Mario Lavanga

Computational Neuroscientist, Data Scientist, Postdoctoral researcher at the Theoretical Neuroscience Group - INS - Aix-Marseille University

About Me

Born in Italy and raised by a medical doctor and a stay-at-home mom, since his childhood Mario has developed an early interest in human physiology and biology. Doubtful to seek a clinical career, he went on to study Biomedical engineering at Polytechnic university of Milan, where he delved into the intertwined realities of physics and biology. During his studies, he had the opportunity to build a strong background in the field of biomedical signal processing. He also joined a top-level students program organized by the two main Italian technical universities, known as Alta Scuola Politecnica (ASP), which provided opportunities to learn a multidisciplinary approaches to solving problems. The skills he acquired have been useful in his working experience, which is characterized by frequent interactions with clinical doctors.

Immediately after graduating, he moved to KU Leuven, Belgium, where he started focusing on mental health and neurodevelopment as part of his PhD studies, under the supervision of Prof. Sabine Van Huffel, Prof. Alexander Caicedo Dorado, Prof. Gunnar Naulears and Prof. Els Ortibus. In January 2017, he received a strategic basic research grant from the Flemish scientific research fund (FWO) to quantify stress in premature babies by means of EEG and ECG signal processing and to understand its impact on the infants’ development. |

After its graduation, he moved to Aix-Marseille University, France, for a postdoc position. His new research focuses on the mathematical modelling of neurodevelopment and ageing based on the Virtual Brain technology developed by Prof. Viktor Jirsa . This project is part of the European effort of the Human Brain Project.

Recent Publications

Lavanga M, Stumme J, Yalcinkaya B H, Fousek J, Jockwitz C, Sheheitli H, Bittner N, Hashemi M, Petkoski S, Caspers S, Jirsa V, The virtual aging brain :a model-driven explanation for cognitive decline in older subjects. bioRxiv. Feb. 20, 2022. doi, 10.1101/2022.02.17.480902.

Lavanga M, De Ridder J, Kotulska K, Moavero R, Curatolo P, Weschke B, et al. Results of quantitative EEG analysis are associated with autism spectrum disorder and development abnormalities in infants with tuberous sclerosis complex. Biomed. Signal Process. Control . 2021 July;68:102658.

Lavanga M, Bollen B, Caicedo A, Dereymaeker A, Jansen K, Ortibus E, Van Huffel S, Naulaers G. The effect of early procedural pain in preterm infants on the maturation of EEG and heart rate variability. PAIN. 2021 May;162(5):1556–66.

Lavanga M, Heremans E, Moeyersons J, Bollen B, Jansen K, Ortibus E, Naulaers G, Van Huffel S, Caicedo A. Maturation of the Autonomic Nervous System in Premature Infants :Estimating Development Based on Heart-Rate Variability Analysis. Front Physiol. 2021 Jan 12.

Projects

Modelling aging and neurodevelopment with The Virtual Brain

The objective of this project is to create a virtual aging cohort based on the 1000 brains study dataset. Based on the structural information of each of the study participant, the functional activity can be replicated with the whole-brain modelling approach of the Virtual Brain. The ultimate goal is to describe the functional trajectory of aging with in-silico approaches and understand which structural processes occurring in aging might affect the brain functions and functional networks. The modelling software has been built in Python and JupyterLab, shared and monitored with Git and executed with SnakeMake and Slurm on hyper-performance computing facilities. The preprint of this study can be found at the following link , while the code will be available soon on this github repository .

The Perinatal Stress Calculator

PhD Thesis Manuscript

Modeling early-life and pain in premature infants

My PhD research focused on the detection of early-life pain and stress in premature infants by means of physiological signal processing. Patients were monitored at different age during the neonatal intensive care unit stay. A variety of clinical oriented metrics were derived to quantify stress and understand the relationship between physiological instabilities and future development outcomes. The results show that infants with a dysmature EEG in the first days of life might be associated with a greater pain experience, that could potentially lead to a lower development outcome and more internalziing behavior. The project was developed in Matlab and the code of the stress detection models along with other automatic development analysis in preemies will soon be available on GitHub. This research was supported with the strategic basich research fellowship by the FWO. The PhD thesis manuscript can be download at this link .

Possible scenarios to introduce a blockchain system for e-voting

The project aimed to introduce possible solutions for a reliable e-voting system based on blockchain techonology. This approach could especially improve the democratic engagement for long-range commuters and citizens living abroad.

