About me

Hi, I'm Alice Schiavone! I am a PhD student in Computer Science at the University of Copenhagen.

My research focuses on the intersection of machine learning and medicine.

Right now, I'm looking into representation learning and distribution alignment via optimal transport. I've worked with medical imaging analysis and natural language processing.

Find me on GitHub or LinkedIn.

I made this website from scratch in 2021, using html, css and simple javascript. In October 2026, I started fixing it and adding functionalities with AI.

MOSAIC: Efficient Multilingual Radiology Report Classification

Open-source framework for analyzing radiology reports across multiple languages with few annotations on local hardware.

Presented at ClinicalNLP at LREC2026.

Keywords: Multilingual NLP, Radiology Reports, Efficient Language Models

Figure 1 from the MOSAIC paper: per-pathology F1 scores across datasets and models

Effective Machine Learning Techniques for Non-English Radiology Report Classification: A Danish Case Study

Challenging classification task on non-English language with a complex annotation protocol.

Results published at the journal AI.

Keywords: Language modeling for Healthcare, Radiology Reports

Deep Learning Methods for the Segmentation of White Matter Hyperintesities

MSs thesis project. Automatic segmentation method for medical imaging analysis. Segmentation of age-related white matter hyperintesities as a biomarker of cognitive decline and mortality.

Results published and presented at MIDL2023.

Keywords: Deep Learning, Medical Imaging Analysis