Alexander Junxiang Chen

Mr. Alexander Junxiang Chen

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
29 September 2021 in Brain Sciences
Verified
Journal Article
Open access
PDF file
JCJay Chandra
Jay Chandra
SMSiva Muthupalaniappan
Siva Muthupalaniappan
ZSZisheng Shang
Zisheng Shang
RDRichard Deng
Richard Deng
RLRaymond Lin
Raymond Lin
DPDaniel Z. Press
Daniel Z. Press
Conventional means of Parkinson’s Disease (PD) screening rely on qualitative tests typically administered by trained neurologists. Tablet technologies that enable data collection during handwriting and drawing tasks may provide low-cost, portable, and instantaneous quantitative methods for high-throughput PD screening. However, past efforts to use data from tablet-based drawing processes to distinguish between PD and control populations have demonstrated only moderate classification ability. Focusing on digitized drawings of Archimedean spirals, the present study utilized data from the open-access ParkinsonHW dataset to improve existing PD drawing diagnostic pipelines. Random forest classifiers were constructed using previously documented features and highly-predictive, newly-proposed features that leverage the many unique mathematical characteristics of the Archimedean spiral. This approach yielded an AUC of 0.999 on the particular dataset we tested on, and more importantly identified interpretable features with good promise for generalization across diverse patient cohorts. It demonstrated the potency of mathematical relationships inherent to the drawing shape and the usefulness of sparse feature sets and simple models, which further enhance interpretability, in the face of limited sample size. The results of this study also inform suggestions for future drawing task design and data analytics (feature extraction, shape selection, task diversity, drawing templates, and data sharing).

11 October 2020
Verified
Conference Paper
Closed access
EAErfan Asaadi
Erfan Asaadi
SBSteven Beland
Steven Beland
ACAlexander Junxiang Chen
Alexander Junxiang Chen
EDEwen Denney
Ewen Denney
DMDragos Margineantu
Dragos Margineantu
HYHuafeng Yu
Huafeng Yu
Dynamic assurance cases (DACs) are a novel concept for the provision of assurance—both during development and, subsequently, continuously in operation—that can be usefully applied to machine learning (ML)-based autonomous systems. We describe the application of a DAC for dependability assurance of an aviation system that integrates ML-based perception to provide an autonomous taxiing capability. Specifically, we present how we: i) formulate and capture risk-based safety and performance objectives, ii) model architectural mechanisms for risk reduction, iii) record the rationale that justifies relying upon autonomy, itself underpinned by heterogeneous items of verification and validation evidence, and iv) develop and integrate a computable notion of confidence that enables a run-time risk assessment and, in turn, dynamic assurance. We also describe our evaluation efforts, currently based on a hardware-in-the-loop simulator surrogate of an airworthy flight platform.

18 August 2022 in medRxiv
Verified
Preprint
Open access
PDF file
JFJeremy Samuel Faust
Jeremy Samuel Faust
CDChengan Du
Chengan Du
BRBenjamin Renton
Benjamin Renton
CLChenxue Liang
Chenxue Liang
ACAlexander Junxiang Chen
Alexander Junxiang Chen
HKHarlan M. Krumholz
Harlan M. Krumholz
Introduction Excess mortality does not depend on labeling the cause of death and is an accurate representation of the pandemic population-level effects. A comprehensive evaluation of all-cause excess mortality in the United States during the first two years of the COVID-19 pandemic, stratified by age, sex, region, and race/ethnicity can provide insight into the extent and variation in harm. Methods With Centers for Disease Control and Prevention (CDC)/National Center for Health Statistics (NCHS) data from 2014-2022, we use seasonal autoregressive integrated moving averages (sARIMA) to estimate excess mortality during the pandemic, defined as the difference between the number of observed and expected deaths. We continuously correct monthly expected deaths to reflect the decreased population owing to cumulative pandemic-associated excess deaths recorded. We calculate excess mortality for the total US population, and by age, sex, US census division, and race/ethnicity. Results From March 1, 2020, through February 28, 2022, there were 1.17 million excess deaths in the United States. Overall, mortality was 20% higher than expected during the study period. Of the excess deaths, 799,477 (68%) were among residents aged 65 and older. The largest relative increase in all-cause mortality was 27% among adults ages 18-49 years. Males comprised most of the excess mortality (57%), but this predominance declined with age. A higher relative mortality occurred among non-Hispanic American Indian/Alaskan Native, non-Hispanic Black, non-Hispanic Native Hawaiian and Other Pacific Islander, Hispanic people. Excess mortality differed by region; the highest rates were in the South, including in the population ages ≥65 years. Excess mortality rose and fell contemporaneously with COVID-19 waves. Conclusion In the first two years of the pandemic, the US experienced 1.17 million excess deaths, with greater relative increases in all-cause mortality among men, in American Indian/Alaskan Native, Black and Hispanic people, and the South.

