FAIrPaCT - Federated Artificial Intelligence Framework to Optimise the Treatment of Pancreatic Cancer
FAIrPaCT I / FAIrPaCT II
Funded by Federal Ministry of Research, Technology and Space
Duration: FAIrPaCT I: 11/2022 - 10/2024
FAIrPaCT II: 11/2024 - 10/2026
Aims of this collaborative project
FAIrPaCT I
The goal of this consortium, consisting of the University Medical Center Göttingen, the University Hospital Giessen and Marburg and the Technical University Munich, is to develop a software
system supported by federated artificial intelligence called FAIrPaCT that will enable the analysis of clinical patient data and molecular cancer cell data from patients with pancreatic cancer across
institutes. Pancreatic cancer is a highly aggressive malignancy with a rising incidence, predicted to become the second leading cause of cancer-related death by 2030 in the industrialised world. Due
to its extraordinarily aggressive, locally invasive tumour biology with a tendency to distant metastases, and the exceptionally high and heterogeneous resistance to conventional
chemotherapy, therapy is often difficult.
Our project combines three of the largest patient cohorts (KFO5002, KFO325, SFB1321) on pancreatic cancer in Germany, which are unique in size and heterogeneity. In combination with innovatively tailored federated artificial intelligence methods, we are able to train robust high-performance models to estimate the probability of success for specific treatment approaches. Moreover, the FAIrPaCT framework will enable the identification of important parameters that drive treatment response, so called markers. These can provide key details about the molecular mechanisms that influence therapy success and thus can support the development of improved drugs and enable personalised treatment strategies. To address the needs of stakeholders such as medical and computational researchers, and patient communities all software packages will be open-source, data will be FAIR, and results will be published. Finally, moving towards cross-cohort analysis enables us to benefit from heterogeneous local data, to build more robust and clinically relevant models, identify globally relevant markers and ultimately make a step further towards artificial intelligence-supported precision medicine.
For more information, please see here.
FAIrPaCT II
Federated Multimodal Artificial Intelligence Framework for Pancreatic Cancer outcome Prediction
FAIrPaCT II builts upon the FAIrPaCT I consortium (BMBF funded 2022-2024), comprising the University Medical Center Göttingen, the University Hospital Giessen and Marburg, and the Technical University Munich. Our approach combines three of the largest patient cohorts (KFO5002, KFO325, SFB1321) on pancreatic cancer in Germany. The initial consortium aimed to create a software system, supported by federated machine learning, facilitating the analysis of clinical information and genetic cancer panel sequencing data from pancreatic cancer patients. Pancreatic cancer, a highly aggressive malignancy, is anticipated to be the second leading cause of cancer-related death by 2030 in the industrialized world. Leveraging state-of-the-art federated artificial intelligence methods, we successfully trained robust high-performance models to estimate the likelihood of success of treatment. In this extension project, our primary objective is to leverage the existing consortium and infrastructure, enhancing our models to incorporate multi-omics data and histopathology images through a multimodal learning approach. Additionally, we aim to broaden our patient cohort to include individuals from the Molecular Tumor Boards at diverse locations. This expansion of our FAIrPaCT framework enables the identification of crucial genetic, as well as molecular and histopathological parameters, facilitating the prediction of treatment responses. These insights can offer pivotal information on therapy success, aiding the development of enhanced drugs and personalized treatment strategies. To disseminate our findings to the scientific community, including medical researchers, biomedical informaticians, clinical practitioners, and patient communities, all software packages will be released as open-source. Further, we will establish an interactive research platform, fostering the sharing and utilization of the resulting models.
For more information, please see here.
Contribution of the Department of Medical Bioinformatics
In FAIrPaCT I our department coordinates the
Workpackage 6: Interpretation and xAI for PDAC Models
This comprises extraction of clinically relevant features from the trained in the FAI models (WP5), statistical evaluation and interpretationof those features.
We will apply and extend in-house developed methods to interpret machine learning models especially in the context of molecular networks. These approaches will
elucidate potential causes and molecular mechanisms that drive the predictive performance of classifiers in PDAC.
In FAIrPaCT II we coordinate the Workpackages 3:
Multimodal Domain-Adaptation and xAI for PDAC Models
WP3 will develop a multimodal domain adaptation approach to enhance our understanding and model accuracy, involving separate modality models, pairwise and triple modal domain adaptation. The adapted multimodal model in a shared domain provides insights into PDAC tumor characteristics, subtype classification, and treatment response. Cross-modality knowledge transfer improves predictions across diverse data, resulting in a robust model. Deliverables include adapted multi-domain adaptation methods, refined multimodal models, cross-modality explanations, and integrated xAI insights based on SHAP and LIME.
Project partners
Location University of Göttingen/ University Medical Center Göttingen
Department of Medicals Informatics (UMG)
Prof. Dr. Anne-Christin Hausschild (Coordinator, PI WP1, WP2)
Prof. Dr. Uli Sax (PI WP3, WP7)
Department of Medical Bioinformatics (UMG)
Prof. Tim Beißbarth (PI, WP6)
Dr. Gregory Chereda (FAIrPaCT I)
Martin Bernaus (PhD) (FAIrPaCT II)
Clinic for General, Visceral and Paediatric Surgery
Clinic for Gastroenterology, Gastrointestinal Oncology and Endocrinology
Prof. Dr. Elisabeth Hessmann (PI, WP3)
Prof. Dr. Günther Schneider
Location Technical University of Munich
Medical Clinic and Polyclinic
Prof. Dr. Maximilian Reichert (PI)
Location Philipps University Marburg
Department of Internal Medicine, Gastroenterology, Endocrinology, Metabolism and Infectiology
Prof. Dr. Matthias Lauth (PI)
That might also interest you