Postdoctoral Position in Multimodal and Generative Artificial Intelligence for Biomedical Imaging and Genetics

The Artificial Intelligence for Good (AIGO) research unit, led by Vittorio Murino, develops machine-learning and deep-learning methods for complex, multimodal and imperfect data, with emphasis on biomedical and healthcare applications, learning under limited or biased supervision, domain shift, few-shot and continual learning, generative models and multimodal foundation models. This position studies multimodal and generative models and representation learning for medical imaging and heterogeneous biomedical data, aiming to learn image-derived representations that capture phenotypic information and predict underlying genotypic factors and genetic risk beyond association-based statistics.

The project, financed by Dompé Farmaceutici, concerns AI-driven prediction of glaucoma-linked genetic variants from OCT retinal imaging. Responsibilities include developing models that link imaging to latent biological or genotypic factors, supervising PhD students, publishing and supporting project proposals. The 12-month position in Genoa pays a gross annual salary of 32,000 to 40,000 euros.

PhD in artificial intelligence, machine learning, computer vision, computer science, engineering, physics, mathematics or related disciplines; documented machine/deep-learning and computer-vision experience, preferably in medical or biomedical imaging and multimodal learning; in-depth knowledge of generative models (GANs, diffusion, encoder-decoder, optimal transport); command of graph neural networks and transformers; excellent Python and PyTorch skills; strong publication record; excellent English. Applications include a detailed CV with publications and contact details of two references.