 article EN 2021 3rd International Conference on Machine Learning, Big Data and Business Intelligence (MLBDBI) 2021-12-01

Objective To evaluate the effects of renin-angiotensin system (RAS) blockades [angiotensin-converting enzyme inhibitors (ACEI) and angiotensin II type 1 receptor blockers (ARB)]on contrast-induced nephropathy (CIN) in patients undergoing angiography. Methods Pubmed, Embase, Cochrane library, Wanfang database CNKI were searched. The literature limited range was from their start year to July 2015. Randomized controlled trials (RCTs) non-randomized influencing CIN assessed. Two...

10.3760/cma.j.issn.1001-7097.2016.05.006 article EN Chin J Nephrol 2016-05-15
Coming Soon ...