profile image

Michele Romani

PhD Student in Brain-Computer Interfaces

Contact Me

Biography

I am Michele Romani, and I come from Asola, a little town in the countryside of Northern Italy near the beautiful Garda lake. As far as I remember, I have always been passionate about understanding our relationship with technology. I clearly remember the first time I used a computer: it was an immediate click. It was not just a mere work tool, but an extension of my brain, and one that allowed me to progressively unleash my creativity and cultivate my skills like anything else before. I delved into drawing, editing, modding games, learnt English, problem solving, and then when years later I begun my studies and started programming seriously, I realized that my initial intuition was correct and I could achieve much more by implementing my own intuitions and ideas into real applications. I strongly believe that my work should have a positive impact on the relationship between humans and technology, and that AI will play a major role in better understanding the underlying cognitive processes of the brain in the next iteration of neuro-scientific research and applications. At the core, I see myself as a "designer": my work lies on the intersection between technology and human behavior, and as such I am focused on enhancing the interaction between users and computers through the design of Human-Centered AI applications.

Work Experience

Research & Development - g.tec Medical Enginering GmbH (Sept 2023 - May 2024)

Development of Brain-Computer Interfaces for Gaming

Schiedlberg, Austria.

Research Engineer - myBrainTechnologies (Apr 2022 - Jul 2023)

Research and development of ML pipelines in Python for neuroscience applications.

Paris, France.

Research Engineer, Intern - University of Twente - myBrainTechnologies (Mar 2021 - Sep 2021)

Research project in Python on emotions classification during music listening.

Enschede, The Netherlands.

Research Developer, Intern - Université Paris-Saclay - LRI, INRIA (Mar 2020 - Jun 2020)

Research project in Unity on collaborative remote environment using Virtual and Augmented Reality and MRTK

Paris, France.

Software Engineer - TAI Solutions (Feb 2017 - Feb 2019)

Development of Java SpringBoot distributed applications with API REST for several b2b projects.

Milan, Italy.

Software Engineer, Intern - TAI Solutions (Aug 2016 - Jan 2017)

Thesis project on the evaluation of MongoDB's performances for distributed b2b applications.

Pisa, Italy.

Featured Projects





brain-hockey"/

Brain Hockey 2024

Neurogame - Brain-Computer Interfaces, Machine Learning, EEG, Unity

BrainHockey is an innovative take on the beloved Pong game, where players use their brainpower to control the paddle instead of traditional keyboard or mouse inputs. Designed for EEG devices, BrainHockey taps into the potential of BCI technology, allowing players to move their paddle simply by focusing and concentrating on the Neurofeedback Targets.

View on Github

green-shield

The Green Shield 2023

Neurogame - Brain-Computer Interfaces, Machine Learning, EEG, Unity

The Green Shield is a revisited version of Space Invaders where the player can use a BCI to create a barrier and stop the alien invasion. It has been developed using the Unicorn Unity interface for G.Tec Unicorn Hybrid Black and Unicorn Headband devices. It can be played as flat screen version on a Windows PC or in VR version on Meta Quest.

View on Github

flower

FLOWER 2023

Design Concept - Machine Learning, EEG, Figma

FLOWer is a revisited design of the classic P300 speller developed during BR41N.IO 2023 hackathon that aims at: - Improving accuracy: analysis of the data, preprocessing and training a CNN - Improving UX: design of a more intuitive flower interface for a region-based paradigm - Improving ITR: design a solution that makes use of NLP/LLM to create a predictive adaptive interface The P300 classifier has been implemented in Python, the interface has been designed in Figma

View on Github

View on Figma

doc-hoc

Doc-Hoc 2022

Design Concept - Machine Learning, EEG, Figma

Doc-Hoc is a concept for a BCI-powered assistant for language learning based on the OpenBCI ecosystem and designed during the BR41N.IO 2022 hackathon. The idea behind Doc-Hoc is to provide analytics on the level of relaxation and concentration to help the students organising their study and avoid stressful sessions

View on Github

View on Figma

music-emotion

Music-Emotion 2021

Research - Machine Learning, EEG

I developed a pipeline to process and classify emotional valence and arousal in the brain activity of users listening to music. The pipeline can be the proof of concept for a future recommending system based on emotions.

