Duc-Khanh To

Duc-Khanh To

Ph.D in Statistics

Lecturer at Department of Probability and Statistics, Faculty of Mathematics and Computer Science, University of Science, Vietnam National University Ho Chi Minh city

CV

Education

  • 1st January 2014 - 31th December 2016. Ph.D in Statistical Sciences, Department of Statistical Sciences, University of Padova.
    Thesis title: Statistical evaluation of diagnostic tests under verification bias.
    Supervisor: Prof. Monica Chiogna
    Co-supervisor: Prof. Gianfranco Adimari
    Date of defence: 24th March, 2017
  • October 2008 – July 2012. B.S. in Mathematics and Computer Sciences, Faculty of Mathematics and Computer Science, University of Science, Vietnam National University - Ho Chi Minh city, Vietnam.
    Title of dissertation: The application of nonparametric regression model for the backward heat problems.
    Supervisor: Prof. Dang Duc Trong

Work experience

  • July 2022 - July 2023: Assistant Professor (RTD-A) - sector 13/D1 – STATISTICS
    • Department of Statistical Sciences
    • University of Padova, Italy
  • December 2020 - June 2022: Postdoc Research Fellows (Type B, Project of Excellence “Statistical methods and models for complex data”)
    • Project title: Advanced methods for the evaluation of diagnostic tests with complex data
    • Coordinator: Prof. Gianfranco Adimari
    • Department of Statistical Sciences
    • University of Padova, Italy
  • December 2018 - November 2020: Postdoc Research Fellows (Type B, Project of Excellence “Statistical methods and models for complex data”)
    • Project title: ROC analysis for complex data
    • Coordinator: Prof. Gianfranco Adimari
    • Department of Statistical Sciences
    • University of Padova, Italy
  • 2nd May 2018 - 1st August 2018: Contract of occasional self–employment for the activity: “Statistical methods for marker detection with a view to cancer stratification”
    • Supervisor: Prof. Monica Chiogna and Prof. Gianfranco Adimari
    • Department of Statistical Sciences
    • University of Padova, Italy
  • 1st May 2017 - 30th April 2018: Postdoc Research Fellows (Type A).
    • Project title: Likelihood approaches for multivariate meta-analysis in situations with possible problems of measurements errors and missing data
    • Supervisor: Prof. Annamaria Guolo
    • Department of Statistical Sciences
    • University of Padova, Italy
  • 2nd May 2018 - 1st August 2018: Contract of occasional self–employment for the activity: “Study of the possibility of combining several biomarkers, with the aim of obtaining new diagnostic tests with greater accuracy than that of individual components.”
    • Supervisor: Prof. Monica Chiogna and Prof. Gianfranco Adimari
    • Department of Statistical Sciences
    • University of Padova, Italy
  • 1st Jan 2014 - 31st Dec 2016: PhD student
    • Department of Statistical Sciences
    • University of Padova, Italy
    • Supervisor: Prof. Monica Chiogna and Prof. Gianfranco Adimari
  • 1st November 2012 – 31th December 2013: Teaching Assistant
    • Faculty of Mathematics and Computer Science
    • University of Science, Vietnam National University - Ho Chi Minh city, Vietnam

Research experience

I have more than ten years of research experience in Statistics. In 2013, after graduating, I worked with Prof. Dang Duc Trong to study the application of nonparametric regression techniques for solving the backward heat problems with statistical discrete data. In 2015 and 2018, I and Prof. Duc Trong Dang, together with other colleagues, published two papers on mathematical and statistical journals.