Experience

TNG - INSTITUT DE NEUROSCIENCES DES SYSTÈMES

Postdoctoral Researcher

October 2020 - Present

INS Website

I am currently working as a postdoctoral researcher at the INS focusing on functional connectivity in the contest of healthy ageing. I am part of the theoretical neuroscience group and I work as part of a team dedicated to the resting state brain. I spend most of my time working on Python applications in Jupyter-lab to compute a variety of biomarkers associated to ageing and reproduce with biophysical models. My team and I are investigating functional connectivity dynamics and static functional connectivity in both the context of pathological and healthy ageing. In parallel, we are deploying a variety of computational strategies to tune our models and reproduce the empirical patterns by means of supercomputer facilities.

Thanks to my doctoral studies, I can bring my perspective in the empirical data analysis, especially related to ageing. On top of that, I provide specific views to enhance the comprehension of our team’s results to the clinical side of the project. I enjoy learning Python and Jupyter as tools for scientific computing, as well as the schedule manager such as Slurm or parallel computing tools such as SnakeMake.

Flemish scientific research fund (FWO)

S.B. PhD Fellow at ESAT departement, STADIUS group, KU Leuven

January 2017 - September 2020

FWO website

During my years as a SB PhD fellow, I developed a model to quantify stress and pain in premature infants and to understand the physiological processes behind the infants’ development. The first part of my research focused on the assessment of development inside the neonatal intesive care unit by means of physiological signal processing. I investigated the scalp-EEG connectivity, the nonlinear properties of the neonatal EEG signal and the fractality of the heart-rate varaibility of the premature infant in the context of NICU development. The quantification of maturation trends by data processing is usually known as brain-age models. The codes associated to each of these studies were developed in Matlab.

The second part of my PhD was focused on the development of models and metrics to quantify pain and stress in premature infants. Based on the investigation of the development processes in the first of my PhD, I combined the information associated to the nonlinear properties of EEG and the heart-rate variability as well as the interaction among EEG signals and between EEG and heart-rate to detect stress or pain in premature infants. These approaches focused on the resting-state activity as well as the specific phenomenon of heart-rate drops, known as bradycardias. The outputs of these analysis were combined in an investigation effort to understand if the accumulation of pain in the intensive care units might lead any difference in maturation trajectories. The results showed greater discontinuity in EEGs were associated to a greater accumulation of pain by preamture infants. The codes associated to each of these studies were developed in Matlab.

KU LEUVEN

PhD Fellow at ESAT departement, STADIUS group

January 2016 - December 2016

STADIUS website

My first year as PhD researcher was fully dedicated to the full comprehension of the specific domains of neonatal physiology and development as well as machine learning and graph theory. These tools were necessary to create and design the maturation and stress models for my doctoral studies.

Education

KU Leuven

Doctor of Engineering Science (PhD) - Electrical Engineering

2016 - 2020

Throughout my PhD studies, I perfectioned my research and communication skills to deliver clinical sounding results based on my mathematical and data science knowledge. The PhD certificate can be download at this link .

Alta Scuola Politecnica

2013 - 2015

In parallel to my master of sciences, I joined a double-degree program organized by the two major Polytechnic university of Italy. The objective was providing a set of skills that could go beyond the knowledge and the scope of a typical brick-and-mortars class. Over two years, I followed a variety of extra seminars and course that go beyond our domain of interests and the success of the programm was evaluated with a multidisciplinary project. The final report and output were not only evaluated on the basis of the provided technical solutions, but also the broader scope and impact for the economy and the society.

Politecnico di Milano

MSc in Biomedical Engineering

2013 - 2015

During my master of sciences years, I developed a passion for the biomedical signal processing and biophysical models to explain physiology in mathematical terms. It was also the period that allowed me to perfection my programming skills, to understand data science, machine learning and their application and to appreciate the necessary mathematical theories to combine programming with data and pattern analysis. The master of sciences diploma supplement can be downloaded at this link .

Politecnico di Milano

BA in Biomedical Engineering

2010 - 2013

Throughout my bachelor, I learnt the basis of engineering science, spanning from chemistry, continuum mechanics to electronics and economics. My last year was also the occasion to get closer to the biomedical data analysis and to get a glimpse into the research world. The bachelor diploma supplement can be downloaded at this link .

Languages

  • Italian Mothertongue

  • English Professional Knowledge

  • French Intermediate Knowledge

  • Dutch Basic Knowledge

Skills & Proficiency

Matlab

Python

Latex

Git

HTML5 & CSS

Slurm

LabView