Lung Cancer Risk Attributable to Active Smoking in China:A Systematic Review and Meta-Analysis

Objective No consensus exists on the relative risk(RR)of lung cancer(LC)attributable to active smoking in China.This study aimed to evaluate the unified RR of LC attributable to active smoking among the Chinese population.Methods A systematic literature search of seven databases was conducted to identify studies reporting active smoking among smokers versus nonsmokers in China.Primary articles on LC providing risk estimates with their 95%confidence intervals(CIs)for“ever”“former”or“current”smokers from China were selected.Meta-analysis was used to estimate the pooled RR of active smoking.Results Forty-four unique studies were included.Compared with that of nonsmokers,the pooled RR(95%CI)for“ever”“former”and“current”smokers were 3.26(2.79–3.82),2.95(1.71–5.08),and 5.16(2.58–10.34)among men,3.18(2.78–3.63),2.70(2.08–3.51),and 4.27(3.61–5.06)among women,and2.71(2.12–3.46),2.66(2.45–2.88),and 4.21(3.25–5.45)in both sexes combined,respectively.Conclusion The RR of LC has remained relatively stable(range,2–6)over the past four decades in China.Early quitting of smoking could reduce the RR to some extent;however,completely refraining from smoking is the best way to avoid its adverse effects.

An Efficient ResNetSE Architecture for Smoking Activity Recognition from Smartwatch

Smoking is a major cause of cancer,heart disease and other afflictions that lead to early mortality.An effective smoking classification mechanism that provides insights into individual smoking habits would assist in implementing addiction treatment initiatives.Smoking activities often accompany other activities such as drinking or eating.Consequently,smoking activity recognition can be a challenging topic in human activity recognition(HAR).A deep learning framework for smoking activity recognition(SAR)employing smartwatch sensors was proposed together with a deep residual network combined with squeeze-and-excitation modules(ResNetSE)to increase the effectiveness of the SAR framework.The proposed model was tested against basic convolutional neural networks(CNNs)and recurrent neural networks(LSTM,BiLSTM,GRU and BiGRU)to recognize smoking and other similar activities such as drinking,eating and walking using the UT-Smoke dataset.Three different scenarios were investigated for their recognition performances using standard HAR metrics(accuracy,F1-score and the area under the ROC curve).Our proposed ResNetSE outperformed the other basic deep learning networks,with maximum accuracy of 98.63%.