Xiaoming Sheng

Dr. Xiaoming Sheng

orcid

Research Focus & Metrics

Academic Career Length

24Years (2003-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 August 2017 in JMIR Research Protocols
Verified
Journal Article
Open access
GLGang Luo
Gang Luo
BSBryan L Stone
Bryan L Stone
MJMichael D Johnson
Michael D Johnson
PTPeter Tarczy-Hornoch
Peter Tarczy-Hornoch
AWAdam B Wilcox
Adam B Wilcox
FNFlory Nkoy
Flory Nkoy
To improve health outcomes and cut health care costs, we often need to conduct prediction/classification using large clinical datasets (aka, clinical big data), for example, to identify high-risk patients for preventive interventions. Machine learning has been proposed as a key technology for doing this. Machine learning has won most data science competitions and could support many clinical activities, yet only 15% of hospitals use it for even limited purposes. Despite familiarity with data, health care researchers often lack machine learning expertise to directly use clinical big data, creating a hurdle in realizing value from their data. Health care researchers can work with data scientists with deep machine learning knowledge, but it takes time and effort for both parties to communicate effectively. Facing a shortage in the United States of data scientists and hiring competition from companies with deep pockets, health care systems have difficulty recruiting data scientists. Building and generalizing a machine learning model often requires hundreds to thousands of manual iterations by data scientists to select the following: (1) hyper-parameter values and complex algorithms that greatly affect model accuracy and (2) operators and periods for temporally aggregating clinical attributes (eg, whether a patient's weight kept rising in the past year). This process becomes infeasible with limited budgets. This study's goal is to enable health care researchers to directly use clinical big data, make machine learning feasible with limited budgets and data scientist resources, and realize value from data. This study will allow us to achieve the following: (1) finish developing the new software, Automated Machine Learning (Auto-ML), to automate model selection for machine learning with clinical big data and validate Auto-ML on seven benchmark modeling problems of clinical importance; (2) apply Auto-ML and novel methodology to two new modeling problems crucial for care management allocation and pilot one model with care managers; and (3) perform simulations to estimate the impact of adopting Auto-ML on US patient outcomes. We are currently writing Auto-ML's design document. We intend to finish our study by around the year 2022. Auto-ML will generalize to various clinical prediction/classification problems. With minimal help from data scientists, health care researchers can use Auto-ML to quickly build high-quality models. This will boost wider use of machine learning in health care and improve patient outcomes.

1 July 2020 in Journal of Orthopaedic Translation
Verified
Journal Article
Open access
WTWei Tong
Wei Tong
XZXiaoguang Zhang
Xiaoguang Zhang
QZQuan Zhang
Quan Zhang
JFJiarui Fang
Jiarui Fang
YLYong Liu
Yong Liu
HTHongtao Tian
Hongtao Tian
Articular cartilage erosion probably plays a substantial role in osteoarthritis (OA) initiation and development. Studies demonstrated that umbilical cord-derived mesenchymal stem cells (UCMSCs) could delay chondrocytes apoptosis and ameliorate OA progression in patients, but the detailed mechanisms are largely uncharacterised. In this study, we aimed to study the effects of UCMSCs on monosodium iodoacetate (MIA)-induced rat OA model, and explore the cellular mechanism of this effect. Intra-articular injection of 0.3 ​mg MIA in 50 ​μL saline was performed on the left knee of the 200 ​g weight male Sprague-Dawley rat to induce rat knee OA. A single dose of 2.5 ​× ​105 undifferentiated UCMSCs one day after MIA or three-time intra-articular injection of 2.5 ​× ​105 UCMSCs on Days 1, 7 ​and 14 were given, respectively. Four weeks after MIA, joints were harvested and processed for paraffin sections. Safranine-O staining, haematoxylin and eosin staining ​and immunohistochemistry of MMP-13, ADAMTS-5, Col-2, CD68 ​and CD4 were performed to observe cartilage erosion and synovium. For in vitro ​studies, migration ability of cartilage superficial layer cells (SFCs) by UCMSCs were accessed by transwell assay. Furthermore, catabolism change of MIA-induced SFCs by UCMSCs was performed by real-rime polymerase chain reaction of Col-X and BCL-2 genes. CCK-8 assay was performed to check proliferation ability of SFCs by UCMSCs-conditioned media. In this study, we locally injected human UCMSCs, which is highly proliferative and noninvasively collectible, into MIA-induced rat knee OA. An important finding is on obviously ameliorated cartilage erosion and decreased OA Mankin score by repeated UCMSCs injection after MIA injection compared with single injection, both of which attenuated OA progression compared with vehicle. Interestingly, we observed significantly increased number of SFCs on the articular cartilage surface, probably related to elevated proliferation, mobilisation and inhibited catabolism marker: Col-X and BCL-2 gene expression of cultured SFCs by UCMSCs-conditioned media treatment in vitro. In addition to the change of unique SFCs, catabolism markers of ADAMTS-5 and MMP-13 were substantially upregulated in the whole cartilage layer chondrocytes as well. Strikingly, MIA-induced inflammatory cells infiltration, on both CD4+ Th cells and CD68+ macrophages, and hyperplasia of the synovium, which was alleviated by repeated UCMSCs injection. Our study demonstrated a critical role of repeated UCMSCs dosing on preserving SFCs function, cartilage structure and inhibiting synovitis during OA progression, and thus provided mechanistic proof of evidence for the use of UCMSCs on OA patients in the future. UCMSCs are a relatively "young" stem cell, and noninvasively collectible. In our study, we clearly demonstrated that it could effectively delay OA progression, possibly through reserving SFCs function and inhibiting synovitis. Therefore, it could be a new promising therapeutic cell source for OA after further clinical trials.

6 June 2012 in BMC Musculoskeletal Disorders
Verified
Journal Article
Open access
PDF file
TTthe WISTAH Hand Study Team
the WISTAH Hand Study Team
AGArun Garg
Arun Garg
KHKurt T Hegmann
Kurt T Hegmann
JWJacqueline J Wertsch
Jacqueline J Wertsch
JKJay Kapellusch
Jay Kapellusch
RHRichard Holubkov
Richard Holubkov
Few prospective cohort studies of distal upper extremity musculoskeletal disorders have been performed. Past studies have provided somewhat conflicting evidence for occupational risk factors and have largely reported data without adjustments for many personal and psychosocial factors.

7 March 2013 in BMC Musculoskeletal Disorders
Verified
Journal Article
Open access
PDF file
AGArun Garg
Arun Garg
KHKurt T Hegmann
Kurt T Hegmann
JMJ Steven Moore
J Steven Moore
JKJay Kapellusch
Jay Kapellusch
Matthew S ThieseMatthew S Thiese
Matthew S Thiese
RHRichard Holubkov
Richard Holubkov
Few prospective cohort studies of workplace low back pain (LBP) with quantified job physical exposure have been performed. There are few prospective epidemiological studies for LBP occupational risk factors and reported data generally have few adjustments for many personal and psychosocial factors.

