Dr. Pedro Miguel Rodrigues is one of the authors of the highly cited article entitled “Cardiovascular Diseases Diagnosis Using an ECG Multi-Band Non-Linear Machine Learning Framework Analysis” published in Bioengineering (ISSN: 2306-5354).

The following is an interview with Dr. Pedro Miguel Rodrigues:

1. Can you tell us a bit about your background and what your research focuses on?
I hold a B.Sc., M.Sc., and Ph.D. in biomedical engineering with specialized expertise in signal processing and artificial intelligence. I am a Professor and Vice-Dean for Lifelong Learning and Partnerships at the Faculty of Biotechnology at Universidade Católica Portuguesa, and a researcher at the Centre for Biotechnology and Fine Chemistry in Porto, Portugal. My primary focus involves applying advanced biosignal processing techniques and machine learning methods to clinical datasets. My research interests revolve around developing algorithms designed to detect and monitor diseases (such as Alzheimer’s, Parkinson’s, cardiovascular conditions, and voice disorders) before their earliest symptoms appear. By working with various biosignals—like EEG, ECG and voice signals—and diagnostic imaging inputs such as MRI, I aim to streamline early detection efforts, minimize subjective interpretation errors, and ultimately offer medical doctors more reliable tools for patient diagnostic evaluation.

2. What made you decide to publish a bioengineering article? Why did you choose MDPI’s Bioengineering?
My team and I were motivated to publish in a venue that places equal emphasis on biology and engineering, given our interdisciplinary approach that intertwines computational techniques with clinical applications. MDPI’s Bioengineering appealed to us through its strong open-access principles, robust international readership, and efficient editorial process. Moreover, it provides a platform where researchers from multiple fields can discover how AI-based diagnostic frameworks apply to a variety of medical contexts, making it a compelling choice for dissemination.

3. Was it important to you that the journal is open access? How does open access publishing advance the field of bioengineering?
It is important to publish open access for several reasons. First, open access ensures the work can be readily examined and implemented by researchers and clinicians operating in diverse settings, including resource-limited regions. Second, it expedites the feedback loop between research and practical implementation, as anyone has access to methods to build upon them directly. This widespread accessibility enables faster progress in addressing pressing biomedical challenges, ultimately benefiting the entire field by nurturing collaboration and transparency.

4. What are the current bottlenecks in this field, and how did you identify your research’s breakthrough point?
One of the most persistent challenges is handling the sheer variation in biosignal quality and data formats across different clinical environments. For instance, older EEG or ECG machines often produce outputs that differ in resolution or file structure from newer devices. Our breakthrough came when we applied advanced feature extraction and domain adaptation techniques, enabling the algorithm to effectively learn from multiple datasets, each with its own unique characteristics. Once we observed that our models maintained strong performance across disparate clinical subpopulations, we realized that our approach was both generalizable and scalable.

5. Why do you think this article has been highly cited?
Conditions such as Alzheimer’s disease exert a profound global impact, affecting nearly every healthcare system. Because of this broad relevance, our work was widely cited by researchers confronting similar challenges across a variety of academic institutions, laboratories, and hospitals worldwide. Moreover, publishing in an open-access format greatly accelerated the distribution of our findings, helping us reach a broader audience much more quickly.

6. Are there follow-up studies planned based on this paper’s findings?
Yes, several lines of research are currently in progress. We are broadening our predictive frameworks by adopting a multimodal approach that integrates EEG, MRI, and genetic (including microbiome) data aiming to harness insights that could further enhance early detection. Additionally, we are working with clinicians who track patients over multiple years, allowing us to evaluate our models’ accuracy and adaptability as various diseases progress.

7. Which technological directions in bioengineering deserve the most attention over the next 5 years?
Over the next five years, bioengineering stands to benefit significantly from AI-driven multi-modal analytics, which will integrate data from advanced sensors, wearables, imaging technologies, and a rapidly expanding array of IoT devices. By combining these continuous data streams with refined machine-learning frameworks, researchers, and clinicians can achieve real-time diagnostics and interventions that transcend traditional healthcare models. Meanwhile, edge cloud computing will play an indispensable role in reducing latency and safeguarding sensitive patient information by processing data on-site—even in resource-constrained or remote areas—which broadens global access to quality healthcare. These converging technologies promise to redefine clinical decision-making, expand the reach of medical services, and fuel ongoing innovation in patient care throughout the coming decade.

8. How is AI reshaping bioengineering research in disruptive ways?
AI is bridging the gap between massive raw datasets and actionable clinical insights by recognizing patterns previously invisible to human observers. These data-driven approaches reduce costs by streamlining diagnostics, minimizing invasive testing, and guiding targeted research on the most promising therapeutics. As AI-driven platforms increasingly integrate diverse data sources, imaging, molecular, signals and wearable sensor data, they bolster clinicians’ decision-making, foster early detection of health risks and pathological status.