Citations

Background/Objectives: To develop and evaluate a deep learning-based quality control system for Lumbar Spinal Digital Radiographs (LSDR), designed to automate and improve their evaluation and reduce reliance on manual reviews. Methods: This retrospective study utilized a deep learning workflow comprising image segmentation, feature extraction, and a classification model. The dataset, including anteroposterior (AP) and lateral (LAT) X-ray images, was expanded through data augmentation techniques. Four U-Net-based models were assessed: standard U-Net, Swin-UNet, Attention U-Net, and Attention U-Net with the weight map, with the latter selected for its superior performance. Extracted features, such as brightness, contrast, and anatomical positioning, were used in an XGBoost classifier, which was evaluated using mean intersection over union (mIoU), accuracy, sensitivity, specificity, and AUC. Results: The Attention U-Net with weighted attention outperformed the other models, achieving high mIoU scores in both AP and LAT views. The XGBoost classifier achieved the best performance in classifying images as “qualified” or “unqualified,” with an AUC of approximately 0.9, high accuracy, and balanced sensitivity and specificity. This approach effectively addressed class imbalances and improved model accuracy compared to traditional machine learning models such as MLP and SVM. Conclusions: The developed automated quality control system demonstrated potential for enhancing image quality, enhancing diagnostic reliability, and optimizing clinical workflow efficiency.

Abstract Introduction: Terminal duct lobular unit (TDLU) involution, an age-related physiological process, is characterized by the reduction of size and numbers. Reduced involution is associated with higher mammographic density, a strong breast cancer risk factor. Both factors predict risk of developing breast cancer among women with benign disease. Given that prior studies have been conducted mainly in White women, we compared the extent of TDLU involution among Western and Asian breast cancer cases to describe potential racial heterogeneity in breast cancer etiology. Method: We obtained three TDLU involution metrics (count/100 mm2, mean span, and mean acini count/TDLU; all inversely correlated with TDLU involution) measured in benign breast tissue sections from 379 Chinese (254 luminal A and 125 core basal phenotype [CBP] cases) and 476 Polish (407 luminal A and 69 CBP) breast cancer cases. Polytomous logistic regression was performed using tertiles of TDLU measures as ordinal dependent variable and race as an independent variable. Covariates included age, body mass index, and parity. Analyses were performed separately by age group (<50, ≥50) and subtype (luminal A and CBP). Results: Among luminal A cases, Chinese had significantly greater TDLU count compared with Polish in both age groups (ORtrend= 11.0; 95% CI=2.80-43.2; Ptrend=0.001 for age<50; ORtrend= 12.1; 95% CI=5.12-28.7; Ptrend=1E-08 for age≥50). In addition, Chinese were more likely to have greater mean span and acini count among younger cases (ORtrend=2.5; 95% CI=1.44-4.28; Ptrend=0.001 for mean span and ORtrend=7.4; 95% CI=4.13-13.4; Ptrend=3E-11 for acini count) but not among older cases. Among CBP cases, Chinese were more likely to have greater acini count compared to Polish only among younger women (ORtrend=2.6; 95% CI=1.12-6.95; Ptrend=0.03 for age<50). Conclusion: We found Chinese were more likely to have reduced TDLU involution compared to Polish cases after accounting for potential confounders, with greater differences for luminal A and younger women. Studies of TDLU involution in diverse populations are needed to confirm and understand whether population difference in TDLU involution metrics are related to age at breast cancer onset and risks for specific tumor subtypes. Citation Format: Hyuna Sung, Changyuan Guo, Jennifer Guida, Shan Zheng, Erni Li, Jing Li, Nan Hu, Joseph Deng, Montserrat Garcia-Closas, Jonine Figueroa, Mark Sherman, Gretchen Gierach, Ning Lu, Xiaohong R. Yang. Racial variation in terminal duct lobular unit (TDLU) involution in Chinese and Polish breast cancer patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5290. doi:10.1158/1538-7445.AM2017-5290