Leveraging knowledge for explainable AI in personalized cancer treatment: challenges and future directions

Emilia Daghir-Wojtkowiak; Javier Alfaro; Michele Mastromattei; Aleksander Palkowski; Mark Stares; Ana Roca-Umbert; Andraz Krajnc; Riccardo Leoni; Anne Boland; Aria Nourbaksh; Ashwin Kallor; Camille Ducki; Davide Venditti; Carla Montesano; Chiara Cipriani; Daniel Faria; Delphine Pflieger; Elisa Zago; Etienne Bardet; Filipa Serrano; Florian Jeanneret; Damien Alouges; Liangwei Yin; Elodine Coquelet; Apolline Bacquet; Francesco Bonchi; Francesco Maiorino; Francesco Torino; Georges Bedran; Jean-Alexandre Long; Laura Balbi; Laurent Guyon; Liana Bevilacqua; Manuel Fiorelli; Marie-Catherine Wagner; Mario Reyes; Mario Roselli; Marta Contreiras Silva; Michal Waleron; Nikolas Dovrolis; Odile Filhol-Cochet; In Hwa Um; Georg Wolflein; Patrícia Eugénio; Pauline Bazelle; Pavlos Golnas; Peter Thorpe; Pierluigi Bove; Piyush Borole; Roberta Bernardini; Rohit Kumar; Rosella Cicconi; Saskia Kaltenbrunner; Saverio Gravina; Simona Brezar; Stefan Symeonides; Steven McGinn; Susana Nunes; Ted Hupp; Yuri Gordienko; Dimitrios Varvaras; Sergii Stirenko; Luciano Xumerle; Stefania Mariani; Assilah Bouzit; Stéphane Gazut; Heiko Poth; Kyriakos Souliotis; Hector Katifelis; Elena Verzoni; Giuseppe Procopio; Sarah Schoch; Francisco Lupiáñez-Villanueva; Sandra Türk; Katarzyna Barud; Dimitri Koroliouk; Juan Caubet; Yamir Moreno; Jean-Luc Descotes; Christina Golna; Valentina Guadalupi; Paolo Garagnani; Maria Gazouli; Jean-François Deleuze; Frans Folkvord; Nikolaus Forgó; David J. Harrison; Håkan Axelson; Armando Stellato; Maurizio Mattei; Ajitha Rajan; Alexander Laird; Christophe Battail; Catia Pesquita; Fabio Massimo Zanzotto

doi:10.3389/fdgth.2025.1637195

Leveraging knowledge for explainable AI in personalized cancer treatment: challenges and future directions

Emilia Daghir-Wojtkowiak, Javier Alfaro, Michele Mastromattei, Aleksander Palkowski, Mark Stares, Ana Roca-Umbert, Andraz Krajnc, Riccardo Leoni, Anne Boland, Aria Nourbaksh, Ashwin Kallor, Camille Ducki, Davide Venditti, Carla Montesano, Chiara Cipriani, Daniel Faria, Delphine Pflieger, Elisa Zago, Etienne Bardet, Filipa SerranoFlorian Jeanneret, Damien Alouges, Liangwei Yin, Elodine Coquelet, Apolline Bacquet, Francesco Bonchi, Francesco Maiorino, Francesco Torino, Georges Bedran, Jean-Alexandre Long, Laura Balbi, Laurent Guyon, Liana Bevilacqua, Manuel Fiorelli, Marie-Catherine Wagner, Mario Reyes, Mario Roselli, Marta Contreiras Silva, Michal Waleron, Nikolas Dovrolis, Odile Filhol-Cochet, In Hwa Um, Georg Wolflein, Patrícia Eugénio, Pauline Bazelle, Pavlos Golnas, Peter Thorpe, Pierluigi Bove, Piyush Borole, Roberta Bernardini, Rohit Kumar, Rosella Cicconi, Saskia Kaltenbrunner, Saverio Gravina, Simona Brezar, Stefan Symeonides, Steven McGinn, Susana Nunes, Ted Hupp, Yuri Gordienko, Dimitrios Varvaras, Sergii Stirenko, Luciano Xumerle, Stefania Mariani, Assilah Bouzit, Stéphane Gazut, Heiko Poth, Kyriakos Souliotis, Hector Katifelis, Elena Verzoni, Giuseppe Procopio, Sarah Schoch, Francisco Lupiáñez-Villanueva, Sandra Türk, Katarzyna Barud, Dimitri Koroliouk, Juan Caubet, Yamir Moreno, Jean-Luc Descotes, Christina Golna, Valentina Guadalupi, Paolo Garagnani, Maria Gazouli, Jean-François Deleuze, Frans Folkvord, Nikolaus Forgó, David J. Harrison, Håkan Axelson, Armando Stellato, Maurizio Mattei, Ajitha Rajan, Alexander Laird, Christophe Battail, Catia Pesquita, Fabio Massimo Zanzotto^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

