iToBoS

iToBoS

iToBoS: Intelligent Total Body Scanner

Duration: April 2021 - March 2025

Project summary:

Skin cancer is the most common human malignancy and its incidence has been increasing in the last decade. Within the general category of skin cancers, melanoma constitutes the main cause of death. Fortunately, melanoma may be cured if treated at an early stage.

However, early-stage melanoma detection is not easy even for expert dermatologists, and due to the labor, time and cost intensive state-of-the-art process of manually screening each patient using hand held dermoscopes, only a small fraction of the skin area can be covered with reasonable expenditure of resources. The risk of missing a melanoma therefore remains significant.

The iToBoS research project, funded by European Union's Horizon 2020 reasearch programme, will develop a novel diagnostic tool in for of an AI-driven total body scanner combining different sources of patient data, to significantly speed up the melanoma screening process, and minimize the associated healthcare costs, while providing transparent feedback to expert clinitians about the AI's diagnoses. The combination of all this personalized information will result in an accurate, detailed and structured assessment of the pigmented skin lesions of the patient.

Project partners:

Universitat de Girona, Optotune AG, IBM Israel - Science and Technology Ltd., Robert Bosch España SA, BARCO NV, National Technical University of Athens, Leibniz Universität Hannover,  Fundació Clínic per a la Recerca Biomèdica, RICOH Spain IT services SLU,  Trilateral Research Limited, Università degli Studi di Trieste, Coronis Computing SL, Torus Actions SAS, V7 Ltd, Isahit, University of Queensland, Magyar Tudományos Akadémia Számít. és Automatizálási Kutatóintézet,
Melanoma Patient Network Europe

Publications:

C. J. Anders, L. Weber, D. Neumannc, W. Samek, K.-R. Müller, S. Lapuschkin. Finding and removing Clever Hans: Using explanation methods to debug and improve deep models. Information Fusion. 2022.

J. Sun, S. Lapuschkin, W. Samek, A. Binder. Explain and improve: LRP-inference fine-tuning for image captioning models. Information Fusion. 2022.