8 March 2023 in medRxiv
Verified
Preprint
Open access
PDF file
JFJeremy Samuel Faust
Jeremy Samuel Faust
BRBenjamin Renton
Benjamin Renton
CDChengan Du
Chengan Du
ACAlexander Junxiang Chen
Alexander Junxiang Chen
SLShu-Xia Li
Shu-Xia Li
HKHarlan M. Krumholz
Harlan M. Krumholz
Introduction The US continued to record all-cause excess mortality after the rollout of vaccines. We sought to quantify excess mortality by state and compare these rates to primary series vaccination completion levels. Methods Observational cohort, US and state-level data. Expected monthly deaths were modeled using pre-pandemic US and state-level data (2015-2020). Mortality data was accessed from CDC public reporting. Results We find that in a two-year period since the rollout of vaccines, the US recorded >874,000 excess deaths. Vaccination rates and excess mortality were most strongly correlated in first two periods before the Omicron variant. Conclusion The association between vaccination and lower excess mortality rates was strongest in 2021 and early 2022, prior to high population rates of infection-acquired immunity. The findings underscore the benefits of the rapid vaccination rollout campaign and the continued need to boost at-risk populations.

2 March 2024 in medRxiv
Verified
Preprint
Open access
PDF file
JFJeremy Samuel Faust
Jeremy Samuel Faust
BRBenjamin Renton
Benjamin Renton
TBTasce Bongiovanni
Tasce Bongiovanni
ACAlexander Junxiang Chen
Alexander Junxiang Chen
KSKaren Dorsey Sheares
Karen Dorsey Sheares
HKHarlan M. Krumholz
Harlan M. Krumholz
BACKGROUND The end of the public health emergency provides an opportunity to fully describe disparities during the Covid-19 pandemic. METHODS In this retrospective cohort analysis of US deaths during the Covid-19 public health emergency (March 2020-April 2023), all-cause excess mortality and years of potential life lost (YPLL) were calculated by race or ethnicity overall and by age groups (ages 1.38 million all-cause excess deaths and ~23 million corresponding YPLL occurred during the pandemic. Had the rate of excess mortality observed among the White population been observed among the total population, >252,300 (18%) fewer excess deaths, and >5,192,000 fewer (22%) YPLL would have occurred. The highest excess mortality rates were among the American Indian/Alaska Native (AI/AN, 822 per 100,000; ~405,700 YPLL) and the Black (549 per 100,000; ~4,289,200 YPLL) populations. The highest relative increase in mortality was observed in the AI/AN population (1.34; 95% CI 1.31-1.37), followed by Hispanic (1.31; 95% CI 1.27-1.34), Native Hawaiian or Other Pacific Islander (1.24; 95% CI 1.21-1.27), Asian (1.20; 95% CI 1.18-1.20), Black (1.20; 95% CI 1.18-1.22) and White (1.12; 95% CI 1.09-1.15) populations. Greater disparities occurred among children and adults 252,000 and 5.2 million fewer YPLL would have been observed had increases in mortality among the total population been similar to the White population.

8 March 2023 in medRxiv
Verified
Preprint
Open access
PDF file
BRBenjamin Renton
Benjamin Renton
CDChengan Du
Chengan Du
ACAlexander Junxiang Chen
Alexander Junxiang Chen
SLShu-Xia Li
Shu-Xia Li
ZLZhenqiu Lin
Zhenqiu Lin
JFJeremy Samuel Faust
Jeremy Samuel Faust
Introduction The US has continued to see excess mortality through the Delta and Omicron periods. We sought to quantify excess mortality on a state level and calculate potential deaths averted if all states matched the excess mortality rates of those with the 10 lowest excess mortality rates. Methods Observational cohort, US and state-level data. Expected monthly deaths were modeled using pre-pandemic US and state-level data (2015-2020). Mortality data was accessed from CDC public reporting. Results We find that during the Delta and Omicron waves, the US recorded over 596,000 excess deaths. 60% of the nation’s total excess mortality during these periods could have been averted if all states had excess mortality rates equal to those with the 10 lowest excess mortality rates. Conclusion With large differences in excess mortality across US states in this 15-month study period, we note that a significant portion of deaths could have been averted with higher vaccination rates, improved mitigation policies, and adherence to other behaviors.