Mathias Perslev

Mathias Perslev is a Guest Researcher at the Department of Computer Science (DIKU), University of Copenhagen, actively contributing to the Machine Learning Section and the SCIENCE AI Centre. His research focuses on machine learning applications in medical imaging, particularly in the segmentation and analysis of complex biomedical data using advanced deep learning techniques.

He completed his PhD in 2023 with a dissertation on "Segmentation of medical images and time series using fully convolutional neural networks", highlighting his expertise in developing novel computational methods for clinical imaging challenges.

His research interests include:

• Deep Learning for Medical Image Analysis
• Convolutional Neural Networks (CNNs) and UNet Architectures
• 3D Medical Imaging & Multi-planar Segmentation
• Neuroimaging & Stroke Lesion Detection
• Kidney, Liver, and Knee Tumor Segmentation
• Biomedical Data Augmentation & Benchmarking

His recent publications demonstrate a clear trajectory towards automated, high-precision medical image segmentation, with applications spanning oncology, neurology, and musculoskeletal disorders. His collaborative work includes participation in international challenges such as KiTS23 and LiTS, contributing to open benchmarks for medical AI.

While no specific awards are listed, his consistent publication record in high-impact journals and conferences underscores his significant contributions to the field.

Currently, he is affiliated with the Machine Learning Section at DIKU, leveraging the department's advanced compute resources and participating in interdisciplinary research initiatives within the SCIENCE AI Centre.