A sandglass ladder residual network(SLRN)is proposed for multi-label chest X-ray classification,thereby improving the accuracy of clinical diagnosis.SLRN consists of 3 key modules:(1)a sandglass convolutional module to simultaneously extract inter-channel and spatial information;(2)a ladder self attention block to achieve different window divisions through shift operations,expand the receptive field,and realize multi-scale feature extraction and fusion;(3)class specific residual attention in the multi-label classification stage to capture the correlation between different labels and the importance of features for accomplishing more accurate classification by adjusting the weights of different features.The proposed model is validated using the IU X-Ray dataset collected by Indiana University and the publicly available Chest X-Ray14 dataset collected by the National Institutes of Health in the United States;and the results demonstrate that SLRN which combines the advantages of convolutional neural network and vision transformer enables the capture of local features and global correlations in images,better handles long-distance dependencies,and assists doctors in clinical diagnosis.

A sandglass ladder residual network(SLRN)is proposed for multi-label chest X-ray classification,thereby improving the accuracy of clinical diagnosis.SLRN consists of 3 key modules:(1)a sandglass convolutional module to simultaneously extract inter-channel and spatial information;(2)a ladder self attention block to achieve different window divisions through shift operations,expand the receptive field,and realize multi-scale feature extraction and fusion;(3)class specific residual attention in the multi-label classification stage to capture the correlation between different labels and the importance of features for accomplishing more accurate classification by adjusting the weights of different features.The proposed model is validated using the IU X-Ray dataset collected by Indiana University and the publicly available Chest X-Ray14 dataset collected by the National Institutes of Health in the United States;and the results demonstrate that SLRN which combines the advantages of convolutional neural network and vision transformer enables the capture of local features and global correlations in images,better handles long-distance dependencies,and assists doctors in clinical diagnosis.

BACKGROUNDTo study the mortality rate and its trend in lung cancer patients in Yanting County, Sichuan, P.R.China, during 1969-1997.

METHODSAccording to the surveillance death data of the residents in Yanting County, the time series of mortality rates of lung cancer, the average changing speed of mortality rate every year and the proportion of death from lung cancer among all malignant diseases were analyzed. The relationship between the age and the death of lung cancer was explored by birth cohort analysis.

RESULTSThe mortality rate of lung cancer increased year by year (Chi-square=457.51, P=0.000). The proportion of death from lung cancer among all malignant diseases remarkably increased year by year (Chi-square=273.29, P=0.000). Both in male and female lung cancer patients, the mortality rate increased with age. And in the later birth group, the mortality rate increased more quickly.

CONCLUSIONSThe mortality rate of lung cancer patients in Yanting County has significantly gone up during the past 28 years. The prevention and treatment of malignant tumor should be focused on lung cancer.