EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Rad4XCNN: A new agnostic method for post-hoc global explanation of CNN-derived features by means of Radiomics

Francesco Prinzi, Carmelo Militello, Calogero Zarcaro, Tommaso Vincenzo Bartolotta, Salvatore Gaglio, Salvatore Vitabile

Computer Methods and Programs in Biomedicine · 2025

DOI: 10.1016/j.cmpb.2024.108576 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Methods and Programs in Biomedicine, 2025. Verfasst von Francesco Prinzi und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedicine, Neuroscience and Advanced Diagnostics (BiND), University of Palermo, Palermo, 90127, Italy. Electronic address: francesco.…. Hauptthemen laut MEDLINE: Breast Neoplasms, Neural Networks, Computer. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 82 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Palermo.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND OBJECTIVE: In recent years, machine learning-based clinical decision support systems (CDSS) have played a key role in the analysis of several medical conditions. Despite their promising capabilities, the lack of transparency in AI models poses significant challenges, particularly in medical contexts where reliability is a mandatory aspect. However, it appears that explainability is inversely proportional to accuracy. For this reason, achieving transparency without compromising predictive accuracy remains a key challenge. METHODS: This paper presents a novel method, namely Rad4XCNN, to enhance the predictive power of CNN-derived features with the inherent interpretability of radiomic features. Rad4XCNN diverges from conventional methods based on saliency maps, by associating intelligible meaning to CNN-derived features by means of Radiomics, offering new perspectives on explanation methods beyond visualization maps. RESULTS: Using a breast cancer classification task as a case study, we evaluated Rad4XCNN on ultrasound imaging datasets, including an online dataset and two in-house datasets for internal and external validation. Some key results are: (i) CNN-derived features guarantee more robust accuracy when compared against ViT-derived and radiomic features; (ii) conventional visualization map methods for explanation present several pitfalls; (iii) Rad4XCNN does not sacrifice model accuracy for their explainability; (iv) Rad4XCNN provides a global explanation enabling the physician to extract global insights and findings. CONCLUSIONS: Our method can mitigate some concerns related to the explainability-accuracy trade-off. This study highlighted the importance of proposing new methods for model explanation without affecting their accuracy.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Breast NeoplasmsNeural Networks, ComputerHumansMachine LearningFemaleReproducibility of ResultsAlgorithmsUltrasonographyDecision Support Systems, ClinicalRadiomics

Stichwörter der Autor:innen

Breast cancerClinical Decision Support SystemsConvolutional neural networksExplainable AIRadiomics

Autor:innen und Institutionen

  • Department of Biomedicine, Neuroscience and Advanced Diagnostics (BiND), University of Palermo, Palermo, 90127, Italy. Electronic address: francesco.prinzi@unipa.it.
  • Institute for High-Performance Computing and Networking (ICAR-CNR), Italian National Research Council, Palermo, 90146, Italy. Electronic address: carmelo.militello@cnr.it.
  • Department of Biomedicine, Neuroscience and Advanced Diagnostics (BiND), University of Palermo, Palermo, 90127, Italy. Electronic address: calogero.zarcaro@unipa.it.
  • Department of Biomedicine, Neuroscience and Advanced Diagnostics (BiND), University of Palermo, Palermo, 90127, Italy. Electronic address: tommasovincenzo.bartolotta@unipa.it.
  • Institute for High-Performance Computing and Networking (ICAR-CNR), Italian National Research Council, Palermo, 90146, Italy; Department of Engineering, University of Palermo, Palermo, 90128, Italy. Electronic address:…
  • Department of Biomedicine, Neuroscience and Advanced Diagnostics (BiND), University of Palermo, Palermo, 90127, Italy. Electronic address: salvatore.vitabile@unipa.it.

Förderung

  • University of Palermo

Zitieren

APA

Prinzi, F., Militello, C., Zarcaro, C., Bartolotta, T. V., Gaglio, S., & Vitabile, S. (2025). Rad4XCNN: A new agnostic method for post-hoc global explanation of CNN-derived features by means of Radiomics. Computer Methods and Programs in Biomedicine. https://doi.org/10.1016/j.cmpb.2024.108576

Vancouver

Prinzi F, Militello C, Zarcaro C, Bartolotta TV, Gaglio S, Vitabile S. Rad4XCNN: A new agnostic method for post-hoc global explanation of CNN-derived features by means of Radiomics. Computer Methods and Programs in Biomedicine. 2025. doi:10.1016/j.cmpb.2024.108576

BibTeX
@article{Prinzi2025,
  author = {Prinzi, Francesco and Militello, Carmelo and Zarcaro, Calogero and Bartolotta, Tommaso Vincenzo and Gaglio, Salvatore and Vitabile, Salvatore},
  title = {Rad4XCNN: A new agnostic method for post-hoc global explanation of CNN-derived features by means of Radiomics},
  journal = {Computer Methods and Programs in Biomedicine},
  year = {2025},
  doi = {10.1016/j.cmpb.2024.108576}
}
Kostenlos starten →