Embracing deep learning-based image analysis

MiaLab supports state-of-the-art deep learning models for medical image classification, segmentation, detection, and registration. It also offers a user-friendly GUI that allows clinical users to review and refine results. Together, these features enable a seamless workflow for a wide range of clinical research applications in medical imaging.

Read More

About MiaLab

MiaLab is a general-purpose image processing platform built on VTK, ITK, and ONNX libraries. It combines interactive image annotation tools with advanced visualization techniques and fully automated deep learning–based processing modules, offering state-of-the-art image analysis capabilities for academic use. The software is compatible with both Windows and Linux systems.

Read More

Our Team

MiaLab was originally developed by Dr. Chunliang Wang during his PhD project under the supervision of Prof. Örjan Smedby, and was later expanded to support a wide range of image processing tasks. With contributions from his colleagues Dr. Simone Bendazzoli and Max Kim, the software was further enhanced to incorporate state-of-the-art deep learning methods through the ONNX library.

Read More

Show Cases

MiaLab has been widely used in numerous research projects within the Medical Image Analysis Group at the Department of Biomedical Engineering and Health Systems, KTH Royal Institute of Technology. Below are example applications in cardiac imaging, pulmonary disease analysis, orthopedic visualization, and neural imaging analysis.

Coronary MRA analysis

LUng lobe segmentation

Pelvic bone segmenattion

Heart chamber segmentation

Liver tumor segmenation and surgery planning

Acknowledgements