In the D2K Capstone program, interdisciplinary teams of students (advanced undergrads, professional master's students, and Ph.D.) work on a semester-long real-world project sponsored by our D2K Affiliate Members.
Fall 2023 Projects
BCM AMD
Illuminating Eyes | Interpretable Machine Learning Pipeline to Identify Genomic Signatures in Age-Related Macular Degeneration.
Age-related Macular Degeneration (AMD) is a complex disease influenced by the multiple genetic variants, environmental stress, and advanced aging. Here we are developing an interpretable machine-learning pipeline to identify crucial genomic signatures in AMD.
Student Team Members
- Duy Ha
- Patrick Yee
- Qingxin Yuan
- Sujitha Ravichandran,
- Tian Xia,
- Wanying Xu
Sponsor and Mentors
- Sponsor: Dr. Rinki Ratnapriya
- D2K Fellow, PhD Mentor: Maryam Khalid
- Faculty Mentor: Arko Barman
Watch 1-minute video highlight →
BCM CSA
Team CSA | From Soda and Tea Consumption to Diagnosing Central Sleep Apnea?
Sleep apnea is a potentially serious sleep disorder in which breathing repeatedly stops and starts. Our project focuses on central sleep apnea (CSA) when the apnea occurs from lack of brain signals. Currently, the cost of detecting CSA is too expensive, and the threshold is too high. Our project aims to reduce the cost and find the most accurate threshold.
Student Team Members
- Minyu Chen
- Jing Hu
- Liuxiao Kang
- Zheran Li
- Huailin Tang
- Risto Trajanov
- Ben Zhao
Sponsor and Mentors
- Sponsor: Dr. Ritwick Agrawal
- D2K Fellow, PhD Mentor: Kai Malcolm
- Faculty Mentor: Arko Barman
Watch 1-minute video highlight →
BCM Growth Trajectories
BCM Growth Trajectories Team | Investigating the Impacts of Neighborhood-Level Factors on Children’s Growth Trajectories
While previous research has indicated patterns of seasonal trends in children’s growth trajectories, their causes are not yet clear. We investigate the relationships between children’s neighborhood-level environment factors (Child Opportunity Index and exposure to artificial light at night) and trends in their growth, using data from a longitudinal study of elementary-school students.
Student Team Members
- Gail Oudekerk
- Elijah Sales
- Caleb Huang
- Zac Andrews
- William Pan
- Jacob Coyle
Sponsor and Mentors
- Sponsor: Dr. Jennette Moreno
- D2K Fellow, PhD Mentor: Kevin McCoy
- Faculty Mentor: Dr. Arko Barman
Watch 1-minute video highlight →
BCM/TCH Cardiomyopathy
BCM/TCH Cardiomyopathy Team | The Use of Machine Learning to Diagnose Diastolic Dysfunction in Pediatric Patients
There is currently no established method for diagnosing pediatric diastolic dysfunction (DD), leading to undetected progression towards diastolic heart failure. To improve early diagnosis and intervention, our project aims to develop a machine learning model capable of classifying various types of pediatric DDs, using echocardiogram data collected from pediatric patients at the Texas Children's Hospital.
Student Team Members
- Anitesh Reddy
- Calvin Aberg
- Lexie Jin
- Nik Sun
- Vicky Liu
- Kaiyuan (Vincent) Wu
Sponsor and Mentors
- Sponsor: Minh Bao Nguyen, Sebastian Acosta
- D2K Fellow, PhD Mentor: Roy Phillips
- Faculty Mentor: Arko Barman
Watch 1-minute video highlight →
CMOR
The Recommenders | 2-WaP: A Graphical Algorithm for Recommender Systems
We propose an algorithm using a graphical approach to provide recommendations to a new user. Our algorithm seeks to provide recommendations that are relevant to users with niche preferences without sacrificing relevance to the average user.
