Glucose
Blood, Sweat, and Sugar | Hidden Variables: The Insulin-Glucagon Project
Glucose and ketone levels are vital to the body’s metabolism. Better understanding and predicting these biomarkers could help improve treatments for disorders like diabetes and Alzheimer’s. This project aims to produce accurate forecasts of glucose and ketone levels using temporal ML models.
Student Team Members
- Ben Allen
- Ruiqi Liu
- Owen Silberg
- Emily Wu
- Ryker Dolese
- Eleazar Martin
Sponsor and Mentors
- Sponsor Mentor: Dr. Ray Simar
- D2K Fellow, PhD Mentor: Roy Philips
- Faculty Mentor: Dr. Lorenzo Luzi
Chronicle
Dust Busters | Filling in the Map: Spatial Interpolation of Small Particle Air Pollution in Houston from Multi-Source Air Quality Data
Particulate matter under 2.5 micrometers (PM2.5) presents great health and environmental risks to the Houston community. We develop a machine learning model to interpolate PM2.5 pollution across the Greater Houston Area using sensor data, weather, traffic, and known polluters. With this, we aim to identify at-risk areas and compare government and non-government sensors to evaluate how sensor distribution affects accuracy. Lastly, we identify key pollution sources and create an interactive map for Houston residents to assess local air quality risks.
Student Team Members
- Isabelle Adeyinka
- Connie Huang
- Matthew Ye
- Bartu Citci
- Reid Westervelt
- Andrew Kim
Sponsor and Mentors
- Sponsor Mentor: Houston Chronicle Team - Rebekah Ward, Alexandra Kanik, Matt Zdun
- D2K Fellow, PhD Mentor: Ali Azizpour
- Faculty Mentor: Dr. Lorenzo Luzi
Flood Forecasting
Flood Sight | Advancing Real-Time Flood Predictions for Situational Awareness
Harris County frequently experiences heavy rainfall and flooding, which disrupts transportation and creates dangerous conditions. Traditional hydraulic models have been shown to be computationally expensive and less adaptable. This project tests deep learning models, particularly LSTMs, for accurately predicting rainfall in urban areas.
Student Team Members
- Alex Zalles
- Anoushka Mahendra-Rajah
- Elaine Ye
- Gloria Song
- Nithya Ramcharan
- Sutter Armistead
Sponsor and Mentors
- Sponsor Mentor: Dr. James Doss-Gollin
- D2K Fellow, PhD Mentor: Yuchen Lu
- Faculty Mentor: Dr. Xinxie Lan
Smithsonian
Paly Pals | AI-Powered Palynology: Automating Large-Scale Fossil Pollen Detection Efforts for Climate Change Research
As rising CO₂ levels pose the risk of extreme future warming, accurate climate models are more critical than ever. Fossilized pollen records help validate these models by enabling reconstructions of past climates. However, these reconstructions rely on identifying thousands of pollen grains across numerous microscope slides – a process that is both time-consuming and labor-intensive. This project automates this process by developing deep-learning models for pollen detection, greatly accelerating data annotation and enabling more scalable paleoclimate reconstructions.
Student Team Members
- Yuhan Wu
- Andrew Ondara
- Audrey (AJ) Kim
- Aaeisha Baharun
- Isauro Sanchez
- Jonathan Lee
Sponsor and Mentors
- Sponsor Mentor: Ingrid Romero, Alexander White, Carlos Jaramillo
- D2K Fellow, PhD Mentor: Krish Kabra
- Faculty Mentor: Arko Barman
Bayou City Waterkeeper
Flow and Order | Sewage Overflow Inequities Associated with Rainfall and Demographics
The impact is to guide remediation of infrastructure challenges in Houston's sewer system. The data makes use of rainfall and geospatial demographic information together with sewage overflow events publicly available by consent decree.
Student Team Members
- Saatchi Sagoo-Jones
- Izzy Goodman
- Jiahui Wei
- Anika Bjerknes
- Katharine Britt
- Alex Kornblum
Sponsor and Mentors
- Sponsor Mentor: Guadalupe Fernandez
- D2K Fellow, PhD Mentor: Tyler Bagwell
- Faculty Mentor: Dr. Xinjie Lan
HFD Train Delay
Data Trainers | Derailed – Evaluating the Effects of Train-Related Delays on Houston Fire Department Operations
Fire and EMS response times in Houston are impacted by road blockages caused by cargo trains. This project aims to combine fire department and train data to develop a unified, dynamic impact metric that scores each dispatch location’s vulnerability to blockages and can update based on new data and policy changes.
Student Team Members
- Vincent Behnke
- Emma Gruben
- Slim Lim
- Peter Stern
- Sawyer Cremer
- Kyle Zhang
Sponsor and Mentors
- Sponsor Mentor: Leonard Chan
- D2K Fellow, PhD Mentor: Meredith Kruse
- Faculty Mentor: Dr. Xinjie Lan
OCFR
Outstanding Coders For Real (OCFR) | Researcher matching system
Rice OCFR is the Office of Corporate and Foundation Relations, and a part of their work involves connecting researchers within Rice to external opportunities. However, making the best matches possible often requires deep technical expertise. We aim to build a researcher matching system powered by a context-aware database of papers to improve the ease, effectiveness, and efficiency of OCFR’s current process.
