
Johann Faouzi
Présentation
My name is Johann Faouzi and I am a postdoctoral research fellow at Paris Brain Institute, Paris, France. My main research interest is machine learning, with applications to neuroscience and open source software.
My PhD thesis was about predicting impulse control disorders in Parkinson's disease. Impulse control disorders are psychiatric disorders characterized by difficulty in self-control of emotions, thoughts and behaviors. These disorders are common in Parkinson's disease and associated with impaired quality of life for patients and increased burden for caregivers. We investigated the predictability of these disorders from associated or suggested clinical and genetic risk factors by training machine learning algorithms. We also investigated the genetics of these disorders using genetic risk scores. Finally, in a more methodological work, we investigated the integration of time-independent data (such as genetic data) in recurrent neural networks.
I also had the opportunity to work on several other projects with many people, including several literature reviews (one on deep learning for brain disorders, one on impulse control disorders in Parkinson's disease, one on machine learning for Parkinson's disease and related disorders, and one on the prediction of mild cognitive impairment in Alzheimer's disease using machine learning), as well as other studies on machine learning applied to neuroscience (a challenge on brain-age prediction and a study on functional brain connectivities in Tourette disorder).
I am also interested in machine learning for time series. I have created a Python package dedicated to time series classification and contributed to other open source Python packages (not necessarly on time series). I am a strong believer of the importance of good programming practices, in particular for people working in academia (both professors and researchers).
Publications
Publications
|
Predicting Impulse Control Disorders in Parkinson's Disease: A Challenging TaskInternational Congress of Parkinson's Disease and Movement Disorders, Sep 2019, Nice, France
Communication dans un congrès
hal-02315533
v1
|
|
Classic machine learning methodsOlivier Colliot. Machine Learning for Brain Disorders, Springer, 2023
Chapitre d'ouvrage
hal-03830094
v4
|
|
Machine learning for Parkinson’s disease and related disordersOlivier Colliot. Machine Learning for Brain Disorders, Springer, 2023, ⟨10.1007/978-1-0716-3195-9_26⟩
Chapitre d'ouvrage
hal-03830073
v2
|
|
Time Series Classification: A review of Algorithms and ImplementationsKetan Kotecha. Machine Learning (Emerging Trends and Applications), Proud Pen, In press, 978-1-8381524-1-3
Chapitre d'ouvrage
hal-03558165
v1
|
|
Machine learning to predict impulse control disorders in Parkinson's diseaseArtificial Intelligence [cs.AI]. Sorbonne Université, 2020. English. ⟨NNT : 2020SORUS048⟩
Thèse
tel-03090079
v2
|