Abstract

Integrating multi-modal patient data to support personalized medicine has gained a lot of interest across different health domains over the past decade. Addressing this challenge requires the development and implementation of an informed, evidence-based AI-driven decision-support system continuously maintained and updated to align with the latest clinical guidelines. A key challenge to ensure its real-life adoption lies in translating the outcomes of complex AI-driven data integration and modeling into a form easily understood by the clinical audience. To ensure explainability, knowledge graphs have emerged as data models integrating multi-omics data sources and representing them as interconnected networks. Knowledge graphs offer a framework which AI models can progressively refine, highlighting the most influential features and relationships facilitating transparency of complex interactions and interdependencies. In this perspective we present major components and challenges upon developing a knowledge-based explainable AI system. Additionally, we showcase a current effort undertaken by the Knowledge at the Tips of your Fingers (KATY) consortium to develop the infrastructure for an explainable system supporting best treatment decision for a renal cancer patient.

Original language	English
Article number	104267
Pages (from-to)	1-10
Number of pages	10
Journal	Frontiers in Digital Health
Volume	7
DOIs	https://doi.org/10.3389/fdgth.2025.1637195
Publication status	Published - 29 Sept 2025

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Personalized cancer treatment
Knowledge graphs
Explainability
AI
Foundation models
Clinical decision-making

Access to Document

10.3389/fdgth.2025.1637195Licence: CC BY

Daghir-Wojtkowiak-et-tal-2025-Leveraging-knowledge-FDTDH-7-1637195-CCBY
Copyright © 2025 Daghir-Wojtkowiak, Alfaro, Mastromattei, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Final published version, 630 KBLicence: CC BY

Handle.net

Persistent link

Cite this

Daghir-Wojtkowiak, E., Alfaro, J., Mastromattei, M., Palkowski, A., Stares, M., Roca-Umbert, A., Krajnc, A., Leoni, R., Boland, A., Nourbaksh, A., Kallor, A., Ducki, C., Venditti, D., Montesano, C., Cipriani, C., Faria, D., Pflieger, D., Zago, E., Bardet, E., ... Zanzotto, F. M. (2025). Leveraging knowledge for explainable AI in personalized cancer treatment: challenges and future directions. Frontiers in Digital Health, 7, 1-10. Article 104267. https://doi.org/10.3389/fdgth.2025.1637195

@article{eb25edbfa96b412fa7ba76787794dae5,

title = "Leveraging knowledge for explainable AI in personalized cancer treatment: challenges and future directions",

abstract = "Integrating multi-modal patient data to support personalized medicine has gained a lot of interest across different health domains over the past decade. Addressing this challenge requires the development and implementation of an informed, evidence-based AI-driven decision-support system continuously maintained and updated to align with the latest clinical guidelines. A key challenge to ensure its real-life adoption lies in translating the outcomes of complex AI-driven data integration and modeling into a form easily understood by the clinical audience. To ensure explainability, knowledge graphs have emerged as data models integrating multi-omics data sources and representing them as interconnected networks. Knowledge graphs offer a framework which AI models can progressively refine, highlighting the most influential features and relationships facilitating transparency of complex interactions and interdependencies. In this perspective we present major components and challenges upon developing a knowledge-based explainable AI system. Additionally, we showcase a current effort undertaken by the Knowledge at the Tips of your Fingers (KATY) consortium to develop the infrastructure for an explainable system supporting best treatment decision for a renal cancer patient.",