Student Team Members
- Daniel Himes
- Joshua Schaffer
- Joshua Yaffee
Sponsor and Mentors
- Sponsor: N/A
- D2K Fellow, PhD Mentor: João Pedro Mattos
- Faculty Mentor: Dr. Xinjie Lan
Watch 1-minute video highlight →
HFD
HFD Team | Assessing Emergency Responses to Motor Vehicle Collisions in the City of Houston
For our project, we generate a motor vehicle collision risk metric and use machine learning algorithms to determine significant contributors. We then visualize our findings using a heatmap to help inform the Houston Fire Department’s efforts in reducing collisions.
Student Team Members
- Benson Chi
- Joshua Fang
- Jasmine Lee
- Jessica Liu
- Quang Nguyen
- Sam Welsh
- John Zhang
Sponsor and Mentors
- Sponsor: Leonard Chen (Houston Fire Department)
- D2K Fellow, PhD Mentor: Yihua Xu
- Faculty Mentor: Dr. Xinjie Lan
Watch 1-minute video highlight →
LivaNova
OwlNova | A Data-Driven Approach to Sales: Optimizing the LivaNova Opportunity Pipeline
LivaNova lacks a reliable method of forecasting the outcome of ongoing sales negotiations of their medical devices with hospitals, making forecasting quarterly revenues difficult. For our project, we train a predictive model to reliably anticipate the results of ongoing sales negotiations, enhancing the accuracy of revenue projections.
Student Team Members
- Vidya Bulusu
- Mason Dooley
- Noah Keogh
- Israel Sierra
- Quent Titre
- Emaad Ullah
- Julian Vargas
Sponsor and Mentors
- Sponsor: Andrew Briggs, Cheyenne Ehman, Kayla Frisoli
- D2K Fellow, PhD Mentor: Giovanni Aiello
- Faculty Mentor: Xinjie Lan
Watch 1-minute video highlight →
MIC
Cardiac Predictors | Predicting Cardiac Output using Non-Invasive Measurements
In this project, we have developed a model for computing cardiac output using non-invasive measurements. The motivation behind the project is that conventional methods for measuring cardiac output like thermodilution are highly invasive and cannot always be applied to all patients, especially in the Pediatric Cardiac Intensive Care Unit (PCICU).
Student Team Members
- Yiwei Lu
- Kaiwen Li
- Huzaifa Ali
- Mattia Saladini
- Matias Romero
Sponsor and Mentors
- Sponsor: Medical Informatics Corp. (Dr. Sebastian Tume, Dr. Sulimon Sattari)
- D2K Fellow, PhD Mentor: Anton Banta
- Faculty Mentor: Dr. Arko Barman
Watch 1-minute video highlight →
TCH Length of Stay
Team TCH Length of Stay | Texas Children’s Heart Center Length of Stay Predictive Model
Our project goal is to create a predictive model to accurately predict the length of stay (LOS) of patients at Texas Children’s Hospital Cardiac Intensive Care Unit (TCH CICU). This model can potentially help provide more accurate patient discharge planning and allow TCH to help more children in need.
Student Team Members
- Abdullah Zaher
- Yijun Zhou
- Harry Wang
- Bobby Yang
- Sai Phani Ram Popuri
- Laura Zhao
Sponsor and Mentors
- Sponsor: Dr. Di Miao, Christian Jenson
- D2K Fellow, PhD Mentor: Nhi Le
- Faculty Mentor: Dr. Xinjie Lan
Watch 1-minute video highlight →
Westlake
The Visionaries | Development of Machine Learning Algorithms for Handwritten Character Recognition
Due to outdated technology in some of their manufacturing sites, Westlake Chemical’s engineers spend countless hours manually transferring handwritten measurements to computers. We develop software capable of automatically converting these measurements handwritten records to its digital form using machine learning.
Student Team Members
- Yuhan Wei
- Yanlin Wu
- Xiaoyu Chen
- Ray Zhang
- Fan Qiao
- Ndidi Nwosu
Sponsor and Mentors
- Sponsor: Gregory Parkison, Michael Dessauer
- D2K Fellow, PhD Mentor: Yuxin Tang
- Faculty Mentor: Xinjie Lan, Arko Barman