A gradient mapping guided explainable deep neural network for extracapsular extension identification in 3D head and neck cancer computed tomography images.

BACKGROUND: Diagnosis and treatment management for head and neck squamous cell carcinoma (HNSCC) is guided by routine diagnostic head and neck computed tomography (CT) scans to identify tumor and lymph node features. The extracapsular extension (ECE) is a strong predictor of patients' survival outcomes with HNSCC. It is essential to detect the occurrence of ECE as it changes staging and treatment planning for patients. Current clinical ECE detection relies on visual identification and pathologic confirmation conducted by clinicians. However, manual annotation of the lymph node region is a required data preprocessing step in most of the current machine learning-based ECE diagnosis studies. PURPOSE: In this paper, we propose a Gradient Mapping Guided Explainable Network (GMGENet) framework to perform ECE identification automatically without requiring annotated lymph node region information. METHODS: The gradient-weighted class activation mapping (Grad-CAM) technique is applied to guide the deep learning algorithm to focus on the regions that are highly related to ECE. The proposed framework includes an extractor and a classifier. In a joint training process, informative volumes of interest (VOIs) are extracted by the extractor without labeled lymph node region information, and the classifier learns the pattern to classify the extracted VOIs into ECE positive and negative. RESULTS: In evaluation, the proposed methods are well-trained and tested using cross-validation. GMGENet achieved test accuracy and area under the curve (AUC) of 92.2% and 89.3%, respectively. GMGENetV2 achieved 90.3% accuracy and 91.7% AUC in the test. The results were compared with different existing models and further confirmed and explained by generating ECE probability heatmaps via a Grad-CAM technique. The presence or absence of ECE has been analyzed and correlated with ground truth histopathological findings. CONCLUSIONS: The proposed deep network can learn meaningful patterns to identify ECE without providing lymph node contours. The introduced ECE heatmaps will contribute to the clinical implementations of the proposed model and reveal unknown features to radiologists. The outcome of this study is expected to promote the implementation of explainable artificial intelligence-assiste ECE detection.

Upcoming imaging concepts and their impact on treatment planning and treatment response in radiation oncology.

For 2018, the American Cancer Society estimated that there would be approximately 1.7 million new diagnoses of cancer and about 609,640 cancer-related deaths in the United States. By 2030 these numbers are anticipated to exceed a staggering 21 million annual diagnoses and 13 million cancer-related deaths. The three primary therapeutic modalities for cancer treatments are surgery, chemotherapy, and radiation therapy. Individually or in combination, these treatment modalities have provided and continue to provide curative and palliative care to the myriad victims of cancer.Today, CT-based treatment planning is the primary means through which conventional photon radiation therapy is planned. Although CT remains the primary treatment planning modality, the field of radiation oncology is moving beyond the sole use of CT scans to define treatment targets and organs at risk. Complementary tissue scans, such as magnetic resonance imaging (MRI) and positron electron emission (PET) scans, have all improved a physician's ability to more specifically identify target tissues, and in some cases, international guidelines have even been issued. Moreover, efforts to combine PET and MR to define solid tumors for radiotherapy planning and treatment evaluation are also gaining traction.Keeping these advances in mind, we present brief overviews of other up-and-coming key imaging concepts that appear promising for initial treatment target definition or treatment response from radiation therapy.