Artificial intelligence Supporting MElanoma patients (ASME)

Project leader - main PI:

• Panagiotis Papapetrou, Professor, DSV, SU

Co-PI

• Maria Bampa, PhD student, DSV, SU

Researchers

• Franco Rugolon, PhD student, DSV, SU

Project period: 2022-06-01 to 2022-12-31

Funding: Marcus and Amalia Wallenberg Foundation

Description

Melanoma is the most dangerous type of skin cancer as it spreads to other organs more rapidly if not treated at an early stage. Treatment is typically surgical, with only a small group of metatastatic patients receiving chemotherapy and/or interferon therapy. About 80% of patients survive melanoma but remain at risk for disease progression for many years, for which there is no successful therapy, making melanoma a chronic, life-threatening and QoL-affecting disease and melanoma patients a patient group most likely to benefit from the AI-enhanced PROMs-based monitoring of ASME.

The goal of ASME is to extend the existing ASCAPE datasets with an ASCAPE-ready melanoma patient dataset to enable both standard ASCAPE and ASME-proposed AI/ML algorithms (for model training, predictions, analytics and simulations) on a type of cancer not originally supported by ASCAPE.

The main objectives of ASME are:

1. Study melanoma patients’ QoL using a set of melanoma patients, containing retrospective dataset of circa 250 patients aiming to uncover interesting hitherto unrecognised patterns
2. Apply ML techniques altogether, these observations indicate that melanoma may have a considerable impact on patients’ lives, including their health-related quality of life (HRQOL)
3. identify factors of metastasis (relapse) and factors of adverse effects of chemotherapy.

ASME’s work will center around Stockholm University’s retrospective dataset containing data of approximately 250 melanoma patients. Its aims will be to:

1. Uncover interesting patterns in the data
2. Extend ASCAPE so that it handles complex and heterogeneous time-evolving medical data sources
3. Demonstrate the applicability of AL/ML algorithms on melanoma data

People

Panagiotis Papapetrou, Professor

sequential and temporal data mining, explainability, healthcare applications

Maria Bampa

learning from complex medical data sources

Franco Rugolon

explainable machine learning for healthcare