keywords = "Personalized cancer treatment, Knowledge graphs, Explainability, AI, Foundation models, Clinical decision-making",

author = "Emilia Daghir-Wojtkowiak and Javier Alfaro and Michele Mastromattei and Aleksander Palkowski and Mark Stares and Ana Roca-Umbert and Andraz Krajnc and Riccardo Leoni and Anne Boland and Aria Nourbaksh and Ashwin Kallor and Camille Ducki and Davide Venditti and Carla Montesano and Chiara Cipriani and Daniel Faria and Delphine Pflieger and Elisa Zago and Etienne Bardet and Filipa Serrano and Florian Jeanneret and Damien Alouges and Liangwei Yin and Elodine Coquelet and Apolline Bacquet and Francesco Bonchi and Francesco Maiorino and Francesco Torino and Georges Bedran and Jean-Alexandre Long and Laura Balbi and Laurent Guyon and Liana Bevilacqua and Manuel Fiorelli and Marie-Catherine Wagner and Mario Reyes and Mario Roselli and Silva, \{Marta Contreiras\} and Michal Waleron and Nikolas Dovrolis and Odile Filhol-Cochet and Um, \{In Hwa\} and Georg Wolflein and Patr{\'i}cia Eug{\'e}nio and Pauline Bazelle and Pavlos Golnas and Peter Thorpe and Pierluigi Bove and Piyush Borole and Roberta Bernardini and Rohit Kumar and Rosella Cicconi and Saskia Kaltenbrunner and Saverio Gravina and Simona Brezar and Stefan Symeonides and Steven McGinn and Susana Nunes and Ted Hupp and Yuri Gordienko and Dimitrios Varvaras and Sergii Stirenko and Luciano Xumerle and Stefania Mariani and Assilah Bouzit and St{\'e}phane Gazut and Heiko Poth and Kyriakos Souliotis and Hector Katifelis and Elena Verzoni and Giuseppe Procopio and Sarah Schoch and Francisco Lupi{\'a}{\~n}ez-Villanueva and Sandra T{\"u}rk and Katarzyna Barud and Dimitri Koroliouk and Juan Caubet and Yamir Moreno and Jean-Luc Descotes and Christina Golna and Valentina Guadalupi and Paolo Garagnani and Maria Gazouli and Jean-Fran{\c c}ois Deleuze and Frans Folkvord and Nikolaus Forg{\'o} and Harrison, \{David J.\} and H{\aa}kan Axelson and Armando Stellato and Maurizio Mattei and Ajitha Rajan and Alexander Laird and Christophe Battail and Catia Pesquita and Zanzotto, \{Fabio Massimo\}",

note = "Funding: The author(s) declare that financial support was received for the research and/or publication of this article. This study was funded by the European Union programme Horizon 2020 under the call H2020-SCI-FA-DTS-2020-1 (contract number 101017453).",

