Hewlett Packard Enterprise Data Science Institute

Hien Van Nguyen, an associate professor in the Department of Electrical and Computer Engineering, served as the lead for an NIH-funded research project. His paper, “Interpreting Radiologist’s Intention from Eye Movements in Chest X-ray Diagnosis,” was recognized as an Outstanding Paper in Content Theme at the 33rd Association for Computing Machinery (ACM) International Conference on Multimedia, a major conference.

ACM Multimedia is dedicated to advancing research and applications across diverse multimedia fields. The conference brings together leading minds from academia and industry to explore cutting-edge developments in areas such as virtual and augmented reality, AI, video, and social data.

Nguyen said he was surprised but honored when he was notified that the team had won.

“This conference is a major venue for multimedia and AI research, so it was very meaningful to see our work recognized there,” he said. “My first thoughts were gratitude for the entire team and our clinical collaborators, because this project sits at the intersection of radiology and AI. It really only works when everyone contributes their expertise. I also felt excited that the community sees value in moving beyond ‘AI that mimics experts’ and toward AI that helps us understand expert decision-making.”

The goal of the team’s research is to teach AI to understand what a radiologist is looking for moment by moment, based on eye movements while reading a chest X-ray.

“Radiologists don’t just ‘look around’ randomly,” Nguyen said. “Their eyes move in patterns that reflect expertise. Sometimes they follow a systematic checklist, sometimes they explore when something looks uncertain, and sometimes they do a quick overall scan and then zoom in on a suspected problem.”

Their method, RadGazeIntent, takes a sequence of gaze fixations along with the X-ray image and predicts which clinical finding the radiologist is likely focusing on at each moment, along with confidence scores.

“As an analogy, imagine watching someone searching for issues in a complex scan,” Nguyen said. “Their eyes reveal whether they’re doing a routine sweep, double-checking something suspicious, or focusing hard on one spot. We’re building tools that can interpret that ‘search strategy’ rather than only copying the final answer.”

Other team members include Ngan Le, an assistant professor in the Department of Computer Science and Computer Engineering at the University of Arkansas, and her Ph.D. student Trong-Thang Pham; Anh Nguyen, an associate professor at the University of Liverpool in England; Carol C. Wu, a professor in the Department of Thoracic Imaging in the Division of Diagnostic Imaging at the University of Texas MD Anderson Cancer Center; and Zhigang Deng, Moores Professor of Computer Science in the College of Natural Sciences and Mathematics.

Nguyen identified three additional research directions the team is now exploring:

• Broader validation and generalization. Their current benchmarks are built from public chest X-ray eye-tracking datasets, but they want to test and refine the approach across more readers, institutions, and clinical settings.
• Expanding beyond chest X-rays. Eye-movement behavior can differ across modalities such as CT or MRI, so extending the approach could reveal what generalizes well and what requires new modeling.
• Intention-aware assistance and training tools. The long-term goal is interactive systems that can collaborate with radiologists—such as educational feedback for trainees or tools that adapt in real time to what the reader is focusing on. This would support clinicians rather than trying to replace them.

Nguyen has published in Nature Scientific Reports about related research, and he is also part of AI projects about more effective organ-targeted drug delivery systems and lung cancer diagnostics. For more information about his lab and his research, visit his group’s website.

On Thursday, April 18, 2024, the University of Houston Hewlett Packard Enterprise Data Science Institute hosted an online lecture titled "Computational Genomics Reveal Insights into the Biology of Non-model Organisms." Richard Meisel from UH's Department of Biology and Biochemistry presented his research on applying genomic methods to non-traditional study organisms. Meisel highlighted how recent technological advancements have made genomic analysis more accessible across various biological fields.  

The lecture focused on projects from Meisel's lab, a key example being their work on the house fly, Musca domestica, which possesses a complex sex determination system. Meisel’s team used advanced sequencing technologies and computational methods to assemble a high-quality genome for the house fly. This improved genome assembly allowed them to investigate the genetic basis of sex determination in this species, revealing variations in the male-determining chromosomes across different populations. The research uncovered how environmental factors, particularly temperature, play a role in the distribution of different male-determining chromosomes.  

Another significant aspect of the talk was the demonstration of how computational genomics can bridge the gap between model and non-model organisms. Meisel’s team used data from the well-studied fruit fly, Drosophila melanogaster, to inform experiments on the house fly, leading to the identification of genes involved in mating behavior.  

