The following is an interview with Prof. Mesin:

1. Can you tell us a bit about your background and what your research focuses on?
I am an electronics engineer with a Ph.D. in applied mathematics. Currently, I serve as an Associate Professor in biomedical engineering. My main research interests lie in the processing and analysis of biomedical data (both signals and images). One focus of my research is on EEG signal acquisition, preprocessing, and functional connectivity analysis.

2. What made you decide to publish a bioengineering article? Why did you choose Bioengineering MDPI?
As an Associate Editor of Bioengineering, I had the opportunity to contribute a review paper in an area closely aligned with the journal’s scope. The choice of MDPI was driven by its strong reputation for rapid and transparent peer review, as well as its commitment to open access dissemination.

3. What was your experience publishing with Bioengineering MDPI?
The experience was very positive. The review process was efficient and timely, and the editorial staff ensured high-quality editing and publication standards.

4. Was it important to you that the journal is open access? How does open access publishing advance the field of bioengineering?
Yes, open access is particularly important for researchers like me. I tend to focus on the scientific substance of my work and less on promotion. Unfortunately, this sometimes means that even my most innovative work goes under-recognized. Open access provides a valuable opportunity for broader dissemination, making research accessible to a larger audience regardless of institutional affiliation or financial resources.

5. What do you hope that readers will get from your paper?
The paper offers a comprehensive tutorial review of data-driven methods for assessing EEG-based functional connectivity. It aims to serve as a reference for researchers by summarizing methods, clarifying technical details, and discussing the impact of acquisition and preprocessing on connectivity estimates.

6. What critical scientific or engineering problems did your research initially aim to address?
We aimed to address the complexity of accurately inferring functional connectivity from EEG data. This includes methodological challenges in preprocessing, signal interpretation, and the selection and validation of appropriate connectivity metrics.

7. What are the current bottlenecks in this field, and how did you identify your research’s breakthrough point?
Key bottlenecks include variability in preprocessing methods, lack of standardization, and the difficulty of distinguishing true from spurious connectivity. Our review highlights the importance of methodological rigor and proposes emerging solutions, such as high-order interactions and graph-based approaches, to advance the field.

8. Which technologies or tools played pivotal roles in designing your methodology?
Our work focuses primarily on data-driven methodologies, including both linear and nonlinear statistical tools. Concepts such as Granger causality, transfer entropy, and multivariate autoregressive modeling are central, alongside tools from information theory and graph theory.

9. Have your experiments or theoretical models undergone significant adjustments? What motivated those changes?
Yes. Throughout the years, our approaches have evolved to incorporate more sophisticated multivariate methods and to address limitations identified in simpler bivariate analyses. These changes were motivated by a deeper understanding of EEG complexity and the risk of misinterpreting indirect connections.

10. Are there any untold “behind-the-scenes” stories worth sharing about this work?
The paper was the result of a genuine collaborative effort between two groups with complementary expertise. Early discussions revealed the need for a unified framework that could bridge theoretical developments and practical EEG applications, which ultimately shaped the paper’s structure.

11. Why do you think this article has been highly cited?
Citation patterns are influenced by complex and often unstable dynamics. A well-known phenomenon is that highly cited papers tend to attract further citations, especially when they receive attention early after publication. In our case, the open access format ensured wide visibility from the outset, and the comprehensive scope of the review likely made it a useful reference for researchers entering or working within the field. Moreover, the collaboration among experienced authors with established scientific reputations may have added credibility and contributed to its early recognition.

12. Are there follow-up studies planned based on this paper’s findings?
Yes, several ongoing studies in our groups are extending the reviewed methodologies to real-world EEG datasets, particularly in clinical and cognitive neuroscience applications.

13. Did your research involve cross-disciplinary collaboration? How did teamwork shape the outcomes?
Definitely. My group was more focused on EEG acquisition and preprocessing, while Prof. Faes’ group contributed theoretical expertise in connectivity metrics. This synergy was essential to producing a balanced and integrative review.

14. How did early career researchers or students contribute to this work?
They played a fundamental role. Early career researchers carried out most of the literature review and technical analysis, under the supervision of senior authors who guided the overall structure and interpretation.

15. What was the greatest technical or theoretical challenge during this research, and how did you overcome it?
One major challenge was organizing a vast and fragmented literature into a coherent and accessible framework. We addressed this by systematically categorizing connectivity metrics across time, frequency, and information-theoretic domains, and by explicitly linking them through their mathematical foundations.

16. Were there difficulties in data acquisition or experimental reproducibility? How were they resolved?
While the paper is primarily a review, we are well aware that reproducibility is a major issue in EEG research. We emphasized best practices in preprocessing and model validation to improve reproducibility and highlighted the importance of transparent methodological reporting.

17. Did ethical concerns (e.g., gene editing, biosafety, etc.) arise? How were they addressed?
No specific ethical concerns arose in this work, as it is a review of methodological approaches and did not involve new experimental protocols with human subjects.

18. Which technological directions in bioengineering deserve the most attention over the next 5 years?
In the coming years, I believe particular attention should be given to the integration of multiple neuroimaging modalities, which can offer complementary spatial and temporal resolution for studying brain function. The development of non-invasive brain–computer interfaces will continue to open new possibilities for clinical and assistive technologies. Equally promising is the evolution of adaptive neurostimulation systems, enabling a deeper and more personalized interaction between the user and the device. Lastly, the incorporation of explainable artificial intelligence into biomedical signal processing holds great potential for improving both diagnostic transparency and clinical trust in AI-based systems.

19. How is AI reshaping bioengineering research in disruptive ways?
AI enables the analysis of complex, high-dimensional data and fostering real-time decision-making in clinical contexts. In EEG analysis, deep learning is being explored for source localization, artifact rejection, and connectivity estimation, though interpretability remains a key challenge.

20. What learning resources would you recommend for newcomers entering this field?
I recommend starting with foundational texts on signal processing and time series analysis, followed by specialized literature on EEG analysis and functional connectivity. Our review paper itself offers a structured entry point. Practical experience with tools like MATLAB, Python, and EEGLAB is also invaluable.