News
| Mar 18, 2026 | I completed the Machine Learning in Production certification by Andrew Ng! 🎓 The Machine Learning in Production course highlighted how building reliable ML systems requires far more than training models. It emphasized the importance of data-centric development, systematic error analysis, and treating ML as an iterative pipeline spanning scoping, data, modeling, and deployment. I also learned how to monitor models under data and concept drift, evaluate performance beyond average accuracy, and design robust ML pipelines. Through hands-on exercises, I deployed computer vision models with Docker and AWS and explored how dataset composition and labeling strategies affect performance. Coming from a background in medical imaging pipelines, these lessons reinforced the importance of data quality and engineering discipline in building trustworthy AI systems. I’m looking forward to continue applying these principles — and many of Andrew’s practical tips — when building robust and reproducible AI systems. Certificate Here! |
|---|---|
| Jan 23, 2026 | 🚀 WEF Hackathon 2026 – Turning Energy Market Data into Actionable Intelligence I participated in the WEF Hackathon 2026, where our team built an Energy Market Intelligence Agent to transform complex energy market time series into actionable insights for decision-makers. The system combined statistical signal analysis with AI agents to detect anomalies, link price movements to underlying drivers such as wind, solar, and consumption, and generate trader-oriented explanations and interactive visualizations. This experience highlighted how accessible AI tooling has become — and how the real challenge lies in understanding business needs and designing systems that support better decisions. |
| Dec 10, 2025 | Presented two posters at the AI conference EurIPS in Copenhagen! 🔥 I attended the european counterpart of the top-tier ML conference NeurIPS in Copenhagen, where I presented my Bayesian mixture modeling framework to evaluate the reliability of repeated measurements in brain networks. The model represents each connection as a probabilistic mixture capturing the presence or absence of a true brain connection and estimates reliability across repeated measurements. We first validated the approach using synthetic repeated-measures data with controlled noise mimicking real neuroimaging variability, before applying it to real datasets. The model successfully detected connections with varying reliability and produced biologically meaningful estimates, while also revealing important limitations in recovering absent connections. Although developed for brain connectivity, the framework is broadly applicable to any network derived from repeated observations where uncertainty quantification is critical. The conference was additionally a great opportunity to learn more about the latest advances in deep learning research, reconnect with friends and former colleagues, and exchange ideas with the community. Exploring the beautiful harbor of Copenhagen was also a wonderful highlight of the trip. |
| Nov 10, 2025 | What an incredible burst of energy, innovation, and vision at Startup Nights 2025! 🔥 Seeing 200+ active startups across Switzerland and neighboring countries absolutely blew me away. Scrolling through countless ideas, I felt genuinely inspired by the potential! My favorite part of the event was meeting people building meaningful healthcare solutions, especially those working in brain imaging, MRI, and biomedical pipelines, areas close to my hearts. The pitching session showcased impressive creativity, from AI tools for elderly care to ear-based breath monitoring for athletes. Watching founders pitch with such clarity and confidence was incredibly inspiring. I also loved volunteering behind the scenes, helping set up and keeping everyone hydrated; a wonderful way to make new friends and feel part of this energizing community. Very grateful to have access to such amazing opportunities! |
| Sep 12, 2025 | I just spent 3 fantastic days in Basel for PharmaCamp 2025! 🧠 On the first day, we gained valuable insights into careers in pharma from industry experts. The next two days were all about the hackathon, where CSEM challenged us to prototype either a smart pipette tip or a microneedle array.I was truly impressed by the dedication of the 25 participants. In just 10 hours, we not only developed sound solutions to real-world problems, but also took on the roles of project managers and market analysts. Very inspiring! I feel especially grateful to have collaborated with an amazing and friendly team of four talented researchers. This experience reinforced my conviction that I am passionate about innovation and that I thrive when working in a connected team towards a common goal. |
| Jul 30, 2025 | New paper out in Cerebral Cortex! 🧠 5 years later the work I did during my master thesis is finally published!  In this study, we investigated how the brain of patient suffering from corpus callosum agenesis (AgCC) preserves a functional default-mode network despite the absence of this major white matter bundle. This work contributes to a better understanding of neuroplastic compensatory mechanisms. Huge credit to Vanessa Siffredi who did most of the heavy lifting to revive the project and bring it to the finish line. 🙌 |
| May 30, 2025 | New paper out in PLOS Biology! 🧠 Defacing MRI images is essential to protect participant privacy, but are there unintended consequences? In this registered report, we showed that defacing significantly alters manual quality assessments of structural MRI images: human raters evaluated the same images differently depending on whether facial features were present. This suggests that defacing removes not only identifiable information but also visual cues critical for assessing image quality. The risk? Poor-quality images may persist despite quality control, potentially undermining the reliability of the study. As a consequence, we recommend to perform quality control before defacing. This work is also the first paper we brought to completion as a registered report, which was not short of challenges, but we learned a ton in the process. Huge thanks to my coauthors for their precious guidance and time investment in this project. |
