Short Bio

I am currently a post-doctoral fellow within the Data Science for Digital Health group at the University of Geneva. My research is dedicated to the development and application of advanced deep learning methodologies to address critical challenges in the medical field. Specifically, I focus on machine and deep learning techniques for predicting infectious disease outcomes in kidney transplant recipients. In addition, I investigate label-noise robust semi-supervised learning frameworks for improved data annotation and classification, as well as the characterization of transcranial Doppler ultrasound signals to support stroke prevention strategies.

Research Interests

Deep Learning, Machine Learning, Self-supervised learning, Solid Organ Transplantation, Infectious Diseases, Model Compression, Medical Imaging, Telecommunications.

Publications

Journals

Vindas, Y., Guépié, B.K., Almar, M., Roux, E., and Delachartre, P., 2024. Trainable pruned ternary quantization for medical signal classification models, in Neurocomputing.

Vindas, Y., Guépié, B.K., Almar, M., Roux, E., and Delachartre, P., 2024. An asymmetric heuristic for trained ternary quantization based on the statistics of the weights: an application to medical signal classification, in Pattern Recognition Letters

Vindas, Y., Roux, E., Guépié, B.K., Almar, M., and Delachartre, P., 2023. Guided deep embedded clustering regularization for multifeature medical signal classification, in Pattern Recognition.

Vindas, Y., Guépié, B.K., Almar, M., Roux, E., and Delachartre, P., 2022. Semi-automatic data annotation based on feature-space projection and local quality metrics: an application to cerebral emboli characterization, in Medical Image Analysis, page 102437, 2022. ISSN 1361-8415. doi: https://doi.org/10.1016/j.media.2022.102437.

Conferences with proceedings

Vindas, Y., Guillaud, M., 2025. Dynamic Channel Charting: Integrating Online Sample Selection with Continual Learning for Streaming CSI Data, in: 2025 IEEE International Conference on Communications (ICC).

Vindas, Y., Guillaud, M., 2025. Weakly-supervised Semantic Space Structuring: Cardiac Cycle Position for Cerebral Emboli Visualization using Contrastive Learning, in: 2025 IEEE International Symposium on Biomedical Imaging (ISBI).

Vindas, Y., Guillaud, M., 2024. Multi-Site Wireless Channel Charting Through Latent Space Alignment, in: 2024 IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC).

Vindas, Y., Guépié, B.K., Almar, M., Roux, E., Delachartre, P., 2023. Soft-labels noise tolerant loss functions for transcranial Doppler ultrasound signal classification, in: 2023 IEEE International Ultrasonics Symposium (IUS).

Vindas, Y., Roux, E., Guépié, B.K., Almar, M., Delachartre, P., 2023 Deep Embedded Clustering regularization for imbalanced cerebral emboli classification using transcranial Doppler ultrasound, in: 2023 European Signal Processing Conference (EUSIPCO)

Vindas, Y., Guépié, B.K., Almar, M., Roux, E., and Delachartre, P., 2022. An hybrid CNN-Transformer model based on multi-feature extraction and attention fusion mechanism for cerebral emboli classification, in: MLHC. 05–06 Aug 2022, PMLR.

Vindas, Y., Roux, E., Guépié, B.K., Almar, M., Delachartre, P., 2021. Semi-supervised annotation of transcranial Doppler ultrasound micro-embolic data, in: 2021 IEEE International Ultrasonics Symposium (IUS), pp. 1–4. doi:10.1109/IUS52206.2021.9593847.

Conferences without proceedings

Vindas, Y., Guépié, B.K., Almar, M., Roux, E., and Delachartre, P., 2023. Classification multi-représentation d'emboles cérébraux à partir d'un dispositif de Doppler transcrânien. in:2023 Intelligence Artificielle en Imagerie Biomédicale (IABM).

PhD Manuscript: Yamil Vindas Yassine. Weakly-supervised learning for emboli characterization with Transcranial Doppler (TCD) monitoring. Medical Imaging. INSA de Lyon, 2023.

Semi-automatic data annotation based on feature space projection and local quality metrics.

Multi-feature medical signal classification.

Imbalanced deep embedded clustering regularization.

Asymmetric trained ternary quantization for deep learning model compression.

Trainable pruned ternary quantization for deep learning model compression.

Presentations

PhD Defense, 2023.

Tecnología al servicio de la salud, Academia Nacional de Ciencias de Costa Rica, 2024.

Teaching

2020-2024: Mathematical and Software Tools 1, IUT Gratte Ciel Lyon 1 (first year students).

2020-2022: Algorithmics and Programming, INSA Lyon (second year students).

2022-2023: Data Science, Faculté Médecine Lyon Est (first year students).

2021 and 2022: Deep Learning for Medical Imaging School (hands-on sessions).

Representative of Ph.D students of CREATIS at the computer science federation of Lyon (FIL), from 2021 to 2023.

Contact

Campus Biotech G6-N3, Chemin des Mines 9, 1202 Genève.

Email: yamil(dot)vindasyassine(at)unige.ch