Groupleader
Main research areas
- 'Rapid Recognition and Reconstruction of Cardiac Planes Using Artificial Intelligence from 3D MRI Scout Images of Heart' in cooperation with the St. Anne's University Hospital (FNUSA) in Brno and the General University Hospital (VFN) in Prague
- 'Effect of Thrombus Permeability and Porosity on Success of Mechanical Recanalization During Ischemic Brain Stroke' in cooperation with the St. Anne's University Hospital (FNUSA) in Brno
- 'Detection of Osteolytic Lesions by Low-Dose Dual Enegry CT in Myeloma Patients with Usage of Deep Neural Networks' in cooperation with the University Hospital Brno (FN Brno)
- 'Analysis of 3D CT Medical Image Data' – long-term cooperating with the international company Philips HealthCare (since 2008) within the framework of a scientific cooperation agreement
Main practical research results
- advanced method of 3D CT angiography of the lower limbs
- fully automatic software for segmentation of deformed and pathological spines including spinal canal detection and vertebral type identification
- fully automatic computer-aided diagnostic tool for the detection, segmentation and classification of bone metastases and tumours in the spine
- fully automatic software tool for alignment of the brain to specific radiological views in head CT scans
- a software tool for the detection and classification of cerebral haemorrhages in head CT scans
- publicly available software tool for automatic alignment of CT image data (whole-body and partial scans) to the LPS (Left-Posterior-Superior) coordinate system
Main research publications
- Deep-MyoSeg: Deep learning-based approach for myocardium segmentation in clinical T1-MOLLI and T2-bSSFP maps
- Combining Deep Learning and Radiomics for Automated, Objective, Comprehensive Bone Marrow Characterization From Whole-Body MRI: A Multicentric Feasibility Study
- VerSe: A Vertebrae Labelling and Segmentation Benchmark for Multi-detector CT Images
- Learning–based vertebra localization and labeling in 3D CT data of possibly incomplete and pathological spine
- Deep convolutional neural network-based segmentation and classification of difficult to define metastatic spinal lesions in 3D CT data
Projects/Industrial contracts
Analysis of specific types of medical image data
Project beginning
2008
Project leader
3D and 4D CT image data registration - research collaboration agreement
Project beginning
2008
Project leader
