Convolutional Neural Networks for the evaluation of cancer in Barrett's esophagus: Explainable AI to lighten up the black-box

Author(s)

de Souza Jr, Luis A. ; Mendel, Robert ^[1] ; Strasser, Sophia ^[1] ; Ebigbo, Alanna ^[2] ; Probst, Andreas ^[2] ; Messmann, Helmut ^[2] ; Papa, Joao P. ^[3] ; Palm, Christoph ^{[1, 4]}

Affiliation

^[1] de Souza Jr, Jr., Luis A., Ostbayer TH Regensburg OTH Regensburg, Regensburg Med Image Comp ReM, Regensburg - Germany

^[2] Univ Klinikum Augsburg, Med Klin 3, Augsburg - Germany

^[3] Sao Paulo State Univ, Dept Comp, UNESP, Sao Paulo - Brazil

^[4] OTH Regensburg, Regensburg Ctr Hlth Sci & Technol RCHST, Regensburg - Germany

• Document type: Journal article
• Source: COMPUTERS IN BIOLOGY AND MEDICINE; v. 135, AUG 2021.
• Web of Science Citations: 0

Abstract

Even though artificial intelligence and machine learning have demonstrated remarkable performances in medical image computing, their level of accountability and transparency must be provided in such evaluations. The reliability related to machine learning predictions must be explained and interpreted, especially if diagnosis support is addressed. For this task, the black-box nature of deep learning techniques must be lightened up to transfer its promising results into clinical practice. Hence, we aim to investigate the use of explainable artificial intelligence techniques to quantitatively highlight discriminative regions during the classification of earlycancerous tissues in Barrett's esophagus-diagnosed patients. Four Convolutional Neural Network models (AlexNet, SqueezeNet, ResNet50, and VGG16) were analyzed using five different interpretation techniques (saliency, guided backpropagation, integrated gradients, input x gradients, and DeepLIFT) to compare their agreement with experts' previous annotations of cancerous tissue. We could show that saliency attributes match best with the manual experts' delineations. Moreover, there is moderate to high correlation between the sensitivity of a model and the human-and-computer agreement. The results also lightened that the higher the model's sensitivity, the stronger the correlation of human and computational segmentation agreement. We observed a relevant relation between computational learning and experts' insights, demonstrating how human knowledge may influence the correct computational learning. (AU)

• FAPESP's process: 17/04847-9 - Barrett's Esophagus Assisted Diagnosis Using Machine Learning
• Grantee: Luis Antonio de Souza Júnior
• Support Opportunities: Scholarships in Brazil - Doctorate
• FAPESP's process: 16/19403-6 - Energy-based learning models and their applications
• Grantee: João Paulo Papa
• Support Opportunities: Regular Research Grants
• FAPESP's process: 14/12236-1 - AnImaLS: Annotation of Images in Large Scale: what can machines and specialists learn from interaction?
• Grantee: Alexandre Xavier Falcão
• Support Opportunities: Research Projects - Thematic Grants
• FAPESP's process: 13/07375-0 - CeMEAI - Center for Mathematical Sciences Applied to Industry
• Grantee: Francisco Louzada Neto
• Support Opportunities: Research Grants - Research, Innovation and Dissemination Centers - RIDC
• FAPESP's process: 19/08605-5 - Computer-assisted diagnosis of Barretts's esophagus using machine learning techniques
• Grantee: Luis Antonio de Souza Júnior
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate