Wan-Ting Hsu

Wan-Ting Hsu

Auto-generated profile, based on publicly available information

Wan-Ting Hsu

Department of Epidemiology, Harvard University T H Chan School of Public Health, Boston, Massachusetts, USA

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
5 July 2021 in Journal of Medical Internet Research
Verified
Journal Article
Open access
JWJo-Hsuan Wu
Jo-Hsuan Wu
TLT. Y. Alvin Liu
T. Y. Alvin Liu
WHWan-Ting Hsu
Wan-Ting Hsu
JHJennifer Hui-Chun Ho
Jennifer Hui-Chun Ho
CLChien-Chang Lee
Chien-Chang Lee
Background: Diabetic retinopathy (DR), whose standard diagnosis is performed by human experts, has high prevalence and requires a more efficient screening method. Although machine learning (ML)–based automated DR diagnosis has gained attention due to recent approval of IDx-DR, performance of this tool has not been examined systematically, and the best ML technique for use in a real-world setting has not been discussed. Objective: The aim of this study was to systematically examine the overall diagnostic accuracy of ML in diagnosing DR of different categories based on color fundus photographs and to determine the state-of-the-art ML approach. Methods: Published studies in PubMed and EMBASE were searched from inception to June 2020. Studies were screened for relevant outcomes, publication types, and data sufficiency, and a total of 60 out of 2128 (2.82%) studies were retrieved after study selection. Extraction of data was performed by 2 authors according to PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), and the quality assessment was performed according to the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Meta-analysis of diagnostic accuracy was pooled using a bivariate random effects model. The main outcomes included diagnostic accuracy, sensitivity, and specificity of ML in diagnosing DR based on color fundus photographs, as well as the performances of different major types of ML algorithms. Results: The primary meta-analysis included 60 color fundus photograph studies (445,175 interpretations). Overall, ML demonstrated high accuracy in diagnosing DR of various categories, with a pooled area under the receiver operating characteristic (AUROC) ranging from 0.97 (95% CI 0.96-0.99) to 0.99 (95% CI 0.98-1.00). The performance of ML in detecting more-than-mild DR was robust (sensitivity 0.95; AUROC 0.97), and by subgroup analyses, we observed that robust performance of ML was not limited to benchmark data sets (sensitivity 0.92; AUROC 0.96) but could be generalized to images collected in clinical practice (sensitivity 0.97; AUROC 0.97). Neural network was the most widely used method, and the subgroup analysis revealed a pooled AUROC of 0.98 (95% CI 0.96-0.99) for studies that used neural networks to diagnose more-than-mild DR. Conclusions: This meta-analysis demonstrated high diagnostic accuracy of ML algorithms in detecting DR on color fundus photographs, suggesting that state-of-the-art, ML-based DR screening algorithms are likely ready for clinical applications. However, a significant portion of the earlier published studies had methodology flaws, such as the lack of external validation and presence of spectrum bias. The results of these studies should be interpreted with caution.

1 January 2022 in Annals of Internal Medicine
Verified
Journal Article
Closed access
CCCho-Han Chiang
Cho-Han Chiang
CCCho-Hung Chiang
Cho-Hung Chiang
JPJohn W Pickering
John W Pickering
KSKiril M Stoyanov
Kiril M Stoyanov
DCDerek P Chew
Derek P Chew
CLChien-Chang Lee
Chien-Chang Lee
High-sensitivity cardiac troponin strategies are commonly used to triage patients with suspected acute myocardial infarction. The 2020 European Society of Cardiology guidelines recommend using the ... Background: The 2020 European Society of Cardiology (ESC) guidelines recommend using the 0/1-hour and 0/2-hour algorithms over the 0/3-hour algorithm as the first and second choices of high-sensiti...

13 January 2020 in PLOS ONE
Verified
Journal Article
Open access
PDF file
ECEric H Chou
Eric H Chou
SMShaynna Mann
Shaynna Mann
THTzu-Chun Hsu
Tzu-Chun Hsu
WHWan-Ting Hsu
Wan-Ting Hsu
CLCarolyn Chia-Yu Liu
Carolyn Chia-Yu Liu
CLChien-Chang Lee
Chien-Chang Lee
To determine the trends of infection sites and outcome of sepsis using a national population-based database. Using the Nationwide Inpatient Sample database of the US, adult sepsis hospitalizations and infection sites were identified using a validated approach that selects admissions with explicit ICD-9-CM codes for sepsis and diagnosis/procedure codes for acute organ dysfunctions. The primary outcome was the trend of incidence and in-hospital mortality of specific infection sites in sepsis patients. The secondary outcome was the impact of specific infection sites on in-hospital mortality. During the 9-year period, we identified 7,860,687 admissions of adult sepsis. Genitourinary tract infection (36.7%), lower respiratory tract infection (36.6%), and systemic fungal infection (9.2%) were the leading three sites of infection in patients with sepsis. Intra-abdominal infection (30.7%), lower respiratory tract infection (27.7%), and biliary tract infection (25.5%) were associated with highest mortality rate. The incidences of all sites of infections were trending upward. Musculoskeletal infection (annual increase: 34.2%) and skin and skin structure infection (annual increase: 23.0%) had the steepest increase. Mortality from all sites of infection has decreased significantly (trend p<0.001). Skin and skin structure infection had the fastest declining rate (annual decrease: 5.5%) followed by primary bacteremia (annual decrease: 5.3%) and catheter related bloodstream infection (annual decrease: 4.8%). The anatomic site of infection does have a differential impact on the mortality of septic patients. Intra-abdominal infection, lower respiratory tract infection, and biliary tract infection are associated with higher mortality in septic patients.

