Explainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization

Primary malignancies in adult brains are globally fatal. Computer vision, especially recent developments in artificial intelligence (AI), have created opportunities to automatically characterize and diagnose tumor lesions in the brain. AI approaches have provided scores of unprecedented accuracy in...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Morteza Esmaeili, Riyas Vettukattil, Hasan Banitalebi, Nina R. Krogh, Jonn Terje Geitung
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	tumor localization black box CNN explainable AI gliomas machine learning Medicine R
Acceso en línea:	https://doaj.org/article/419fbc418eee4fbf9d8ce1756e6bb850
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:419fbc418eee4fbf9d8ce1756e6bb850
record_format	dspace
spelling	oai:doaj.org-article:419fbc418eee4fbf9d8ce1756e6bb8502021-11-25T18:08:04ZExplainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization10.3390/jpm111112132075-4426https://doaj.org/article/419fbc418eee4fbf9d8ce1756e6bb8502021-11-01T00:00:00Zhttps://www.mdpi.com/2075-4426/11/11/1213https://doaj.org/toc/2075-4426Primary malignancies in adult brains are globally fatal. Computer vision, especially recent developments in artificial intelligence (AI), have created opportunities to automatically characterize and diagnose tumor lesions in the brain. AI approaches have provided scores of unprecedented accuracy in different image analysis tasks, including differentiating tumor-containing brains from healthy brains. AI models, however, perform as a black box, concealing the rational interpretations that are an essential step towards translating AI imaging tools into clinical routine. An explainable AI approach aims to visualize the high-level features of trained models or integrate into the training process. This study aims to evaluate the performance of selected deep-learning algorithms on localizing tumor lesions and distinguishing the lesion from healthy regions in magnetic resonance imaging contrasts. Despite a significant correlation between classification and lesion localization accuracy (<i>R</i> = 0.46, <i>p</i> = 0.005), the known AI algorithms, examined in this study, classify some tumor brains based on other non-relevant features. The results suggest that explainable AI approaches can develop an intuition for model interpretability and may play an important role in the performance evaluation of deep learning models. Developing explainable AI approaches will be an essential tool to improve human–machine interactions and assist in the selection of optimal training methods.Morteza EsmaeiliRiyas VettukattilHasan BanitalebiNina R. KroghJonn Terje GeitungMDPI AGarticletumor localizationblack box CNNexplainable AIgliomasmachine learningMedicineRENJournal of Personalized Medicine, Vol 11, Iss 1213, p 1213 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	tumor localization black box CNN explainable AI gliomas machine learning Medicine R
spellingShingle	tumor localization black box CNN explainable AI gliomas machine learning Medicine R Morteza Esmaeili Riyas Vettukattil Hasan Banitalebi Nina R. Krogh Jonn Terje Geitung Explainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization
description	Primary malignancies in adult brains are globally fatal. Computer vision, especially recent developments in artificial intelligence (AI), have created opportunities to automatically characterize and diagnose tumor lesions in the brain. AI approaches have provided scores of unprecedented accuracy in different image analysis tasks, including differentiating tumor-containing brains from healthy brains. AI models, however, perform as a black box, concealing the rational interpretations that are an essential step towards translating AI imaging tools into clinical routine. An explainable AI approach aims to visualize the high-level features of trained models or integrate into the training process. This study aims to evaluate the performance of selected deep-learning algorithms on localizing tumor lesions and distinguishing the lesion from healthy regions in magnetic resonance imaging contrasts. Despite a significant correlation between classification and lesion localization accuracy (<i>R</i> = 0.46, <i>p</i> = 0.005), the known AI algorithms, examined in this study, classify some tumor brains based on other non-relevant features. The results suggest that explainable AI approaches can develop an intuition for model interpretability and may play an important role in the performance evaluation of deep learning models. Developing explainable AI approaches will be an essential tool to improve human–machine interactions and assist in the selection of optimal training methods.
format	article
author	Morteza Esmaeili Riyas Vettukattil Hasan Banitalebi Nina R. Krogh Jonn Terje Geitung
author_facet	Morteza Esmaeili Riyas Vettukattil Hasan Banitalebi Nina R. Krogh Jonn Terje Geitung
author_sort	Morteza Esmaeili
title	Explainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization
title_short	Explainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization
title_full	Explainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization
title_fullStr	Explainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization
title_full_unstemmed	Explainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization
title_sort	explainable artificial intelligence for human-machine interaction in brain tumor localization
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/419fbc418eee4fbf9d8ce1756e6bb850
work_keys_str_mv	AT mortezaesmaeili explainableartificialintelligenceforhumanmachineinteractioninbraintumorlocalization AT riyasvettukattil explainableartificialintelligenceforhumanmachineinteractioninbraintumorlocalization AT hasanbanitalebi explainableartificialintelligenceforhumanmachineinteractioninbraintumorlocalization AT ninarkrogh explainableartificialintelligenceforhumanmachineinteractioninbraintumorlocalization AT jonnterjegeitung explainableartificialintelligenceforhumanmachineinteractioninbraintumorlocalization
_version_	1718411554749153280

Explainable Artificial Intelligence for Human-Machine Interaction in Brain Tumor Localization

Ejemplares similares