Computational Imaging Research Lab | Projects

AutoPiX has started in Nov 2024 and will connect imaging and AI to improve the care of patients with arthritis. It will develop novel technologiy in automated image analysis, therapy monitoring and prediciton, and the creation of foundation models to boost science and care in the area of arthritis. 

ARTEMIS has started in January 2024 and will develop novel machine learning and simulation approaches to accelerate the translation of virtual twins towards a personalised management of fatty liver patients.

ONSET (Start 2021) is developing machine learning techniques for the detection of novel emerging diseases and the training of prediction models from early scarse observational data during a pandemic outbreak. It is part of a a collaboration with IAEA/ZODIAC, and funded by the FWF.

The AI-POD project has started in May 2023 and is developing AI tools for the prediction of risk of cardiovascular diseases in obese persons. It links imaging with continual activity tracking to improve risk scores. 

PREDICTOME (Start 2021) develops machine learning models for the prediction of response to neoadjuvant chemotherapy in breast cancer. It analyses the dynamics of early treatment response in imaging data and epigenomics profiles to understand their relationship and predictive value. It is funded by the WWTF.

RIFTAIR is a registry study to collect imaging data and relevant clinical data of patients with infectious lung diseases who have undergone CT examinations. Data collection includes data from before the ongoing COVID-19 pandemic, and continues as a prospective study.

MALBACS (Start 2022) aims at enhancing MRI breast cancer (BC) screening for women a high risk with help of machine learning. We will develop novel approaches for personalized screening, early detection, and a reduction of false positives leading to unnecessary biopsies. It is funded by the CCC.

EUCAIM (Start 2023) develops a federated European infrastructure for cancer imaging data funded under the DIGITAL programme. The project starts with 21 clinical sites from 12 countries and aims to have at least 30 distributed data providers from 15 countries by the end of the project. 

TRABIT - funded by European Union's Horizon 2020 program is an interdisciplinary and intersectoral joint effort of computational scientists, clinicians, and the industry in the field of neuroimaging. Its aim is to train a new generation of innovative and entrepreneurial researchers to bring quantitative image computing methods into the clinic.

AAMIR - Active Appearance Models in Musculo Skeletal Radiology, funded by the Austrian Science Fund FWF, (2004-2007). The accurate quantification of the progression of rheumatoid arthritis is a decisive factor during its treatment. Until now mainly manual quantification procedures are utilized. They are time consuming and lack reproducibility as well as accuracy. In a project a fully automated method for the assessment of RA is developed.

AORTAMOTION - Computerized Motion Analysis of the Aortic Arch to Assess the Influence of Pathologies, Supraaortic Transposition, and Stent-Graft Placement on its Deformation Patterns During the Cardiac Cycle,funded by Austrian National Bank Anniversary Fund, (2009 - 2012) The aim of this project is the development and application of a computerized system to assess the cardiac cycle dependent deformation dynamics of the thoracic aorta with ECG-gated computed tomography angiography. We are developing methods to extract deformation models, and to quantify the motion characteristics, that are crucial for an understanding of the effects of vessel transposition, and stenting.

BIGMEDILYTICS - funded by European Union's Horizon 2020 research and innovation program, coordinated by PHILIPS ELECTRONICS NEDERLAND, BigMedilytics will transform Europe’s Healthcare sector by using state-of-the-art Big Data technologies to achieve breakthrough productivity in the sector by reducing cost, improving patient outcomes and delivering better access to healthcare facilities simultaneously, covering the entire Healthcare Continuum – from Prevention to Diagnosis, Treatment and Home Care throughout Europe.

BONEMATCH - Bone Multi-modal Automated Trabecular Histomorphometry, funded by the EU 7th Framework through a Marie Curie Intra-European Fellowship, (2012 -) The aim of this project is to move closer to in-situ observation of bone remodelling and microarchitecture. This will be achieved by developing methods for detection of osteoclastic sites and eroded trabecular surfaces in micro-CT, and ultimately HR-pQCT images, using information obtained from bone histology. This project will build on the progress and results of the FELUX/BIOBONE and XBONE projects in order to develop a platform of bone imaging techniques aimed at the realization of image-based bone biopsy.

