Lymphadenopathy (LAP) after COVID-19 vaccination in patients with a diagnosis of cancer has been challenging. We analyzed imaging and clinical features from early cases of axillary LAP in six COVID-19 vaccine recipients with a history of breast cancer. Among the patients with a history of breast cancer and recent COVID-19 vaccine administration, six patients who showed isolated axillary LAP were gathered. Radiologic features were reviewed from breast ultrasound, chest computed tomography, and breast magnetic resonance imaging. Clinical and pathological information were obtained for analysis. The interval between ultrasound detection of LAP and last COVID-19 vaccine administration ranged from 14 to 28 days (mean 21.67 days). Round shape of the lymph node and irregular cortex were noted in 2 and 0 cases, respectively. Mean maximum cortical thickness, length to width ratio and interval aggravation in maximum cortical thickening were 4.2 mm, 1.34, and 2.81-fold with cut-off value of 3 mm, 1.5, 2.0-fold, respectively. We observed axillary LAP ipsilateral to a recent vaccine administration persisting longer than what the Centers for Disease Control and Prevention announced. In our patients, COVID-19 vaccine-related LAP tended to show increased cortical thickness without cortical irregularity. Oncologist as well as radiologist should be familiar with the fact that COVID-19 vaccines, regardless of vaccine type or dosage, can frequently cause ipsilateral axillary LAP, showing some suspicious features more often than others, and can persist longer than anticipated so that both over- and underdiagnosis can be avoided. We report our observations in six patients and provide an exhaustive review of the published literature

Lim J, et al. J Korean Med Sci. 2011 Mar;26(3):457-460. http://dx.doi.org/10.3346/jkms.2011.26.3.457

The purpose of this study was to develop a mammography-based deep learning (DL) model for predicting the risk of breast cancer in Asian women. This retrospective study included 287 examinations in 153 women in the cancer group and 736 examinations in 447 women in the negative group, obtained from the databases of two tertiary hospitals between November 2012 and March 2022. All examinations were labeled as either dense breast or nondense breast, and then randomly assigned to either training, validation, or test sets. DL models, referred to as image-level and examination-level models, were developed. Both models were trained to predict whether or not the breast would develop breast cancer with two datasets: the whole dataset and the dense-only dataset. The performance of DL models was evaluated using the accuracy, precision, sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (AUC). On a test set, performance metrics for the four scenarios were obtained: image-level model with whole dataset, image-level model with dense-only dataset, examination-level model with whole dataset, and examination-level model with dense-only dataset with AUCs of 0.71, 0.75, 0.66, and 0.67, respectively. Our DL models using mammograms have the potential to predict breast cancer risk in Asian women.

Objective: This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise.Methods: Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics—compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)—from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model.Results: The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees.Conclusion: The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.

To evaluate the negative predictive value (NPV) of ultrasound (US) elastography for non-palpable Breast Imaging Reporting and Data System (BI-RADS) category 3 lesions on ultrasonography and to determine whether US elastography is helpful in reducing the number of BI-RADS category 3 lesions on ultrasonography. Two hundred seventy-six consecutive, non-palpable BI-RADS category 3 lesions in 256 women who underwent US elastography and US-guided core biopsy, and who had at least 12 months of follow-up data, comprised our study group. The BI-RADS final assessment category and elasticity score were prospectively and independently classified. The rate of malignancy and NPV according to the elasticity score were analysed. We also investigated whether there was a subset of BI-RADS category 3 lesions that were of benign histology but negative on elastography. Of the 276 non-palpable BI-RADS category 3 lesions, three lesions (1.0%) were finally confirmed as ductal carcinomas in situ. No cancers were found in the remaining 273 lesions with benign biopsy histology at a mean follow-up of 39.4 months (range, 12 to 72 months). The NPV of a negative elasticity score (elasticity score of 1) was 99.3% (165 of 166). If BI-RADS category 3 lesions showing a negative elasticity score were downgraded to BI-RADS category 2, 60.4% (165 of 273) of them with benign histology could have been safely followed without biopsy with an increased malignancy rate from 1% (3 of 276) to 1.8% (2 of 110), which is not significantly higher (P=0.626). US elastography has the potential to reduce the number of BI-RADS category 3 lesions on ultrasonography.