Interpretable deep learning approach for oral cancer classification using guided attention inference network

Kevin Chew Figueroa, Bofan Song, Sumsum Sunny, Shaobai Li, Keerthi Gurushanth, Pramila Mendonca, Nirza Mukhia, Sanjana Patrick, Shubha Gurudath, Subhashini Raghavan, Tsusennaro Imchen, Shirley T. Leivon, Trupti Kolur, Vivek Shetty, Vidya Bushan, Rohan Ramesh, Vijay Pillai, Petra Wilder-Smith, Alben Sigamani, Amritha SureshMoni Abraham Kuriakose, Praveen Birur, Rongguang Liang

Research output: Contribution to journalArticlepeer-review

Abstract

Significance: Convolutional neural networks (CNNs) show the potential for automated classification of different cancer lesions. However, their lack of interpretability and explainability makes CNNs less than understandable. Furthermore, CNNs may incorrectly concentrate on other areas surrounding the salient object, rather than the network's attention focusing directly on the object to be recognized, as the network has no incentive to focus solely on the correct subjects to be detected. This inhibits the reliability of CNNs, especially for biomedical applications. Aim: Develop a deep learning training approach that could provide understandability to its predictions and directly guide the network to concentrate its attention and accurately delineate cancerous regions of the image. Approach: We utilized Selvaraju et al.'s gradient-weighted class activation mapping to inject interpretability and explainability into CNNs. We adopted a two-stage training process with data augmentation techniques and Li et al.'s guided attention inference network (GAIN) to train images captured using our customized mobile oral screening devices. The GAIN architecture consists of three streams of network training: classification stream, attention mining stream, and bounding box stream. By adopting the GAIN training architecture, we jointly optimized the classification and segmentation accuracy of our CNN by treating these attention maps as reliable priors to develop attention maps with more complete and accurate segmentation. Results: The network's attention map will help us to actively understand what the network is focusing on and looking at during its decision-making process. The results also show that the proposed method could guide the trained neural network to highlight and focus its attention on the correct lesion areas in the images when making a decision, rather than focusing its attention on relevant yet incorrect regions. Conclusions: We demonstrate the effectiveness of our approach for more interpretable and reliable oral potentially malignant lesion and malignant lesion classification.

Original languageEnglish (US)
Article number015001
JournalJournal of biomedical optics
Volume27
Issue number1
DOIs
StatePublished - Jan 1 2022

Keywords

  • Guided attention inference network
  • Interpretable deep learning
  • Oral cancer

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Biomaterials
  • Biomedical Engineering

Fingerprint

Dive into the research topics of 'Interpretable deep learning approach for oral cancer classification using guided attention inference network'. Together they form a unique fingerprint.

Cite this