SISC: End-to-End Interpretable Discovery Radiomics-Driven Lung Cancer Prediction via Stacked Interpretable Sequencing Cells

Abstract

Lung cancer is the leading cause of cancer-related death worldwide. Computer-aided diagnosis (CAD) systems have shown significant promise in recent years for facilitating the effective detection and classification of abnormal lung nodules in computed tomography (CT) scans. While hand-engineered radiomic features have been traditionally used for lung cancer prediction, there have been significant recent successes achieving state-of-the-art results in the area of discovery radiomics. Here, radiomic sequencers comprising of highly discriminative radiomic features are discovered directly from archival medical data. However, the interpretation of predictions made using such radiomic sequencers remains a challenge. A novel end-to-end interpretable discovery radiomics-driven lung cancer prediction pipeline has been designed, build, and tested. The radiomic sequencer being discovered possesses a deep architecture comprised of stacked interpretable sequencing cells (SISC). The SISC architecture is shown to outperform previous approaches while providing more insight in to its decision making process. The SISC radiomic sequencer is able to achieve state-of-the-art results in lung cancer prediction, and also offers prediction interpretability in the form of critical response maps. The critical response maps are useful for not only validating the predictions of the proposed SISC radiomic sequencer, but also provide improved radiologist-machine collaboration for effective diagnosis.

Combining computed tomography and biologically effective dose in radiomics and deep learning improves prediction of tumor response to robotic lung stereotactic body radiation therapy

“…Similar results were obtained by Lao et al., 60 who demonstrated that the deep features extracted via transfer learning performed better than radiomic features in prediction of overall survival in glioblastoma patients. However, DL models are still perceived as “black boxes,” meaning that it is difficult to determine how they arrive at their predictions, which impairs their use by clinicians as part of their clinical practice 61,62 . To address this issue, we searched for the radiomic features most related to the DL features in the DL CT,BED model.…”

Survey of explainable artificial intelligence techniques for biomedical imaging with deep neural networks

“…The proposed architecture provided better insights into the model's decisions and achieved better performance compared to the considered approaches. The proposed radiomic sequencer achieved correct predictions that can potentially improve clinical adoption [182].…”

“…The results were incorporated to increase the radiologists' confidence level for clinical diagnosis [180]. A stacked interpretable architecture was proposed for the classification of abnormal lung nodules for lung cancer prediction on LIDC-IDRI dataset [182]. The proposed architecture provided better insights into the model's decisions and achieved better performance compared to the considered approaches.…”

A survey on the interpretability of deep learning in medical diagnosis

“…By comparing it with three popular interpretability methods (CAM, VBP, and LRP), it was found that the proposed method can generate a more accurate and clear visual interpretation map. Similarly, the stacked interpretable sequencing cells (SISC) architecture was introduced [ 67 ] to predict lung cancer by generating attentive maps to provide explanations. It can not only achieve remarkable results for lung prediction but also highlight the crucial regions.…”