2 May 2018 in The American Journal of Clinical Nutrition
Verified
Journal Article
Closed access
PDF file
HJHuifeng Jin
Huifeng Jin
HCHaojie Cheng
Haojie Cheng
WCWei Chen
Wei Chen
XSXiaoming Sheng
Xiaoming Sheng
MLMark A Levy
Mark A Levy
JTJunqiang Tian
Junqiang Tian
The single nucleotide polymorphism of the gene 5,10-methylenetetrahydrofolate reductase (MTHFR) C677T (or rs1801133) is the most established genetic factor that increases plasma total homocysteine (tHcy) and consequently results in hyperhomocysteinemia. Yet, given the limited penetrance of this genetic variant, it is necessary to individually predict the risk of hyperhomocysteinemia for an rs1801133 carrier. We hypothesized that variability in this genetic risk is largely due to the presence of factors (covariates) that serve as effect modifiers, confounders, or both, such as folic acid (FA) intake, and aimed to assess this risk in the complex context of these covariates. We systematically extracted from published studies the data on tHcy, rs1801133, and any previously reported rs1801133 covariates. The resulting metadata set was first used to analyze the covariates’ modifying effect by meta-regression and other statistical means. Subsequently, we controlled for this modifying effect by genotype-stratifying tHcy data and analyzed the variability in the risk resulting from the confounding of covariates. The data set contains data on 36 rs1801133 covariates that were collected from 114,799 participants and 256 qualified studies, among which 6 covariates (sex, age, race, FA intake, smoking, and alcohol consumption) are the most frequently informed and therefore included for statistical analysis. The effect of rs1801133 on tHcy exhibits significant variability that can be attributed to effect modification as well as confounding by these covariates. Via statistical modeling, we predicted the covariate-dependent risk of tHcy elevation and hyperhomocysteinemia in a systematic manner. We showed an evidence-based approach that globally assesses the covariate-dependent effect of rs1801133 on tHcy. The results should assist clinicians in interpreting the rs1801133 data from genetic testing for their patients. Such information is also important for the public, who increasingly receive genetic data from commercial services without interpretation of its clinical relevance. This study was registered at Research Registry with the registration number reviewregistry328.

6 June 2019 in JMIR Research Protocols
Verified
Journal Article
Open access
GLGang Luo
Gang Luo
BSBryan L Stone
Bryan L Stone
CKCorinna Koebnick
Corinna Koebnick
SHShan He
Shan He
DADavid H Au
David H Au
FNFlory Nkoy
Flory Nkoy
Both chronic obstructive pulmonary disease (COPD) and asthma incur heavy health care burdens. To support tailored preventive care for these 2 diseases, predictive modeling is widely used to give warnings and to identify patients for care management. However, 3 gaps exist in current modeling methods owing to rarely factoring in temporal aspects showing trends and early health change: (1) existing models seldom use temporal features and often give late warnings, making care reactive. A health risk is often found at a relatively late stage of declining health, when the risk of a poor outcome is high and resolving the issue is difficult and costly. A typical model predicts patient outcomes in the next 12 months. This often does not warn early enough. If a patient will actually be hospitalized for COPD next week, intervening now could be too late to avoid the hospitalization. If temporal features were used, this patient could potentially be identified a few weeks earlier to institute preventive therapy; (2) existing models often miss many temporal features with high predictive power and have low accuracy. This makes care management enroll many patients not needing it and overlook over half of the patients needing it the most; (3) existing models often give no information on why a patient is at high risk nor about possible interventions to mitigate risk, causing busy care managers to spend more time reviewing charts and to miss suited interventions. Typical automatic explanation methods cannot handle longitudinal attributes and fully address these issues. To fill these gaps so that more COPD and asthma patients will receive more appropriate and timely care, we will develop comprehensible data-driven methods to provide accurate early warnings of poor outcomes and to suggest tailored interventions, making care more proactive, efficient, and effective. By conducting a secondary data analysis and surveys, the study will: (1) use temporal features to provide accurate early warnings of poor outcomes and assess the potential impact on prediction accuracy, risk warning timeliness, and outcomes; (2) automatically identify actionable temporal risk factors for each patient at high risk for future hospital use and assess the impact on prediction accuracy and outcomes; and (3) assess the impact of actionable information on clinicians' acceptance of early warnings and on perceived care plan quality. We are obtaining clinical and administrative datasets from 3 leading health care systems' enterprise data warehouses. We plan to start data analysis in 2020 and finish our study in 2025. Techniques to be developed in this study can boost risk warning timeliness, model accuracy, and generalizability; improve patient finding for preventive care; help form tailored care plans; advance machine learning for many clinical applications; and be generalized for many other chronic diseases. PRR1-10.2196/13783.