Contact network models are recent alternatives to equation-based models in epidemiology. In this paper, the spread of disease is modeled on contact networks using bond percolation. The weight of the edges in the contact graphs is determined as a function of several variables in which case the weight is the product of the probabilities of independent events involving each of the variables. In the first experiment, the weight of the edges is computed from a single variable involving the number of passengers on flights between two cities within the United States, and in the second experiment, the weight of the edges is computed as a function of several variables using data from 2012 Kenyan household contact networks. In addition, the paper explored the dynamics and adaptive nature of contact networks. The results from the contact network model outperform the equation-based model in estimating the spread of the 1918 Influenza virus.

Anomaly detection in graph-based data is an emerging field in machine learning with many relevant applications. Although some algorithms have been developed, current models lack consistency on real-world data and often have problems with overfitting. The paper presents a new model to address these challenges. The model is a combination of graph neural network layers in a variational autoencoder framework. The graph convolution layers learn complex relationships between the node attributes, while also taking the graph structure into account. By using the variational autoencoder framework, the model is less likely to have overfitting problems. The model outperformed existing models on four of five real-world datasets with organic outliers, two out of three real-world datasets with synthetic outliers, and was the second best on the other two datasets. The model demonstrates the potential to combine the variational autoencoder architecture with graph convolution layers in graph deep learning tasks on outlier detection.

Background: Surveillance mammography is recommended for all women with a history of breast cancer. Risk-guided surveillance incorporating advanced imaging modalities based on individual risk of a second cancer could improve cancer detection. However, personalized surveillance may also amplify disparities. Methods: In simulated populations using inputs from the Breast Cancer Surveillance Consortium (BCSC), we investigated race- and ethnicity-based disparities. Disparities were decomposed into those due to primary breast cancer and treatment characteristics, social determinants of health (SDOH) and differential error in second cancer ascertainment by modeling populations with or without variation across race and ethnicity in the distribution of these characteristics. We estimated effects of disparities on mammography performance and supplemental imaging recommendations stratified by race and ethnicity. Results: In simulated cohorts based on 65,446 BCSC surveillance mammograms, when only cancer characteristics varied by race and ethnicity, mammograms for Black women had lower sensitivity compared with the overall population (64.1% vs. 71.1%). Differences between Black women and the overall population were larger when both cancer characteristics and SDOH varied by race and ethnicity (53.8% vs. 71.1%). Basing supplemental imaging recommendations on high predicted second cancer risk resulted in less frequent recommendations for Hispanic (6.7%) and Asian/Pacific Islander women (6.4%) compared with the overall population (10.0%). Conclusions: Variation in cancer characteristics and SDOH led to disparities in surveillance mammography performance and recommendations for supplemental imaging. Impact: Risk-guided surveillance imaging may exacerbate disparities. Decision-makers should consider implications for equity in cancer outcomes resulting from implementing risk-guided screening programs. See related In the Spotlight, p. 1479

Supplementary Table 3. Sensitivity and cancer risk prediction logistic regression model coefficients