year = "2025",

month = sep,

day = "29",

doi = "10.3389/fdgth.2025.1637195",

language = "English",

volume = "7",

pages = "1--10",

journal = "Frontiers in Digital Health",

issn = "2673-253X",

}

Daghir-Wojtkowiak, E, Alfaro, J, Mastromattei, M, Palkowski, A, Stares, M, Roca-Umbert, A, Krajnc, A, Leoni, R, Boland, A, Nourbaksh, A, Kallor, A, Ducki, C, Venditti, D, Montesano, C, Cipriani, C, Faria, D, Pflieger, D, Zago, E, Bardet, E, Serrano, F, Jeanneret, F, Alouges, D, Yin, L, Coquelet, E, Bacquet, A, Bonchi, F, Maiorino, F, Torino, F, Bedran, G, Long, J-A, Balbi, L, Guyon, L, Bevilacqua, L, Fiorelli, M, Wagner, M-C, Reyes, M, Roselli, M, Silva, MC, Waleron, M, Dovrolis, N, Filhol-Cochet, O, Um, IH, Wolflein, G, Eugénio, P, Bazelle, P, Golnas, P, Thorpe, P, Bove, P, Borole, P, Bernardini, R, Kumar, R, Cicconi, R, Kaltenbrunner, S, Gravina, S, Brezar, S, Symeonides, S, McGinn, S, Nunes, S, Hupp, T, Gordienko, Y, Varvaras, D, Stirenko, S, Xumerle, L, Mariani, S, Bouzit, A, Gazut, S, Poth, H, Souliotis, K, Katifelis, H, Verzoni, E, Procopio, G, Schoch, S, Lupiáñez-Villanueva, F, Türk, S, Barud, K, Koroliouk, D, Caubet, J, Moreno, Y, Descotes, J-L, Golna, C, Guadalupi, V, Garagnani, P, Gazouli, M, Deleuze, J-F, Folkvord, F, Forgó, N, Harrison, DJ, Axelson, H, Stellato, A, Mattei, M, Rajan, A, Laird, A, Battail, C, Pesquita, C & Zanzotto, FM 2025, 'Leveraging knowledge for explainable AI in personalized cancer treatment: challenges and future directions', Frontiers in Digital Health, vol. 7, 104267, pp. 1-10. https://doi.org/10.3389/fdgth.2025.1637195

