BackgroundLung cancer is the leading cause of cancer-related mortality, and accurate prediction of patient survival can aid treatment planning and potentially improve outcomes. In this study, we proposed an automated system capable of lung segmentation and survival prediction using graph convolution neural network (GCN) with CT data in non-small cell lung cancer (NSCLC) patients.MethodsIn this retrospective study, we segmented 10 parts of the lung CT images and built individual lung graphs as inputs to train a GCN model to predict 5-year overall survival. A Cox proportional-hazard model, a set of machine learning (ML) models, a convolutional neural network based on tumor (Tumor-CNN), and the current TNM staging system were used as comparison.FindingsA total of 1,705 patients (main cohort) and 125 patients (external validation cohort) with lung cancer (stages I and II) were included. The GCN model was significantly predictive of 5-year overall survival with an AUC of 0.732 (p < 0.0001). The model stratified patients into low- and high-risk groups, which were associated with overall survival (HR = 5.41; 95% CI:, 2.32–10.14; p < 0.0001). On external validation dataset, our GCN model achieved the AUC score of 0.678 (95% CI: 0.564–0.792; p < 0.0001).InterpretationThe proposed GCN model outperformed all ML, Tumor-CNN, and TNM staging models. This study demonstrated the value of utilizing medical imaging graph structure data, resulting in a robust and effective model for the prediction of survival in early-stage lung cancer.

Background: The success of social distancing implementations of severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2) depends heavily on population compliance. Mathematical modelling has been used extensively to assess the rate of viral transmission from behavioural responses. Previous epidemics of SARS-Cov-2 have been characterised by superspreaders, a small number of individuals who transmit a disease to a large group of individuals, who contribute to the stochasticity (or randomness) of transmission compared to other pathogens such as Influenza. This growing evidence proves an urgent matter to understand transmission routes in order to target and combat outbreaks. Objective: To investigate the role of superspreaders in the rate of viral transmission with various levels of compliance. Method: A SEIRS inspired social network model is adapted and calibrated to observe the infected links of a general population with and without superspreaders on four compliance levels. Local and global connection parameters are adjusted to simulate close contact networks and travel restrictions respectively and each performance assessed. The mean and standard deviation of infections with superspreaders and non-superspreaders were calculated for each compliance level. Results: Increased levels of compliance of superspreaders proves a significant reduction in infections. Assuming long-lasting immunity, superspreaders could potentially slow down the spread due to their high connectivity. Discussion: The main advantage of applying the network model is to capture the heterogeneity and locality of social networks, including the role of superspreaders in epidemic dynamics. The main challenge is the immediate attention on social settings with targeted interventions to tackle superspreaders in future empirical work. Conclusion: Superspreaders play a central role in slowing down infection spread following compliance guidelines. It is crucial to adjust social distancing measures to prevent future outbreaks accompanied by population-wide testing and effective tracing.