View on Github

mindworms

Mindworms 2021

Neurogame - Brain-Computer Interfaces, Machine Learning, Unity, EEG

Mindworms is a Worms-like game in Unity, developed to showcase the possibility of using brain activity as an input for a game. In Mindworms, the power of the player's attacks is modulated by the sensorimotor rhythm of the players.

View on Github

mindworms

Amuleto, Sincere Pill Dispenser 2020

Design - Unreal Engine, Illustrator, Figma

Amuleto is a sincere and courageous pill dispenser designed to create an experience that helps patients cope with their diagnosis, both from the social and health perspectives. The project was the outcome of a Multisensory Design process that aimed at including multiple human senses in the design of a product.

Watch demo

mindworms

XR Music Lab 2020

Serious game - Mixed Reality, Unity

XR Music lab is a creative application for mixing music usign two different types of virtual interaction. In the VR app the player learns the basics of music composition by turning on totem beats. In the AR app the player can use an augmented console with leap motion and a webcam to transpose their hands in the virtual environment. Both applications have been built using Unity.

View on Github

Watch demo

mindworms

Exhibition of a lifetime 2020

Art & Entrepreneurship - Virtual Reality, Unity

Exhibition of a Lifetime is a VR art gallery where spectators are driven through a visual and audio experience that represents the four phases of life: childhood, youth, adulthood and seniority. The goal of the project was to explore innovative ways for people to interact with art pieces

View on Github

mindworms

Gamerly 2020

Design - Figma

Gamerly is a platform for game developers to find coaching and other peers to work with on new game projects. We designed Gamerly as a challenge to innovate the market of game development by adding a networking component to build new teams and with the possibility of requesting coaching from senior members of the community.

View on Figma

mindworms

Turkey Jump! 2019

Videogame - Unity

Turkey Jump! is a 2D platformer game developed in Unity, as part of a study on alternative input methods for videogames. The game is a clone of the famous and unforgiving Flappy Bird, but instead of touch pressure, the jumps of the turkey Kevin are triggered using the volume of the player's voice.

View on Github

mindworms

IOT Vulnerability 2018

Webapp - Security, IOT, ELK

We developed an analysis tool to find vulnerable IOT devices using the Shodan search engine and visualize them on the ELK stack with several combination of metrics. The pipeline has been automated using Apache NiFi.

View on Github

mindworms

RoadBuddy 2016

App - Android

RoadBuddy is an app that we designed and developed for bikers to organize their trips together. Based on the Google Maps API, the core idea is to share a trip and stops for a group of people driving together on different vehicles.

View on Github

mindworms

Throw Away 2015

Design - Illustrator

ThrowAway is a design prototype of an interactive tangible serious game to teach children how to properly recycle waste.

Watch Demo

Research

Author

Music-Emotion: Towards real-time recognition of affective states using a wearable Brain-Computer Interface 2021

This research set out to investigate the feasibility of performing Emotion-Recognition using Melomind, a wearable neural interface manufactured by myBrainTechnologies. Melomind is capable of recording EEG signals, that can be processed using machine learning algorithms in the form of a classification task of the emotional dimensions of valence and arousal...

Read Now

Github

AR Notes: Augmenting Collaborative Tools 2020

Over the last decades of HCI research, it has become clear that humans interact more naturally when in presence of a tangible interface that recalls them of a functionality they already know from a previous non-digital experience. However, digitalization and automation of many tools we use on a daily basis have made complicated integrating tangibles and software in...

Read Now

Design Report

Github

Co-author

Hybrid Harmony: A Multi-Person Neurofeedback Application for Interpersonal Synchrony 2021

Recent years have seen a dramatic increase in studies measuring brain activity, physiological responses, and/or movement data from multiple individuals during social interaction. For example, so-called “hyperscanning” research has demonstrated that brain activity may become synchronized across people as a function of a range of factors. Such findings...

Read Now