Yassine T.

Consulting and development in automation, data analytics, and machine learning.

Research assistant in the Centre for Developing Brain at King's College London. Working on the prediction of neurodevelopmental outcomes from neonatal brain connectivity data with deep learning methods. Paper published: https://doi.org/10.1101/2020.09.28.317180

Machine learning engineer, developing models for maritime data analysis. Developed computer vision and natural language processing solution currently in use by clients.

Universal Knowledge Society is King's College London student based society. We aim to show how cross-disciplinary thinking is crucial in innovation in today’s world. We will establish common grounds between quantitative and qualitative fields through conferences and problem-solving workshops. We want to encourage students to think with more universality. We organize conferences to show how this has already been exploited before: how mathematics has revolutionized music, how anthropological studies are having a huge impact on economic theories; how artificial intelligence is shaping the political world of tomorrow... The goal of these conferences would be to “raise awareness” on the fact that tomorrow’s world will be shaped by cross-disciplinary ideas. We hence want to motivate one to dig deeper into other disciplines as well as his/her own, in order to establish inter-disciplinary connections, thus increasing one’s innovative potential. In addition, we organize workshops whereby students would directly experience this cross-disciplinary approach and its benefits. This would consist of problem-solving activities with various types of problems (e.g. societal, engineering, political problems) where students from different disciplines would collaborate. This will allow students from different backgrounds to work together to achieve innovative solutions. The goal of these workshops is for students to concretely experience and witness the interest of combining several types of knowledge within one single framework.

The purpose of this project is to perform feasibility analysis and design quantitative T1 and T2 MR imaging strategy specially tailored for the fetal brain. The major challenge for fetal MRI is unpredictable fetal motion combined with maternal breathing. State-of-the-art quantitative MRI techniques, however, rely on multiple lengthy 3D acquisitions that are incompatible with constant fetal motion. This internship consisted in developing slice-selective (2D) quantitative MRI sequences, which are resistant to motion and specifically tuned to be sensitive to typical T1 and T2 values of the fetal brain.

Research intern. Worked on Neural networks (AI) and Zernike Decomposition to solve for tissue stiffness on images. Used C++ and ROOT (CERN).