In 2014, I was selected as a Ph.D. student in Statistics at the Department of Statistical Sciences with the first rank in the ranking list of CARIPARO Ph.D. scholarship at the University of Padova, Italy. During three years of my Ph.D. program, I was working on the subject “Statistical evaluation of diagnostic tests under verification bias” which is an interested topic in Medical Statistics. In particular, I and my supervisors (Prof. Monica Chiogna and Prof. Gianfranco Adimari) have investigated several methods, both parametric and nonparametric, to correct for verification bias in the evaluation of diagnostic test accuracy when the disease status has three categories (e.g., non-diseased, intermediate, diseased). For illustration, we applied our proposed methods to evaluate the accuracy of ovarian cancer-related tumor marker CA125 (which is a highly glycosylated sialomucin that is expressed on epithelial cell surface, especially on ovarian cancer cells) and to evaluate the ability of cerebrospinal fluid tau protein to discriminate between cognitively normal and mild cognitive impairment or Alzheimer’s disease. In addition, I developed an R package (named bcROCsurface) and a web interface (ShinyApps) to support researchers in the evaluation of predictive ability of diagnostic tests. As a result, I was awarded the Prize for Research 2017 of the department.

From May 2017 to April 2018, I was a postdoctoral research fellow at the department for the project “Likelihood approaches for multivariate meta-analysis in situations with possible problems of measurements errors and missing data” under the advice of Prof. Annamaria Guolo. We developed a pseudo-likelihood approach for multivariate meta-analysis of test accuracy studies with multiple thresholds. This method is based on the ideas of composite likelihood approach and multivariate mixed-effect model. Our proposed method is applied to two real meta-analyses on the accuracy of diagnostic tests to detect significant proteinuria from pre-eclampsia of pregnancies and to identify type 2 diabetes mellitus.

From November 2018 to December 2020, I was a postdoctoral research fellow at the department for my proposed research project “ROC analysis for complex data”. The project has been funded within the Project of Excellence “Statistical methods and models for complex data” of the department, from 2018 to 2020. Within this project, I studied some improvement for estimation of the volume under the ROC surface under nonignorable missing data mechanism, and methods to estimate the ROC surface and the volume under the ROC surface with clustered data also in the presence of covariates.

From 2020 to 2022, I proposed a research project entitled “Advanced methods for the evaluation of diagnostic tests with complex data”. The project has been funded within the Project of Excellence “Statistical methods and models for complex data” of the department. Within this project, I focus on some of these non-standard or complex situations and, in particular, aims to tackle and solve some open questions in the context of ROC analysis with clustered data, covariate effects and optimal use of diagnostic test or biomarker. For clustered data, I and my colleagues (Prof. Monica Chiogna and Prof. Gianfranco Adimari) proposed some new estimation processes for finding the optimal thresholds of a diagnostic test (biomarker), underlying the linear mixed-effects models. As an application, we study the use of the Lysosomal Associated Membrane Protein Family Member 5 gene expression as a biomarker to distinguish among three types of glutamatergic neurons, namely Layer 2/3 Intratelencephalic (L2/3 IT), Layer 4 (L4) and Layer 5 Pyramidal Tract (L5 PT) neurons. In addition, I developed an R package, named ClusROC, to support researches in the evaluation of predictive ability of classifiers (diagnostic tests or biomarkers) with clustered data. With regards to problem of covariate effects to diagnostic test accuracy, we proposed a new method to adjust for covariate effects in the estimation of volume under a ROC surface (VUS). The method is based on the induced-regression methodology, and the estimation can be done by a semiparametric generalized estimating equations (GEE) or by local linear regression (LL). We applied this method to the study of the effect of age on the ability of three cerebrospinal fluid (CSF) biomarkers, namely Aβ1-42 (amyloid-β1-42), Tau (total tau) protein and pTau (phosphorylated tau) protein, to distinguish the stages of Alzheimer’s disease, i.e., cognitively normal (CN), mild cognitive impairment (MCI), and Alzheimer’s disease (AD).

I also collaborated with some colleagues, biomedical engineers (Department of Information Engineering, University of Padova), to study some topics within the project “Movement Biomechanics in Parkinson’s Disease”.

Currently, I am studying the application of empirical likelihood-based inference for ROC analysis.

Skills

  • ROC analysis
  • Missing data
  • R package
  • Linear mixed-effects model
  • Location-scale regression model
  • Biomarker evaluation

Publications

See publications page.