Throughout the presentation, Meisel emphasized the importance of integrating various approaches, including computational genomics, organismal experiments and comparative studies between model and non-model organisms. This multifaceted strategy allows researchers to address fundamental questions in biology from both mechanistic and evolutionary perspectives. Meisel highlighted the technical challenges of genomic assembly, such as dealing with repetitive DNA sequences. Long-read sequencing technologies and novel computational approaches have helped overcome these obstacles.

Richard Meisel, Ph.D. completed his postdoctoral training in the Department of Molecular Biology and Genetics at Cornell University. The Meisel lab uses lab experiments, genomic data and molecular biology techniques to study a broad range of questions in evolutionary biology. The research addresses how environmental variation and sex differences influence genetic and phenotypic diversity within populations and between species. Meisel’s lecture is part of an ongoing series hosted by the UH Hewlett Packard Enterprise Data Science Institute, aimed at showcasing cutting-edge research that leverages high-performance computing resources for scientific discovery. 

In a strategic collaboration connecting academia and industry expertise, the Hewlett Packard Enterprise Data Science Institute partnered with University Career Services to host a day-long AI Industry Symposium and Career Mixer at the University of Houston. The event, held on March 1, brought together scholars and professionals from diverse industry sectors.

Claudia Neuhauser, director of the HPE DSI, opened the event with a brief introduction of the institute and set the tone for a day of rich discussions. After the welcome, sector experts gave presentations on AI in research, technological advancements, and industry practices, covering oil and gas, energy, supply chain management and other areas.  

The symposium concluded with a panel on career success in AI-dominated industries, featuring speakers from Hewlett Packard Enterprise, Shell, ConocoPhillips, Patterson-UTI, Chevron, Humana and Metegrity.  

The career mixer that followed gave young professionals the opportunity to interact with representatives from ten leading organizations in the field. Recruiters shared insights on career paths and industry demands, outlining potential future collaborations.

Nearly 100 students, alumni, and members of the UH community attended the event. “It shows that we are meeting the interests of our young researchers and providing them with real-world connections to fruitful career paths,” Neuhauser said.  

This event is expected to shape the trajectory of programming, as Neuhauser hopes it will serve as a model for further integration of research, teaching, and career development at the University.

On February 29, Ernesto Lima gave a hybrid Lunch and Learn presentation with the Society of HPC Professionals at the Texas Advanced Computing Center (TACC). Lima is a research associate at UT Austin’s Center for Computational Oncology.

Breast cancer affects millions worldwide. Around 15-20 percent of cases are HER2-positive breast cancers, which are tumors that promote aggressive, rapid cancer cell growth. Current HER2-positive treatments— the most notable being trastuzumab— have proven effective, often given with chemotherapies, such as doxorubicin, to increase response rates. In his study, Lima investigates one main question: how can we optimally combine these two treatments to maximize control and reduce tumor size?

To tackle this problem, Lima and his research team utilized data from a murine model of HER+ breast cancer to analyze changes in tumor volume under various trastuzumab and doxorubicin treatment protocols. Adopting a Bayesian framework, they developed and calibrated ten mathematical models that capture the experiment tumor dynamics and the direct effects of the drug-to-drug interactions. The model calibration was done through Python using a Markov chain Monte Carlo (MCMC) ensemble sampler.

After selecting the model with the highest Bayesian information criterion, they applied the optimal control theory to find two ideal treatment protocols that both deliver trastuzumab prior to doxorubicin. The first one used the same experimental doses for both drugs, predicting an additional 45% tumor burden reduction. However, to limit cardiotoxicity, Lima and his team reduced the total dose of doxorubicin to approximately 43% for the second protocol, which achieved equivalent tumor control as the experimental protocol.

Currently, Lima and his research team are performing the necessary experiments to confirm, or improve, the optimized treatment protocol.

Lima is a research associate at the Center for Computational Oncology at the Oden Institute for Computational Engineering and Sciences and a member of the Life Sciences group at the Texas Advanced Computing Center (TACC). His research focuses on the development of numerical methods, innovative tumor growth models, treatment optimization, and model selection and calibration.

On March 21, 2024, the University of Houston Hewlett Packard Enterprise Data Science Institute hosted an online research lecture titled "Accelerated Materials Design Driven by Machine Learning" featuring Mingjian Wen, an assistant professor of Chemical and Biomolecular Engineering at UH's Cullen College of Engineering.  

Wen’s lecture highlighted two main challenges in creating reliable artificial intelligence models for materials science: uncertainty quantification for neural network models and the incorporation of prior physical knowledge and constraints into such models.  

Wen's research employs graph neural networks to model the elastic properties of materials. From the predicted elastic tensor, one can obtain various mechanical properties, such as bulk modulus (resistance to volume change) and shear modulus (resistance to sliding deformation) across different crystal systems. This capability is particularly valuable because many materials exhibit anisotropy, behaving differently depending on the direction of applied force.

