About me:

Passionate about transforming healthcare, I aspire to lead innovations in patient-centric care and personalized treatments. With a background in Biomedical Engineering, Data Science, and Medicine from my PhD from Johns Hopkins University, my focus is on developing AI/ML algorithms for medical data. These algorithms, applied in areas like Cardiology, Critical Care, and Oncology, address the need for personalized treatments, moving beyond ‘one-size-fits-all’ approaches.

Since 2013, I’ve worked on diverse health projects, collaborating with teams of clinicians, scientists, engineers, and statisticians. My dedication stems from early experiences witnessing patients suffering and dying since my childhood. In addition to academic research, I’ve contributed to the healthcare industry at Apple, Medtronic, Perthera AI, and Teladoc Health. My expertise spans various medical data formats, from electronic health records, digital signals, claims, to genomics and medical imaging, utilizing ML and causal inference methodologies to drive impactful outcomes and help make healthcare easier for clinicians and patients. Please see my LinkedIn and Google Scholar for my most up-to-date professional and academic profiles.

Currently, I work as a Principal Machine Learning Research Scientist at Optum AI (formerly Optum Labs/UnitedHealth Group R&D), UnitedHealth Group.

Portfolio:

My Short CV - Last Update: Sep 2022

My one-page General Resume - Last Update: Sep 2022

I. DISCLOSABLE PROJECTS:

Dissertation Research: Novel Time-to-Event Machine Learning Approach for Integration of Longitudinal Data and Image Data

An Early Prediction Algorirthm For Response to Treatment in Kawasaki Patients

Better Prediction of Outcomes in Pediatric and Adult Cardiac Surgery

Prediction of Trajectory of Exercise Capacity and Functional Status in Patients with Structural Heart Disease

A Machine Learning-Based Prediction of Cardiac Arrest Outcome Using a Large Multi-Center Database - Manuscript Accepted for Anaesthesia Critical Care & Pain Medicine

Utilizing electronic health record and bed-side monitoring temporal data, our post-cardiac arrest neurologic outcome and mortality prediction machine learning model outperforms the baseline APACHE model by area under the receiver operating curve of 12% and 15% respectively, showing both the efficacy of integrating time series data and the generalizability of utilizing a diverse multicenter database. Feature analysis revealed previously unknown factors which were associated with post-CA recovery. Results attest to the effectiveness of ML models for post-CA predictive modeling and suggest that PTS recorded in very early phase after resuscitation encode short-term outcome probabilities.

Machine Learning Methods for Survival Analysis: Are They Good Enough? - Manuscript In Preparation

Recent advances in data acquisition, storage, and artificial intelligence have enabled the use of machine learning (ML) in processing large, high-dimensional data to support decision-making in medicine and many other fields. ML methods in general, and Deep Learning (DL) in particular, have shown various successes in classification and regression tasks, and several ML and DL methods have been successfully adapted for the task of survival analysis. Despite the claims from the original authors that these ML and DL methods can address limitations posed by traditional statistical survival methods, there have been limited studies that demonstrate superior performance gains by comprehensively comparing these ML methods against the traditional methods. In this invited talk, I will summarize current ML methods in survival analysis to date, talk about their limitations, and present some results of my benchmark study that compare ML survival methods to the traditional Cox model, using cardiovascular event in the young adults as a motivating example.

Effect of Arterial Catheters on Outcomes in the ICU: A Causal Inference Approach - Manuscript In Preparation

Arterial catheters (AC) placement is a common procedure in the intensive care unit (ICU) to invasively monitor blood pressure, get frequent blood samples, and monitor blood gases. However, the benefit of AC is questionable and controversal, with comparable reading from non-invasive blood pressure measurements and AC’s asscociated complications (e.g., infections, limb ischemia, and pseudoaneurysm). With that, there is a call for better studies to define their role and type of patient to best benefit from them. We aim to determine the correlation between AC placement and days alive and vasopressor free in adults admitted to the ICU on vasopressors. A causal inference approach using propensity score matching was employed to construct two exchangeable groups to estimate the average treatement effect of AC on the outcomes. The outcomes include (1) days alive and vasopressor free - primary outcome, (2) death, and (3) length-of-stay (secondary outcomes). Results suggest that AC is associated with longer duration of vassopressor use and higher mortality.

Periatrial Fat Quality Predicts Atrial Fibrillation Ablation Outcome

Automatic Abnormality Classification and Breast Cancer Detection in Mammograms (Breast X-ray Images)

Age Prediction Based on Statistical Shape Models from MRI brain images

Projects that I have provided mentorship and/or consultation:

An Early Warning System of Hypoxemia in the ICU
Better Diagnosing ADHD (Attention Deficit Hyperactivity Disorder) Using Pre-Diagnostic Questionnaire
Predicting Acute Kidney Injury After Elevated Risk Non-cardiac Surgery
Delirium Prediction and Subtype Identification Using a Large Multi-Center ICU Database
Risk Factors For Neurologic Injury and Mortality in Critically Ill Children Supported on ECMO
Predicting Respiratory Decompensation in Pediatric ICU
A Computational Model to Predict Neurological Recovery Following Moderate and Severe Traumatic Brain Injuries
Advanced Risk Stratification and Prediction of Venous Thromboembolism in Critically Ill Patients

II. SELECTED PUBLICATIONS:

Please see my Google Scholar page for more up-to-date publications.

III. HONORS AND AWARDS:

Star Research Achievement Award | Society of Critical Care Medicine 2021 Virtual Conference
The award recognizes excellence in critical care research

Star Research Achievement Award | Society of Critical Care Medicine 2020
The award recognizes excellence in critical care research

ACCM Research Day Award | Department of Anesthesiology and Critical Care Medicine at Johns Hopkins 2019 & 2020
Best poster presentation

Young Investigator Award | Resuscitation Science Symposium, American Heart Association 2019
The award recognizes excellent submissions in the field of cardiac and trauma resuscitation science that are conducted by investigators in their early careers

President’s Fellow | Trinity College, Dept. of Engineering 2017
One senior student is selected as the best student from each major to represent their program of study

Presentation Award | Annual Biomedical Research Conference, American Society for Microbiology 2016
The award recognizes students who gave the best presentations in each discipline

Junior Engineering Prize | Trinity College, Dept. of Engineering 2016
The award recognizes one rising senior engineering major who, voted by the Engineering faculty, has demonstrated outstanding academic achievement and shown evidence of professional development

Full Scholarship | Trinity College 2013-2017
All tuition, room, board, and health insurance are covered for 4 years of college

Research Grant from the Daniel and Janet Mordecai Foundation | Children’s Hospital Colorado 2014
The grant provides stipends for a summer research and travel expenses to present research

One of Ten Young Promising Faces | Vietnamese Fund for Young Talents and National Committee Youth of Vietnam 2014
The award honors 10 Vietnamese under 35 years-old who stood out in fields of study, scientific research, production, society, sport, arts, and national defense

IV. MISCELLANEA:

Outside of work, I enjoy dancing (breakdancing, tutting, and general hip hop), calisthenics, watching Youtube, and talking with friends.
I also love manga, soccer (Come On Chelsea!), and the NBA.