AS ARTIFICIAL intelligence becomes increasingly integrated into healthcare, Thomasian clinicians said the technology should enhance — not replace — physicians by improving diagnostic accuracy and personalized treatment while preserving the human judgment essential to patient care.
The UST Hospital Department of Internal Medicine and the Thomasian Internists’ Group of Educators and Researchers Inc. (Tiger Inc.) explored AI-integrated diagnostics, precision medicine and the ethical limits of emerging technologies during a two-day postgraduate course aimed at preparing clinicians for an increasingly data-driven healthcare landscape.
Internist and gastroenterologist Dr. Bryan Christopher Lao said predictions that AI would make physicians and medical schools obsolete by 2035 overlooked the more pressing challenge confronting clinicians: cognitive overload.
With 85% of medical leaders already adopting AI, Lao said, the technology could help physicians navigate the growing complexity of electronic health records (EHRs), multimorbidity and polypharmacy rather than replace clinical expertise.
“The goal of AI nowadays is not just for data retrieval, but for actionable intelligence,” Lao said.
He said physicians must understand how AI reaches its conclusions before incorporating it into clinical practice.
“Machine learning focuses on recognizing patterns, while deep learning uses neural networks to analyze complex features. Beyond just identifying what is there, AI calculates the probability of what will happen,” he said.
AI as clinical assistant
Endocrinologist Dr. Dionise Ysabelle Bawal said AI could allow physicians to devote more time to patient care by reducing routine clinical work and expanding preventive healthcare through hybrid telemedicine models.
“In 2020, telemedicine emerged as a rescue. In 2026, it’s now meant to prevent. We can do continuous monitoring and be proactive in our patients,” she said.
She cited the increasing use of consumer-grade wearable devices such as continuous glucose monitors (CGMs), smart rings and smartwatches, which enable physicians to remotely monitor patients and intervene before complications arise.
Radiologist Dr. John Patrick Almazan described AI as a tireless “second reader” that could help reduce fatigue-related errors in busy radiology departments.
“AI acts as a safety net to prevent fatigue-related misses. AI is not foolproof, but it causes fewer misses on the busy list,” he said.
Almazan said radiologists were evolving into consultants who oversee a patient’s “clinical digital twin” — a virtual model that combines imaging, genomic, pathology and laboratory data to provide a more comprehensive picture of health.
Dr. Dennis Flores, vice president of the UST Medical Alumni Association, said advances in genomics were likewise transforming infectious disease management by allowing physicians to tailor treatment to individual patients instead of relying solely on broad-spectrum therapies.
He said pathogen sequencing could identify antimicrobial resistance markers within hours instead of days, allowing clinicians to begin appropriate treatment sooner.
Human judgment remains indispensable
While AI offers unprecedented speed and analytical capability, speakers emphasized that physicians remained responsible for validating every recommendation the technology produces.
Dr. Clarissa Mendoza, chair of the UST Hospital Cardiology Section, said physicians’ medical training would enable them to critically evaluate AI-generated information.
“AI is everything, but [it cannot be] hook, line, and sinker. The advantage we have is medical knowledge. We have the background to discern what is true,” she said.
Dr. Maria Cristina Maranion, chief program officer for clinical programs of the UST Faculty of Medicine and Surgery, said effective AI use also depended on crafting precise prompts.
She recommended platforms such as Perplexity or Claude for high-stakes clinical decision support, while reserving ChatGPT or Gemini for administrative tasks to improve efficiency without compromising professional standards.
Maranion added that AI-assisted clinical work must always be accompanied by transparency and independent verification. She also predicted that AI-powered diagnostic support could eventually become part of the standard of care.
“Soon, not using AI to help determine diagnosis or treatments could be seen as malpractice. Gone were the days [of] one size fits all. We need to prescribe the right drug for the right patient at the right time,” she said.
Risks of overreliance
While AI offers significant clinical advantages, speakers warned that excessive dependence on automated recommendations could weaken physicians’ diagnostic reasoning.
Pulmonologist Dr. Earl Louis Sempio said medicine should adopt emerging technologies cautiously because errors have immediate consequences for patients.
“If medicine is the first to use a particular technology and you find out the pitfalls while already using it, patients will die,” Sempio said. “It’s better that other industries use it first, so we see the potential problems before we adopt it.”
Physicians remain legally accountable under the Medical Act of 1959 because AI is neither a licensed practitioner nor a legal person, he said.
“If the data the doctor uses is faulty at the start, the harm happens at the ‘byteside’ before you even get to the bedside,” Sempio said. “You’re the sole bearer of the responsibility.”
Sempio said AI should instead relieve physicians of repetitive cognitive tasks, allowing them to spend more time interacting with patients and making informed clinical decisions.
Cardiologist Dr. Alena Pias Bantolo emphasized that clinical judgment remained uniquely human.
“Humans and AI are actually an incomparable concept,” she said. “Clinical judgment is what differentiates us. It is perceived as an intuitive art that is rooted in observation, introspection, and dialogue.”
Internist Dr. Dylan Jansen Taytayan echoed the concern, warning that excessive reliance on automated systems risks reducing medicine to a transactional process.
The postgraduate course, titled “Unlocking Strategies for Tomorrow: Artificial Intelligence and Precision Medicine in Current Clinical Practice,” was held on April 17 and 18 at the St. John Paul II Building.