COBAQUO - Computer Based Quantification of Osteoporosis and Bone Alignment , funded by the Austrian National Bank Anniversary Fund, (2008 - 2010). Osteoporosis is an increasing problem for the Western world´s aging population. In Europe, 40% of women and 20% of men are, by the age of 80, expected to have suffered from an osteoporotic spine. The goals of the project are to develop and evaluate fully automated methods to assess the extent and progression of osteoporotic spinal fractures and the dimension of axial mal-alignment of the lower limb in radiographs.

DACHMM - Whole Body Image Analysis for Diagnosing Patients with Monoclonal Plasma Cell Disorders funded by the Deutsche ForschungsGesellschaft (DFG) and Austrian Science Fund (FWF). (2016-) The project aims at modelling infiltration patterns of Multiple Myeloma blood cancer longitudinaly over progression time, and at automatically measuring disease progression and treatment response based on multi-modal (CT, MR) imaging data.

FABRIC - exploring the Formation and Adaptation of the BRaIn Connectome funded by the European Union FP7 - Marie Curie Intra European Fellowships. (2013-2015) The project utilizes in utero fMRI and DTI to explore the emergence of large-scale anatomical pathways that presumably render information exchange channels for the forming of functional activity. Such observations will eventually be integrated into a temporal atlas of the fetal connectome development.

This project aims at developing new preprocessing strategies for in-utero functional MRI with the focus on new high resolution reconstruction and motion correction techniques, quality assessment of preprocessing pipelines and longitudinal analysis of developing functional connectivity in fetuses.

FETALMORPHO - Quantitative Morphometry of Fetal Brain Development for Disease Modeling and Diagnosis, funded by Austrian National Bank Anniversary Fund, (2012 -) We are building a spatio-temporal model - an atlas - of healthy fetal brain development from in utero magnetic resonance imaging (MRI) data from a large group of individuals, and study specific developmental diseases with help of this model.

FETLAS - Assessing Fetal Brain Development Based on a Spatio-Temporal in vivo Atlas Learned from Ultra-Fast Magnetic Resonance Images, funded by a DOC-fFORTE-fellowship of the Austrian Academy of Sciences, (2010 - 2012). This project's goal is to establish a spatio-temporal atlas for the developing human brain during a specific time period. In this context, we will develop methods that synchronously build the atlas as quantitative reference for physicians, as well as performing an automated segmentation of cortical and sub-cortical cerebral structures in fetal Magnetic Resonance Images.

FELUX/BIOBONE - Longitudinal clinical evaluation of bone architecture and biomechanical changes in transplantation osteoporosis, funded by Austrian National Bank Anniversary Fund, (2009 - ). This project aims to establish longitudinal FE modelling and advanced texture-based image analysis as novel, non-invasive monitoring tools for the in vivo-assessment of changes in bone microarchitecture due to osteoporosis.

KHRESMOI - Medical Information Analysis and Retrieval. funded by the European Union FP7. A multi-lingual, multi-modal search and access system for biomedical information and documents. The system will allow access to biomedical data: from many sources, analyzing and indexing multi-dimensional (2D, 3D, 4D) medical images.

OPTIMA - Christian Doppler Laboratory on Ophtalmic Image Analysis. funded by the Christian Doppler Gesellschaft. The OPTIMA project aims at individualizing patient management and at lowering treatment and monitoring needs in ophtalmic diseases to make the most effective ocular treatment available to all patients and physicians. We are conducting methodological research on big spatio-temporal data analysis.

PULMARCH - Computerized 3D Pulmonary Architecture Analysis, funded by the Austrian Science Fund FWF, (2010 - 2014) The goal of this project is to create a computer-aided diagnosis (CAD) system to differentiate between the textural expression of usual interstitial pneumonia (UIP, the histopathological counterpart of IPF) and NSIP, while providing increased sensitivity/specificity over today's differential diagnosis.

Information regarding the project PREDICT will be provided soon.
Team: Matthias Perkonigg, Daniel Sobotka, Georg Langs

VISCERAL - Visual Concept Extraction Challenge in Radiology funded by the European Union FP7. VISCERAL will organize two competitions on information extraction and retrieval involving medical image data and associated text that will benchmark the state of the art and define the next big challenges in large scale data processing in medical image analysis.

XBONE - Assessment of bone micro structure in the diagnosis of osteoporosis, funded by PHILIPS, (2008 - 2012). In this project we are focusing on the effect of osteoporis on the bone micro structure. In particular we study changes beyond the bone mineral density, and their depiction in high resolution computed tomography data.