Wen stressed the importance of incorporating physical constraints and uncertainty quantification into these models. He explained how a dropout technique can be used to assess the reliability of predictions, especially when extrapolating beyond the training data. This approach allows researchers to identify when a model might be unreliable for certain materials or conditions.

The lecture highlighted practical applications of these AI-driven methods. Wen demonstrated how the models could accelerate materials innovation by rapidly screening thousands of potential candidates and identifying those with exceptional properties such as strength. Wen emphasized that the true power of these methods lies in combining traditional scientific knowledge with machine learning techniques. This synergy between human expertise and AI tools has the potential to revolutionize materials design and discovery across various industries, creating more effective and physically-justified models for materials science.

Mingjian Wen, Ph.D is a Presidential Frontier Faculty Fellow at the University of Houston. His current work focuses on developing data-driven approaches for computational understanding and design of energy materials and contributing to open cyberinfrastructure for materials modeling. Prior to UH, Wen worked as a postdoctoral scholar on the Materials Project at Lawrence Berkeley National Lab. Wen's lecture is part of an ongoing series hosted by the HPE Data Science Institute, aimed at showcasing cutting-edge research that leverages high-performance computing resources for scientific discovery. 

Machine learning techniques are revolutionizing nuclear data evaluation and reactor design. Vladimir Sobes, assistant professor in the Department of Nuclear Engineering at the University of Tennessee, Knoxville, has dedicated his career to exploring the application of modern artificial intelligence and machine learning algorithms to current problems in nuclear engineering. 

In the realm of nuclear data evaluation, Sobes underscored the significance of ML algorithms in automating tasks such as hyper-parameter tuning and learning complex functions, particularly in the realm of uncertainty quantification. Sobes illustrated the structured nature of nuclear data, stressing the importance of automated feature identification—a task traditionally performed manually by experts. ML/AI, he argued, offers the promise of enhancing efficiency, reproducibility, and uncertainty quantification in nuclear data evaluation, thereby enabling experts to devote their time to more critical tasks. 

Transitioning to the topic of reactor design, AI has the potential surpass human capabilities in autonomous optimization, although Sobes cautioned against the indiscriminate use of surrogate models.  

Sobes demonstrated the effectiveness of the Gaussian Process Learning Algorithm for design optimization using a scenario related to the development of a spacecraft antenna. NASA engineers asked an AI algorithm to imagine an antenna. Despite an initial skepticism of its unintuitive design, rigorous testing demonstrated superior performance compared to human-designed alternatives. In 2006, the NASA ST5 spacecraft antenna, the same model proposed by an evolutionary computer design, was sent to space. 

AI/ML technologies are reshaping the landscape of nuclear engineering, from streamlining nuclear data evaluation processes to pushing the boundaries of reactor design beyond what was previously thought possible. Sobes' insights highlighted the symbiotic relationship between human expertise and AI, emphasizing the potential for collaboration to drive innovation in nuclear science. 

The presentation concluded with a quote by Arthur C. Clarke, adapted to signify the transformative potential of advanced optimization methodologies: "Any sufficiently advanced optimization methodology is going to be indistinguishable from intelligence."  

Sobes envisions a future where AI supplements human expertise, enabling breakthroughs in nuclear science and technology. 

The HPE Data Science Institute has partnered with University Career Services for a day of discovery and opportunity at the inaugural Artificial Intelligence (AI) Industry Symposium and Career Mixer. The event is open to students, alumni, faculty and staff from UH as well as industry professionals in the Houston area.
 

The event, sponsored by Chevron, addresses the emergent need for practical discussion on AI across industries and on preparing students to join this evolving workforce. Attendees will have the chance to connect with industry professionals and gain insights on how to thrive in increasing AI-enhanced environments. Panelists will include representatives from Chevron, ConocoPhillips, Hewlett Packard Enterprise, Humana, Metegrity, Patterson-UTI, and Shell.
 

After the symposium, students will be able to meet face-to-face with companies at the career mixer.
 

Claudia Neuhauser, director of the HPE DSI, describes the event as “a necessary and timely collaboration” between the institute and University Career Services. “As AI becomes established and integrated into corporate environments, we are finding ways to bridge academic training to meet these professional opportunities.”
 

The AI Industry Symposium and Career Mixer will be held on Friday, March 1 in the Houston Room of UH Student Center South. Attendees may register for the full event or individual sessions here. Recruiting employers can register to engage with UH students at the career mixer here. This event is free for all attendees thanks to event sponsor Chevron.

For more information, please visit the event website.