Student Team Members
- Grace Wang
- Emma Li
- Janhvi Somaiya
- Joy Yu
- Soham Ambre
Sponsor and Mentors
- Sponsor Mentor: Leah Aschmann Idaly Swindle
- D2K Fellow, PhD Mentor: Joao Pedro Rodrigues Mattos
- Faculty Mentor: Arko Barman
LivaNova
BrainStormers | Understanding VNS Patient Travel Trends
The Brainstormers have partnered with LivaNova to better understand the drug-resistant epileptic patients’ journey from diagnosis to follow-up in hopes of optimizing the VNS procedure and follow-up care. Travel patterns, influenced by geography, disease severity, and socioeconomic factors, highlight disparities in access to specialized treatment. By predicting patient travel distances for VNS therapy, LivaNova can develop targeted investment strategies, identify underserved areas, and improve healthcare delivery.
Student Team Members
- Didi Zhou
- Meera Borle
- Sharazad Ali
- Kristen Nwafor
- Alex Kim
- Nathan Nguyen
Sponsor and Mentors
- Sponsor Mentor: Megan Christy, Cheyenne Ehman, Kayla Frisoli
- D2K Fellow, PhD Mentor: Nhi Le
- Faculty Mentor: Xinjie Lan
NASA Spacecraft Segmentation
Segment the Satellite | Segmentation of an Unknown Spacecraft for In-Space Inspection
This project aims to develop a segmentation algorithm to segment and isolate any spacecraft from its background in real time. This paves the way for safer, more cost-effective in-space inspections and servicing, propelling NASA’s future missions and the wider aerospace sector forward.
Student Team Members
- Sophie Chikhladze
- Hamza Shili
- Jeffery Huang
- Taeho Choe
- Aditi Balaji
Sponsor and Mentors
- Sponsor Mentor: James Berck
- D2K Fellow, PhD Mentor: Janmajay Singh
- Faculty Mentor: Arko Barman
HERC
District Match | Bridging the Data Gap: Empowering Texas School Districts with Analytics
Texas school districts face growing pressure to meet accountability standards, making data-driven decision-making more essential than ever. There is currently no standard metric to compare data between similar school districts. Our project provides a dashboard to help faculty find demographically similar school districts using a distance-based algorithm, enabling them to compare and improve upon educational outcomes.
Student Team Members
- Bianca Schutz
- Everett Adkins
- Manav Mathur
- Sachin Shurpalekar
- Trey McCray
Sponsor and Mentors
- Sponsor Mentor: Dr. Erin Baumgartner, emb10@rice.edu
- D2K Fellow, PhD Mentor: Konstantin Larin, kl83@rice.edu
- Faculty Mentor: Dr. Xinjie Lan, xl116@rice.edu
HGO
Analyze4HGO | Uncovering Donor Journeys: A Hidden Markov Model and Bayesian Framework for Predicting Conversion at Houston Grand Opera
This project utilizes Hidden Markov Models and Bayesian inference to predict which donors will convert to patron-level and to identify potential high-level donors for Houston Grand Opera. By analyzing donor engagement data, such as contribution history and event interactions, we aim to uncover patterns that inform targeted strategies for cultivating top-tier patrons and maximizing long-term support.
Student Team Members
- Krish Kumar
- Tyler Field
- Jonah Lubin
- Tolu Asupoto
- Lavender Juma
- Matthew Hong
Sponsor and Mentors
- Sponsor Mentor: Claire Padien-Havens and Grace Tsai
- D2K Fellow, PhD Mentor: Janet Fu
- Faculty Mentor: Chad Shaw
HFD Station Optimization
OptimiStation Team | Houston Fire Department Dispatch Optimization Considering Train Delays
This project simulates emergency dispatch decisions to optimize response times using two methods. One method uses straight-line distance, while the other incorporates live train data to avoid blocked crossings. The goal is to demonstrate how real-time train information can improve dispatch efficiency and reduce delays.
Student Team Members
- Bayzhan Mukatay
- Benjamin Wang
- Eva Moughan
- Franco Gomez
- Isabelle Ruble
- Sol Kim
- Zachary Katz
Sponsor and Mentors
- Sponsor Mentor: Leonard Chan, Michael Marino
- D2K Fellow, PhD Mentor: Kang An
- Faculty Mentor: Arko Barman
NASA Crater
Lunar Lens | NASA Crater Detector
This project aims to develop an automated system that detects and outlines lunar craters in spacecraft imagery, allowing spacecraft to determine their location without relying on Earth-based communication. This capability is essential for enabling safe, autonomous navigation during future missions between Earth and the Moon.
Student Team Members
- Juan Hevia
- Tian Le
- Anekha Sokhal
- Henry Tran
- Jeremy Xu
- Madeleine Harrell
Sponsor and Mentors
- Sponsor Mentor: Kyle Smith
- D2K Fellow, PhD Mentor: Ananya Muguli
- Faculty Mentor: Arko Barman
NASA Orbital Transfer
Houston Rockets | Machine Learning for Orbital Transfer Optimization
Inside’s NASA’s mission planning group, there’s a tool called Copernicus which enables NASA to simulate and optimize the orbits of their satellites. Copernicus is slow and often requires manual engineering input. We address this problem with a machine learning algorithm to output optimal values to be fine-tuned with Copernicus, saving time, energy, and compute.
Student Team Members
- Facundo Arredondo
- Warren Weissbluth
- Judy Zhu
- Kyle Tam
- Hongyi Li
- Josue Jacome
Sponsor and Mentors
- Sponsor Mentor: Maxon V. Widner IV , Anu Kamath
- D2K Fellow, PhD Mentor: David Mis, Muhammad Sameer
- Faculty Mentor: Dr. Chad Shaw