☕️
Hello, I’m Maicel Eugenio Monzón Pérez. I’m a physician, biostatistician, and data scientist, passionate about transforming health research through science, statistics, and artificial intelligence. I graduated as a General Practitioner in 2003 and later pursued a Master’s in Health Informatics, the Biostatistics specialty (first and second degree), and finally a PhD in Medical Sciences at the University of Medical Sciences of Havana. Throughout my career, I have worked on the design and analysis of biomedical research, with a particular focus on clinical trials and the application of machine learning models to solve real-world problems in health. I also enjoy programming in R and Python, creating tools that enhance research efficiency and improve clinical decision-making. Today, I share my experience as an Assistant Professor at the National School of Public Health and as a Clinical Trial Reviewer at the Center for State Control of Medicines, Equipment, and Medical Devices. My purpose is clear: to strengthen a more robust, innovative, and evidence-based health system, bringing together medicine, statistics, and technology to generate real impact in people’s lives.
General Practitioner (Basic Medical Degree)
University of Medical Sciences of Villa Clara
Master’s in Health Informatics
Center for Cybernetics Applied to Medicine (CECAM)
Specialist in Biostatistics
National School of Public Health
PhD in Medical Sciences
University of Medical Sciences of Havana
Sequel to the ‘Lie to the Professor’ experiment. I challenged the silicon colossi with absurd correlations. Not one spotted the nonsense—revealing an algorithmic ‘horror vacui’ that poses a threat to scientific integrity.
I asked my residents to invent hemoglobin data in 60 seconds. I used six forensic techniques to catch them. Spoiler: Statistics always wins.
Practical Guide with R for Developing Robust Predictive Models in Clinical Settings
IMRaD is the publication standard, but it need not dictate the writing process. This post argues for a ‘reverse’ writing strategy—from findings back to the introduction—as a method to enhance honesty and clarity, distinguishing it from HARKing and aligning narrative with genuine scientific discovery.
This analysis examines the essential pillars that define Large Language Models (LLMs), from their technical conception to their practical implications and inherent challenges
