A new PHOREVER publication presents promising results on supporting early detection of pancreatic cancer using extracellular vesicles (EVs) combined with adaptive machine learning techniques. This work has been accepted at EMBC 2025.
The study analyzed EV characteristics together with clinical and laboratory data from patients with Pancreatic Ductal Adenocarcinoma (PDAC) and non-cancer samples. The resulting model achieved:
-
Accuracy: 0.96
-
Sensitivity: 0.95
-
Specificity: 0.97
These results suggest that EV-based, minimally invasive approaches supported by machine learning could assist in earlier pancreatic cancer diagnosis.
This publication is authored by K. Kourou, A. Iosif, I. Angelioudaki, V. Kigka, A. G. Tzingounis, V. Pantazatou, M. Vlachou, N. Memos, A. Kataki, M. M. Konstadoulakis and I. D. Fotiadis.
Read the full publication here.