Celebrating Doctor Trusting’s contributions to CoLabo and his next steps in science
By Andrés Colubri —
In this post, the Colubri Lab as a whole would like to send their heartfelt congratulations to Dr. Trusting Inekwe, who successfully defended his Ph.D. dissertation in Computer Science at Worcester Polytechnic Institute (WPI) earlier this summer. But beyond that, I want to personally highlight Trusting’s very important role in the lab since joining in 2021, as well as his scientific development and production over the years, now that he’s moving on to the next stage in his career.
Trusting was in fact one of the first members of my lab; he reached out to me with an email on May 17 2021 that started with the following lines: “Dear Prof Colubri, I would like to enquire with you about the possibility of having internship or co-op in your lab.” In another part of the email, Trusting said that he was “passionate about the research and development of interactive systems and software tools that intersect science, art, and programming with a goal of improving the livelihood of people.” In the midst of the pandemic, we met for a coffee near the WPI campus a few weeks later, and he became part of CoLabo from that time onwards.
Given the lab’s work at the intersection of bioinformatics, epidemiological modeling, AI/ML, and data visualization, Trusting’s vision for blending data science with impactful healthcare applications fit perfectly within this mission from the start. And so, Trusting was able to quickly start working on several research projects, mentoring junior lab members, and eventually finding the direction for this thesis work, all the while becoming a much appreciated member of the team, as can be seen in the pictures below:
From the very beginning, he demonstrated not just technical ability but also the capacity to guide and inspire others, creating a collaborative and inclusive culture in the lab.
From developing mobile health tools to applying advanced AI/ML for studying impact of COVID-19
Trusting’s first project in the lab allowed him to explore his interest in combining software engineering and data science, by building mobile apps that could use health data from health frontline workers as input for machine learning models for disease diagnosis and prognosis. This work aimed at using clinical prediction models that we published with many collaborators over the years (this, this, this, this, this, this, and this), and creating a simple framework to make those models available into an easy-to-use app.
The project aimed at showing how the gap between advanced computational models and frontline healthcare workers could be bridged, potentially enabling faster decision-making in resource-limited environments. Trusting had the opportunity to gain experience in Android mobile development by working on an existing code base of an app for collection of clinical data (here), improving on it, while also mentoring talented high-school and college student interns in the lab: Isha Nagireddy (Massachusetts Academy of Math and Science at WPI), Shrestha Mishra (Coppell High School in Coppell, Texas), and Faith Shim (Brown University in Providence, Rhode Island).
Trusting’s high-school mentees went on to pursue advanced STEM opportunities, reflecting the positive influence of his guidance. (Isha Nagireddy, for example, got accepted into Cornell University this Fall, while receiving several recognitions before graduating from Mass Academy like the NCWIT 2024 AiC Award, but that would be a story for another article.)
As we worked throughout the project, we also discovered how recent Python technologies such as Streamlit made the development of complex native apps less critical for this kind of use. Thanks to the Streamlit framework, which was first suggested by Shrestha, we were able to create a flexible proof-of-concept that would take any model in TensorFlow Lite (now LiteRT) format, use it to generate a data input form to apply the model on data instance provided by the user, and then calculate the Shapley values to visualize the contribution of each feature in the data to the final model’s prediction. We made all of this work available in this documented repository for anyone who may be interested in it.
However, what it turned to be Trusting’s main graduate work and eventually a fascinating journey through machine learning and AI, came through of a project that we inherited from a past lab intern from Harvard University, Brian Wee, who conducted an initial computational analysis of electronic medical records of over 200,000 patients from UMass Memorial Hospital. The purpose of Brian’s analysis was to study the impact of the COVID-19 pandemic on the disruptions in care for individuals with underlying conditions like diabetes mellitus and cardiovascular disease.
Following this initial lead, and taking over the UMass Memorial dataset, under the main supervision of Professor Emmanuel Agu at WPI, Trusting took a deep dive into the use of various traditional machine learning models to explore causality in biomarker changes before and after the pandemic, then moved on to deep learning and approaches to optimize the hyperparameters in neural network models, and ended up at the cutting edge of AI, applying transformer models for biomarker prediction using time series data. This was truly an impressive trajectory of work that literally traversed through the entire history of machine learning from logistic regression to transformer models with a meaningful application in biomedical science.
This evolution not only showcased Trusting’s ability to master diverse modeling approaches but also positioned our lab at the forefront of biomedical AI research. All of this work led to three publications:
- Inekwe T, Mkandawire W, Wee B, Agu E, Colubri A. Biomarker Trajectory Prediction and Causal Analysis of the Impact of the Covid-19 Pandemic on CVD Patients using Machine Learning. 2024 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE), Wilmington, DE, USA, 2024, pp. 1–12. https://ieeexplore.ieee.org/document/10614425
- Inekwe T, Agu E, Mkandawire W. Colubri A. A Genetic Neural Architecture Search Framework to Predict Biomarker Status in Cardiovascular Disease Patients during Pandemics. 2025 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE), New York, NY, USA, 2025, pp. 45–56. https://ieeexplore.ieee.org/document/11121123
- Inekwe T, Agu E, Mkandawire W, Colubri A. A Multi-target Bayesian Transformer Framework for Predicting Cardiovascular Disease Biomarkers during Pandemics. arXiv. 2025 2509.01794v1. https://arxiv.org/abs/2509.01794
Trusting’s full thesis dissertation can be downloaded from the Digital WPI repository.
As part of this work, Trusting also collaborated with Tzu-Chun Chu, a PhD candidate in the Department of Epidemiology and Biostatistic at the University of Georgia, with whom the lab has collaborated before. Tzu-Chun used the UMass Memorial dataset for her own thesis project as well, applying a series of statistical approaches complementary to Trusting’s, including Interpretable Bayesian Networks and Interrupted Time Series Analysis, to understand the impact of COVID-19 on chronic diseases in general, and on cardiovascular manifestations more specifically. The work of Trusting and Tzu-Chun on this dataset demonstrates how rigorous application of statistical and machine learning techniques on clinical datasets can yield valuable insights with significant public health impact.
After four years of having Trusting in my lab and seeing him develop as an accomplished computer scientist specializing in AI models for biomedical informatics, while being a wonderful part of the group all these years, reminds me of what drives us to keep our efforts in university education and research, even during difficult times. In his own words, Trusting describes his experience in the Colubri lab as “an opportunity of a lifetime” and also “that little seed of a letter grew to bear many fruits.” 🙏
Congratulations again, Dr. Trusting, and all the best wishes from the Colubri lab on your next endeavors, starting with a postdoctoral researcher position at SUNY Polytechnic Institute in Utica, NY! 🎉
