EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Applications of interpretable deep learning in neuroimaging: A comprehensive review

Lindsay Munroe, Mariana da Silva, Faezeh Heidari, Irina Grigorescu, Simon Dahan, Emma C Robinson, Maria Deprez, Po-Wah So

Imaging Neuroscience · 12.07.2024 · 2

DOI: 10.1162/imag_a_00214 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Imaging Neuroscience, online erschienen am 12.07.2024. Verfasst von Lindsay Munroe und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neuroimaging, King's College London, London, United Kingdom. Die Arbeit wurde bisher 29-mal zitiert (Crossref) und nennt 161 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIA NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Clinical adoption of deep learning models has been hindered, in part, because the “black-box” nature of neural networks leads to concerns regarding their trustworthiness and reliability. These concerns are particularly relevant in the field of neuroimaging due to the complex brain phenotypes and inter-subject heterogeneity often encountered. The challenge can be addressed by interpretable deep learning (iDL) methods that enable the visualisation and interpretation of the inner workings of deep learning models. This study systematically reviewed the literature on neuroimaging applications of iDL methods and critically analysed how iDL explanation properties were evaluated. Seventy-five studies were included, and ten categories of iDL methods were identified. We also reviewed five properties of iDL explanations that were analysed in the included studies: biological validity, robustness, continuity, selectivity, and downstream task performance. We found that the most popular iDL approaches used in the literature may be sub-optimal for neuroimaging data, and we discussed possible future directions for the field.

Schlagwörter

Stichwörter der Autor:innen

explainable AIinterpretable deep learningintrinsic interpretabilityneuroimaging

Autor:innen und Institutionen

  • Department of Neuroimaging, King's College London, London, United Kingdom.
  • School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom.
  • Institute of Clinical Medicine, University of Eastern Finland, Kuopio, Finland.

Förderung

  • NIA NIH HHS

Zitieren

APA

Munroe, L., da Silva, M., Heidari, F., Grigorescu, I., Dahan, S., Robinson, E. C., Deprez, M., & So, P. W. (2024). Applications of interpretable deep learning in neuroimaging: A comprehensive review. Imaging Neuroscience, 2. https://doi.org/10.1162/imag_a_00214

Vancouver

Munroe L, da Silva M, Heidari F, Grigorescu I, Dahan S, Robinson EC, et al. Applications of interpretable deep learning in neuroimaging: A comprehensive review. Imaging Neuroscience. 2024;2. doi:10.1162/imag_a_00214

BibTeX
@article{Munroe2024,
  author = {Munroe, Lindsay and da Silva, Mariana and Heidari, Faezeh and Grigorescu, Irina and Dahan, Simon and Robinson, Emma C and Deprez, Maria and So, Po-Wah},
  title = {Applications of interpretable deep learning in neuroimaging: A comprehensive review},
  journal = {Imaging Neuroscience},
  year = {2024},
  volume = {2},
  doi = {10.1162/imag_a_00214}
}
Kostenlos starten →