THE second run of the Ateneo “Breakthroughs” public lecture series shifts from how network-driven processes define power dynamics to create a collective identity that shape communities, and how intelligent systems and data-driven technologies deepen human understanding and improve lives.
At the “Smarter Sight: Building Intelligent Visual Systems for Public Good” session, Dr. Patricia Angela Abu, associate professor and chairman of the Department of Information Systems and Computer Science at Ateneo de Manila University, said that artificial intelligence (AI) is often portrayed as a future replacement for human specialists. She insists the reality is far more grounded.
“AI works best as a second set of eyes,” she told the crowd after she began narrated a frustrating parking spot search in a now dismantled Makati shopping center.
At the Ateneo Laboratory for Intelligent Visual Environments (Alive), which she leads, researchers are building computer vision systems that assist clinicians by flagging areas in medical images that require closer review. These tools do not diagnose. Instead, they help doctors navigate the growing volume of imaging data in modern health care with greater consistency and speed.
“AI is better understood as a tool that supports image analysis and screening,” Abu said. “It helps surface patterns that might otherwise be difficult to detect consistently.”
One of the lab’s ongoing projects focuses on odontogenic sinusitis, an infection that originates in the teeth but affects the sinuses. Using the YOLO (You Only Look Once) architecture, a model designed for fast and efficient object detection, the team developed a system that scans dental images for signs of infection and highlights them for clinical review.
The choice of YOLO was deliberate. Its ability to perform real-time detection while remaining computationally lightweight makes it suitable for environments where resources are limited and conditions mirror many health care settings in the Philippines.
That constraint points to a broader challenge: moving AI from the laboratory into everyday clinical use.
Models that perform well in controlled research settings often struggle when exposed to real-world variability. Differences in imaging equipment, patient profiles, and data quality can all affect performance. For Abu and her team, building an accurate model is only the starting point; ensuring that it works reliably across diverse conditions is the more demanding task.
“Systems have to remain consistent across different datasets and environments,” she said. “That’s where much of the work really is.”
To address this, Alive collaborates with research groups and clinical partners in Taiwan, where integration between universities and hospitals allows technologies to be tested using real clinical workflows. These partnerships provide access to varied datasets while reinforcing the importance of data governance, patient privacy, and responsible use of medical information.
Beyond health care, the same approach to visual analysis is being applied to a very different environment: city streets.
Alive’s work in urban sensing treats the city as a measurable system. Using computer vision and machine learning, the lab generates data on vehicle flow, pedestrian movement, parking occupancy, and even road conditions. The goal is to understand how urban spaces are actually used — not how they are assumed to function.
“Urban mobility starts with observing what is happening on the ground,” Abu said. “We look at how people, vehicles, and infrastructure interact in real environments.”
One immediate application is parking. By detecting whether spaces are occupied or available in real time, these systems can inform drivers where slots are open, reducing the time spent circling congested areas. This does not solve the larger issue of too many vehicles and too little space, but it makes existing infrastructure more efficient.
It is a small but practical intervention — one that reflects a broader philosophy behind the lab’s work. Whether analyzing a medical scan or a city street, the objective is the same: generate reliable data that leads to better decisions.
The inspiration for this work began in a far more ordinary setting: the parking spot hunt, circling a crowded mall in Makati with her mother eventually ending up very late for a wedding reception. What began as a routine frustration evolved into a professional question: how can technology make urban systems easier to navigate?
Her path from engineering into computer science was shaped by that same curiosity — an interest in how complex systems, from traffic networks to medical imaging, can be made more efficient and more responsive to real-world conditions.
As a leader in a field still largely dominated by men, her trajectory also reflects the importance of mentorship and support.
“It has been a meaningful journey, especially having mentors who helped guide my next steps and encouraged me to think through ideas more deeply,” Abu said. “Their support — particularly those who pushed me forward at moments when I doubted myself — has been instrumental in shaping my work and in gradually building what has become Alive today.”
This perspective also shapes how computer science is taught at Ateneo. Through service-learning and partnerships with communities and institutions, students are encouraged to apply technical skills to real-world problems. The emphasis is not only on building systems, but on understanding the environments in which those systems operate.
Under Abu’s leadership, Alive has also become a platform for what she describes as “social good” AI — projects that connect technical innovation with public need. Whether developing tools to assist radiologists in resource-constrained settings or systems that make everyday urban challenges like parking more manageable, the work remains grounded in a simple principle: data should serve people.