Talks

  • 24 August 2022: Invited speaker in Seminar on Applied Statistics.
    • Topic: Optimal thresholds selection methods for early detection of disease status.
    • Vietnam Institute for Advanced Study in Mathematics (VIASM), Hanoi, Vietnam

Presentations

Oral presentations

  • To, D. K., Adimari, G. and Chiogna, M. (2022). Covariate adjustment methods for the evaluation of biomarkers in multi-class setting. (oral). Invited session “ROC methods for the evaluation of biomarkers (EO619)”, 15th International Conference of the ERCIM WG on Computational and Methodological Statistics (CMStatistics 2022), London, 17th - 19th December, 2022.

  • To, D. K., Adimari, G. and Chiogna, M. (2022). Confidence regions for optimal sensitivity and specificity of a diagnostic test. 51th Scientific Meeting of the Italian Statistical Society - SIS 2022, Caserta, Italy, 22-24 June, 2022.

  • To, D. K., Adimari, G. and Chiogna, M. (2019). A mean score equation-based approach to correct for nonignorable verification bias in estimation of the volume under the ROC surface. 32nd European Meeting of Statisticians, Palermo, Italy, 22-26 July, 2019.

  • Duc Khanh To and Annamaria Guolo (2018). A pseudo-likelihood approach for multivariate meta-analysis of test accuracy studies with multiple thresholds. 23rd International Conference on Computational Statistics, Iasi, Romania, 18-31 August, 2018.

  • To Duc, K. (2017). ROC surface analysis in presence of verification bias. 10th International Conference of the ERCIM WG on Computational and Methodological Statistics and 11th International Conference on Computational and Financial Econometrics, London, 18th December, 2017.

  • To Duc, K., Chiogna, M. and Adimari, G. (2016). Nonparametric estimation of ROC surfaces in presence of verification bias. The 22nd International Conference on Computational Statistics, Oviedo, Spain, 23-26 August, 2016.

  • To Duc, K., Chiogna, M. and Adimari, G. (2016). Bias-corrected estimation methods for receiver operating characteristic surface. The 18th International Conference on Biometrics and Biostatistics, Amsterdam, Netherlands, 4-5 August, 2016.

  • Khanh T. D., Trong D. D., Tuan N. H., Minh N. D. (2013). Nonparametric regression in a statistical modified Helmholtz equation using Fourier spectral regularization. The 8th Congress Mathematical Vietnam, Nha Trang, Vietnam, 10-14 August, 2013.

  • Thang, P. L., Khanh T. D., Hoang, N. D. (2013). Some applications of the Plackett-Burman design for investigating optimal medium components with soya bean to grow Bacillus subtilis for protein production. Statistics and its Interactions with Other Disciplines, Ho Chi Minh city, Vietnam, June, 2013.

Poster presentations

  • Khanh T. D., Trong D. D., Tuan N. H., Minh N. D. (2013). A two-dimensional backward heat problems with random data. Statistics and its Interactions with Other Disciplines, Ho Chi Minh city, Vietnam, June, 2013.

Teaching

See teaching page.

Service

Reviewer

I serve as a peer-reviewer in my field. A subset of my peer reviews can be found on my Web of Science profile.

Awards and Scholarship

  • (2022) Achieved the National Scientific qualification as associate in the Italian higher education system, in the call 2021/2023 (Ministerial Decree n. 553/2021 and 589/2021) for the disciplinary field of 13/D1 - Statistics. (Academic Recruitment Field 13/D - Statistics and Mathematical Methods for Decisions, according to the National Classification). The validity of the qualification is eleven years, starting from the 01/02/2023 and will expire on the 01/02/2034.

  • (2018) Prize for Research 2017, Department of Statistical Sciences, University of Padova.

  • (2014 - 2016) CARIPARO Ph.D scholarship, Department of Statistical Sciences, University of Padova.

  • (2011) Awards for Scientific Research Student, rank 4th, University of Science, Vietnam National University - Ho Chi Minh city, Vietnam.

  • (2010 - 2012) Fellowship of honor program, University of Science, Vietnam National University - Ho Chi Minh city, Vietnam.