TY - JOUR

T1 - Leveraging knowledge for explainable AI in personalized cancer treatment

T2 - challenges and future directions

AU - Daghir-Wojtkowiak, Emilia

AU - Alfaro, Javier

AU - Mastromattei, Michele

AU - Palkowski, Aleksander

AU - Stares, Mark

AU - Roca-Umbert, Ana

AU - Krajnc, Andraz

AU - Leoni, Riccardo

AU - Boland, Anne

AU - Nourbaksh, Aria

AU - Kallor, Ashwin

AU - Ducki, Camille

AU - Venditti, Davide

AU - Montesano, Carla

AU - Cipriani, Chiara

AU - Faria, Daniel

AU - Pflieger, Delphine

AU - Zago, Elisa

AU - Bardet, Etienne

AU - Serrano, Filipa

AU - Jeanneret, Florian

AU - Alouges, Damien

AU - Yin, Liangwei

AU - Coquelet, Elodine

AU - Bacquet, Apolline

AU - Bonchi, Francesco

AU - Maiorino, Francesco

AU - Torino, Francesco

AU - Bedran, Georges

AU - Long, Jean-Alexandre

AU - Balbi, Laura

AU - Guyon, Laurent

AU - Bevilacqua, Liana

AU - Fiorelli, Manuel

AU - Wagner, Marie-Catherine

AU - Reyes, Mario

AU - Roselli, Mario

AU - Silva, Marta Contreiras

AU - Waleron, Michal

AU - Dovrolis, Nikolas

AU - Filhol-Cochet, Odile

AU - Um, In Hwa

AU - Wolflein, Georg

AU - Eugénio, Patrícia

AU - Bazelle, Pauline

AU - Golnas, Pavlos

AU - Thorpe, Peter

AU - Bove, Pierluigi

AU - Borole, Piyush

AU - Bernardini, Roberta

AU - Kumar, Rohit

AU - Cicconi, Rosella

AU - Kaltenbrunner, Saskia

AU - Gravina, Saverio

AU - Brezar, Simona

AU - Symeonides, Stefan

AU - McGinn, Steven

AU - Nunes, Susana

AU - Hupp, Ted

AU - Gordienko, Yuri

AU - Varvaras, Dimitrios

AU - Stirenko, Sergii

AU - Xumerle, Luciano

AU - Mariani, Stefania

AU - Bouzit, Assilah

AU - Gazut, Stéphane

AU - Poth, Heiko

AU - Souliotis, Kyriakos

AU - Katifelis, Hector

AU - Verzoni, Elena

AU - Procopio, Giuseppe

AU - Schoch, Sarah

AU - Lupiáñez-Villanueva, Francisco

AU - Türk, Sandra

AU - Barud, Katarzyna

AU - Koroliouk, Dimitri

AU - Caubet, Juan

AU - Moreno, Yamir

AU - Descotes, Jean-Luc

AU - Golna, Christina

AU - Guadalupi, Valentina

AU - Garagnani, Paolo

AU - Gazouli, Maria

AU - Deleuze, Jean-François

AU - Folkvord, Frans

AU - Forgó, Nikolaus

AU - Harrison, David J.

AU - Axelson, Håkan

AU - Stellato, Armando

AU - Mattei, Maurizio

AU - Rajan, Ajitha

AU - Laird, Alexander

AU - Battail, Christophe

AU - Pesquita, Catia

AU - Zanzotto, Fabio Massimo

N1 - Funding: The author(s) declare that financial support was received for the research and/or publication of this article. This study was funded by the European Union programme Horizon 2020 under the call H2020-SCI-FA-DTS-2020-1 (contract number 101017453).

PY - 2025/9/29

Y1 - 2025/9/29

N2 - Integrating multi-modal patient data to support personalized medicine has gained a lot of interest across different health domains over the past decade. Addressing this challenge requires the development and implementation of an informed, evidence-based AI-driven decision-support system continuously maintained and updated to align with the latest clinical guidelines. A key challenge to ensure its real-life adoption lies in translating the outcomes of complex AI-driven data integration and modeling into a form easily understood by the clinical audience. To ensure explainability, knowledge graphs have emerged as data models integrating multi-omics data sources and representing them as interconnected networks. Knowledge graphs offer a framework which AI models can progressively refine, highlighting the most influential features and relationships facilitating transparency of complex interactions and interdependencies. In this perspective we present major components and challenges upon developing a knowledge-based explainable AI system. Additionally, we showcase a current effort undertaken by the Knowledge at the Tips of your Fingers (KATY) consortium to develop the infrastructure for an explainable system supporting best treatment decision for a renal cancer patient.

AB - Integrating multi-modal patient data to support personalized medicine has gained a lot of interest across different health domains over the past decade. Addressing this challenge requires the development and implementation of an informed, evidence-based AI-driven decision-support system continuously maintained and updated to align with the latest clinical guidelines. A key challenge to ensure its real-life adoption lies in translating the outcomes of complex AI-driven data integration and modeling into a form easily understood by the clinical audience. To ensure explainability, knowledge graphs have emerged as data models integrating multi-omics data sources and representing them as interconnected networks. Knowledge graphs offer a framework which AI models can progressively refine, highlighting the most influential features and relationships facilitating transparency of complex interactions and interdependencies. In this perspective we present major components and challenges upon developing a knowledge-based explainable AI system. Additionally, we showcase a current effort undertaken by the Knowledge at the Tips of your Fingers (KATY) consortium to develop the infrastructure for an explainable system supporting best treatment decision for a renal cancer patient.

KW - Personalized cancer treatment

KW - Knowledge graphs

KW - Explainability

KW - AI

KW - Foundation models

KW - Clinical decision-making

UR - https://www.scopus.com/pages/publications/105018812173

U2 - 10.3389/fdgth.2025.1637195

DO - 10.3389/fdgth.2025.1637195

M3 - Article

SN - 2673-253X

VL - 7

SP - 1

EP - 10

JO - Frontiers in Digital Health

JF - Frontiers in Digital Health

M1 - 104267

ER -

Leveraging knowledge for explainable AI in personalized cancer treatment: challenges and future directions

Abstract

UN SDGs

Keywords

Access to Document

Handle.net

Other files and links

Fingerprint

Cite this