Wesley De Neve received his M.Sc. degree in Computer Science from Ghent University in 2002, completing a thesis on functional metadata integration within Apple's QuickTime framework. He then pursued a Ph.D. degree at the Multimedia Lab of Ghent University and the iMinds research institute, earning his doctorate in Computer Science Engineering in 2007 with a dissertation on format-independent multimedia content adaptation. During his doctoral studies, he contributed to the MPEG-21 Digital Item Adaptation standard within the MPEG consortium.
Following his doctoral studies, Wesley De Neve was awarded a Brain Korea 21 (BK21) fellowship, which brought him to South Korea as a postdoctoral researcher at ICU and KAIST. There, he shifted focus from multimedia representation to multimedia content analysis. In 2011, he returned to Ghent University - iMinds as a postdoctoral fellow, while maintaining an adjunct appointment at KAIST. He established the Social and Visual Intelligence (SaVI) cluster, focusing on large-scale machine learning for social media and visual data.
Since 2014, Wesley De Neve has been a full-time professor at Ghent University Global Campus (GUGC) in South Korea, where he co-founded the Center for Biosystems and Biotech Data Science. He teaches Informatics and Bioinformatics and leads research in biotech data science. His work involves analyzing biomedical images and biological sequences, with applications including parasite detection, microplastic identification, tumor spread analysis, and genome annotation.
Wesley De Neve currently supervises eight Ph.D. students and has overseen seven Ph.D. completions. Since 2020, his team has published 14 SCI(E) journal papers and 17 conference papers on machine learning-based computer vision and sequence analysis.
He also serves as Secretary of the GUGC Campus Council and is a member of both the Director of Studies Team and the Executive Board at GUGC.
Domains of expertise
- Artificial intelligence
- Biomedical image processing
- Computer vision; Deep learning
- Machine learning and decision making
Selected projects & research platforms
SpliceRover
SpliceRover is a deep learning-based tool designed for the identification of splice sites in genomic DNA sequences. Splice sites are the boundaries between exons and introns in genes, which are crucial for RNA splicing during gene expression. Accurate detection of these sites is essential for gene structure annotation, understanding alternative splicing, and interpreting genomic variants.
http://bioit2.irc.ugent.be/rover/
PIPAC
Peritoneal metastasis (PM) occurs in the advanced stages of ovarian and gastrointestinal cancers. Patients with PM have a poor prognosis, and their quality of life is severely compromised. Pressurized intraperitoneal aerosol chemotherapy (PIPAC) is a promising treatment option, but treatment responses are difficult to predict. Currently, standard clinical, microscopic, and medical imaging modalities are limited in their ability to quantify PM and evaluate PIPAC responses.
The PIPAC interdisciplinary research project addresses both challenges by combining expertise in cancer oncology and artificial intelligence. It focuses on developing novel computer vision techniques, grounded in deep machine learning, to quantify PM and evaluate PIPAC responses in a reproducible manner. The adopted approach balances data requirements with computational complexity, model effectiveness, and robustness.
Tryp
Trypanosomiasis, a neglected tropical disease (NTD), poses a significant public health challenge in sub-Saharan Africa and Latin America. The World Health Organization (WHO) emphasizes the urgent need for practical, field-adaptable diagnostic methods and rapid screening tools to mitigate the impact of NTDs. Although artificial intelligence has demonstrated promise in disease screening, progress is hindered by the scarcity of high-quality, curated datasets.
To address this challenge, we developed the Tryp dataset, which is a comprehensive collection of microscopy images featuring unstained thick blood smears containing the Trypanosoma brucei brucei parasite. The Tryp dataset is the largest publicly available resource of its kind and is designed to support the development and benchmarking of AI-driven diagnostic tools for trypanosomiasis.
Selected publications
- Utku Özbulak, Baptist Vandersmissen, Azarakhsh Jalalvand, Ivo Couckuyt, Arnout Van Messem, and Wesley De Neve, “Investigating the significance of adversarial attacks and their relation to interpretability for radar-based human activity recognition systems,” Computer Vision and Image Understanding, vol. 202, no. 103111, 9 pages, 2021 [IF: 4.886 (2021) – peer reviewed]
- Mijung Kim, Jong Chul Han, Seung Hyup Hyun, Olivier Janssens, Sofie Van Hoecke, Changwon Kee, and Wesley De Neve. “Medinoid: computer-aided diagnosis and localization of glaucoma using deep learning,” Applied Sciences - Basel, vol. 9, no. 15, 19 pages, 2019. [IF: 2.474 (2019) – peer reviewed]
- Jasper Zuallaert, Frederic Godin, Mijung Kim, Arne Soete, Yvan Saeys, and Wesley De Neve, “SpliceRover: interpretable convolutional neural networks for improved splice site prediction,” Bioinformatics, vol. 34, no. 24, pp. 4180–4188, 2018. [IF: 4.531 (2018) – peer reviewed]
Work details
wesley.deneve@ghent.ac.kr
Ghent University Building, Incheon Global Campus,119-5 Songdomunhwa-ro, Yeonsu-gu, Incheon-si 21985, Republic of Korea
ORCID number: 0000-0002-8190-3839