1 February 2020 in Hypertension
Verified
Journal Article
Closed access
PDF file
WHWan-Ting Hsu
Wan-Ting Hsu
BGBrandon Patrick Galm
Brandon Patrick Galm
GSGregory Schrank
Gregory Schrank
THTzu-Chun Hsu
Tzu-Chun Hsu
SLShih-Hao Lee
Shih-Hao Lee
CLChien-Chang Lee
Chien-Chang Lee
Antagonists of the renin-angiotensin-aldosterone system (RAAS), including ACEIs (angiotensin-converting enzyme inhibitors) and ARBs (angiotensin II receptor blockers), may prevent organ failure. We, therefore, investigated whether specific RAAS inhibitors are associated with reduced mortality in patients with sepsis.We conducted a population-based retrospective cohort study using multivariable propensity score–based regression to control for differences among patients using different RAAS inhibitors. A multivariable-adjusted Cox proportional-hazards regression model was used to determine the association between RAAS inhibitors and sepsis outcomes. To directly compare ACEI users, ARB users, and nonusers, a 3-way propensity score matching approach was performed. Results were pooled with previous evidence via a random-effects meta-analysis. A total of 52 727 patients were hospitalized with sepsis, of whom 7642 were prescribed an ACEI and 4237 were prescribed an ARB. Using propensity score–matched analyses, prior ACEI use was associated with decreased 30-day mortality (hazard ratio, 0.84 [95% CI, 0.75–0.94]) and 90-day mortality (hazard ratio, 0.83 [95% CI, 0.75–0.92]) compared with nonuse. Prior ARB use was associated with an improved 90-day survival (hazard ratio, 0.88 [95% CI, 0.83–0.94]). These results persisted in sensitivity analyses focusing on patients without cancer and patients with hypertension. By contrast, no beneficial effect was found for antecedent β-blockers exposure (hazard ratio, 0.99 [95% CI, 0.94–1.05]). The pooled estimates obtained from the meta-analysis was 0.71 (95% CI, 0.58–0.87) for prior use of ACEI/ARB.The short-term mortality after sepsis was substantially lower among those who were already established on RAAS inhibitor treatment when sepsis occurred.

3 March 2016 in Thorax
Verified
Journal Article
Open access
PDF file
CLChih-Cheng Lai
Chih-Cheng Lai
MLMeng-Tse Gabriel Lee
Meng-Tse Gabriel Lee
SLShih-Hao Lee
Shih-Hao Lee
WHWan-Ting Hsu
Wan-Ting Hsu
SCShy-Shin Chang
Shy-Shin Chang
CLChien-Chang Lee
Chien-Chang Lee
Background Epidemiological data suggest that statins improve the clinical outcome of respiratory infections. We sought to examine whether statin therapy decreases the risk of active TB. Methods We conducted a nested case-control study on data obtained from a national health insurance claims database between 1999 and 2011. The use of statins was classified as current, recent, past or chronic use. Three conditional logistic regression models were used to estimate the incidence rate ratios (RRs). The first assessed the effect of statin use without further adjustment; the second adjusted (individually) for 75 potential confounders; and the third adjusted for the Disease Risk Score (DRS). Results A total of 8098 new TB cases and 809 800 control patients were examined. All four types of statin users showed a decreased risk of active TB. Chronic use (>90 days in a calendar year) of statins was associated with the lowest unadjusted risk of TB (RR 0.74; 95% CI 0.63 to 0.87). The protective effect of active TB remained after adjusting for individual confounders (RR 0.66; 95% CI 0.56 to 0.78) and after DRS adjustment (RR 0.62; 95% CI 0.53 to 0.72). The effect estimates obtained for chronic and current use of statins were very similar. We also found that the active TB protection increased with increasing length of statin prescription. Conclusions We found that statin therapy was associated with a decreased risk of active TB, and the length of statin therapy affected the TB protection. Given the observational nature of this study, the protective effect against active TB must be confirmed in future randomised trials.

1 August 2017 in Journal of the American Medical Directors Association
Verified
Journal Article
Open access
WHWan-Ting Hsu
Wan-Ting Hsu
AEAmin Esmaily-Fard
Amin Esmaily-Fard
CLChih-Cheng Lai
Chih-Cheng Lai
DZDarshan Zala
Darshan Zala
SLSie-Huei Lee
Sie-Huei Lee
CLChien-Chang Lee
Chien-Chang Lee
Studies investigating the association between antipsychotic use and the risk of cerebrovascular accident (CVA) showed inconsistent results. Conduct a systematic review and meta-analysis to evaluate whether use of antipsychotics is associated with increased risk of CVA. Major electronic databases were searched from 1970 to October 2016 for observational studies investigating the risk of CVA among users of antipsychotics. Pooled estimates of odds ratios (ORs) and 95% confidence intervals (CIs) were obtained by random effects meta-analysis. Of 1171 citations identified, 10 studies were considered eligible. Significant increase in risk of CVA was associated with first-generation antipsychotics (OR 1.49; 95% CI 1.24-1.77) but not with second-generation antipsychotics (OR 1.31; 95% CI 0.74-2.30). Use of any antipsychotics in patients with dementia was associated with a low risk of CVA (OR 1.17; 95% CI 1.08-1.26). The available evidence suggests use of with first-generation antipsychotics as opposed to second-generation antipsychotics significantly increased the risk of CVA.