Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

In International journal of radiation oncology, biology, physics

PURPOSE : Accurate lymph node (LN) malignancy classification is essential for treatment target identification in head and neck cancer (HNC) radiation therapy. Given the constraints imposed by relatively small sample sizes in real-world medical applications, to classify LN malignancy status accurately, we proposed an attention-guided-classification (AGC) scheme that 1) incorporates human knowledge (i.e., LN contours) into model training to guide model's "learning" direction, alleviating the critical requirement of large training samples by deep learning approaches; 2) doesn't require accurate delineation of LNs in the inference stage, but can highlight the discriminative region nearby the LN which is important for malignancy determination.

METHODS AND MATERIALS : In the proposed AGC scheme, there is an attention-guided-CNN (agCNN) module followed by a classification-CNN (cCNN) module. The input of the proposed AGC scheme is a region-of-interest (ROI) containing the LN and its surrounding tissues. The agCNN is designed to find the discriminative region in the ROI, which outputs an activation map whose voxel values indicate the importance of the voxels in malignancy prediction. Through multiplying the activation map with the ROI, we obtain the input for the cCNN which finally outputs the LN malignancy probability. To demonstrate the effectiveness of the proposed scheme, we have performed experimental studies using PET and contrast-enhanced CT from 129 surgical HNC patients, including 791 LNs with pathological ground truth of malignancy status. Five folder cross validation was used to evaluate the performance.

RESULTS : The sensitivity, specificity, accuracy, and area under the receiver operating characteristic (ROC) curve (AUC) values obtained by the proposed AGC scheme were 0.91, 0.93, 0.92 and 0.98 respectively, significantly outperforming conventional CNN and radiomics approaches at a significance level of 0.05 under paired ROC comparison statistical test.

CONCLUSIONS : We developed an AGC scheme which can highlight the discriminative region in the image for LN malignancy prediction, outperforming a conventional radiomics method that requires accurate segmentation and a standard CNN model without involving segmentation.

Chen Liyuan, Dohopolski Michael, Zhou Zhiguo, Wang Kai, Wang Rongfang, Sher David, Wang Jing