| May 04, 2025 | Just back from three inspiring days in Boston, where I had the chance to collaboratively contribute to the NiPreps software ecosystem! I spent the time coding alongside like-minded peers, learning from domain experts, and expanding the scope of our tools. My contribution was a deep-dive discussion on the needs to assess quality of Positron Emission Tomography (PET) imaging. This led me to develop a new workflow—drawing inspiration from MRIQC’s functional processing—to generate quality reports tailored to PET. Curious about the progress? You can follow the pull request here! Also had the pleasure of reconnecting with old friends. Fun times ! :) |
| Apr 29, 2025 | Australia here I come! I am thrilled to have been awarded a travel grant to attend the Annual Meeting of the Organization for Human Brain Mapping (OHBM) 2025 in Brisbane, Australia. This opportunity allows me to pursue a long-standing ambition: to premiere the open release of our large multi-modal MRI dataset, the Human Connectome Phantom (HCPh) dataset. My goal is to engage with the neuroimaging community and encourage the adoption of this dataset to explore important and fascinating research questions. At OHBM, I’ll participate in BrainHack—a three-day collaborative coding event—present a poster showcasing our first analyses of the HCPh dataset related to the intra-subject reliability of functional MRI, and, give a talk to share our latest insights drawn from this unique resource. |
| Mar 06, 2025 | Officially Dr. Provins! |
| Jan 14, 2025 | I successfully passed my private defense!  The committee praised my rigorous methodology, the clarity of my thesis manuscript and presentation, and my substantial contributions to the field. Wrapping up this Ph.D. has been both a challenge and a privilege. These four years were filled with hard work, learning, and moments of doubt, but also excitement, discovery, and growth. I’m proud of what I’ve accomplished and grateful for everyone who supported me along the way—now onto the final step, the public defense! |
| Sep 06, 2024 | The data collection of our large multimodal MRI acquisition on a single subject (HCPh) is finished! A single subject underwent 72 sessions of a comprehensive MRI protocol. I organized, managed and carried out the collection of the last 36 sessions that we acquired across 3 scanners including 2 full-time clinical scanners. |
| May 05, 2024 | I am happy to share more about my topic of predilection, quality control 🌟, in a poster at OHBM24 in Seoul, South Korea! In this poster, I am presenting how we leverage a positive control task to assess the integrity of the analysis workflow. More about this important work here. |
| Nov 14, 2023 | We received in principle acceptance by the prestigious journal Nature Methods for our second Stage 1 Registered Report. My team and I have been working hard towards this moment and are thrilled to share the news. We will now immediately start the data collection. |
| Jul 22, 2023 | My first talk at a conference! 😃 I will be presenting what an ideal QC workflow looks like for us, highlighting QC guidelines to improve the reliability of your study, and presenting how to leverage MRIQC and fMRIPrep to assess data quality. Come listen to this educational if you want to implement better quality control (which is crucial for reliability!!) I’m also looking forward to getting insights from the other presenters and discussing this important topic that I’ve been heavily involved in. Thank you to the research topic organizers for putting together this important initiative and giving me the opportunity to participate! |
| Jun 23, 2023 | I am presenting a poster at OHBM22 in Glasgow, Scotland! I am presenting my poster entitled “Defacing biases manual and automated quality assessments of structural MRI with MRIQC”. Pass by my poster number WTh566 tomorrow thursday 23.06 at 1.45pm to share questions, ideas or suggestions. A preview of the poster content is available here. |
| Jun 01, 2023 | I am starting a doctoral research visit in Stanford University! 🌲 In my Ph.D., I am essentially contributing to expand the tools that were initiated here, so it is an honor to be at the source 😊. Through my stay, I hope to gain perspective around open science stakes and solutions and to expand my professional network through meaningful collaboration. |
| May 31, 2023 | My first registered report received Stage 1 in principle acceptance from the Peer Community in Registered Report. In this confirmatory study, I will investigate whether the preprocessing step of defacing T1w images impact how their quality is perceived by humans and machines. |
| May 05, 2023 | Fantastic team got us a poster at OHBM23 in Montreal, Canada! Incredibly proud to have led a project at the BrainHack Geneva with a team talented enough to get ourselves a poster at OHBM!  In this work, we leverage the image quality metrics generated by MRIQC to predict the extent of head motion in T1w images using machine learning. Using feature extraction, we also showed that signal-to-noise ratio is the most important feature for this prediction. Pass by our poster to get to know more about our methods and the results  |
| May 05, 2023 | I am presenting a poster at ISMRM22 in London! Next week, I will be presenting my abstract called “Quality control and nuisance regression of fMRI, looking out where signal should not be found” at the ISMRM conference. Here is the link to my abstract. Meet me at 14:45 on Thursday 12 May at the posters session, if you have some questions, suggestions or ideas. |
| Dec 19, 2022 | My first paper got accepted! In this paper, I present the quality control protocol we developed for functional MRI. For more information, checkout the projects page. |