医学
疾病负担
胆道癌
胆囊癌
队列
自回归积分移动平均
疾病负担
代群效应
全球卫生
人口学
疾病
癌症
环境卫生
内科学
公共卫生
计算机科学
时间序列
病理
机器学习
吉西他滨
社会学
作者
Shuping Qiu,Wanting Zeng,Jilin Zhang,Jianfeng Xie,Xiaoping Chen
摘要
Gallbladder and biliary tract cancer (GBTC) is a serious disease burden. A comprehensive assessment of the disease burden is essential for improving prevention and treatment strategies. The estimated annual percentage change, Joinpoint regression analysis and age-period-cohort model (APCM) were used to comprehensively evaluate the current status and trend of GBTC burden from 1990 to 2021 from the Global Burden of Disease Study. From the perspective of deep learning, a hierarchical weighted long short-term memory network model (SW-LSTM) is proposed for trend prediction to overcome the shortcomings of traditional models. The global GBTC burden increased non-linearly with age, which was higher in women than in men. With the increase of SDI, the gender difference showed a decreasing trend. Significant period and cohort effects were observed for the indicators in the remaining regions except for some indicators in the low-and low-middle-SDI regions. Age-standardised indicators in the high, high-middle and middle SDI regions showed a downward trend, while the remaining regions showed an upward trend. The proportion of age-standardised mortality rate attributable to high BMI increased with the increase of SDI. The prediction results showed that the SW-LSTM model outperformed the APCM and ARIMA models in prediction accuracy. The SW-LSTM model proposed in this paper can provide more accurate prediction information to assist in the development of more targeted prevention strategies. In view of the impact of GBTC on global health, especially among women and the elderly, effective measures should be taken to reverse the increasing trend of GBTC.
科研通智能强力驱动
Strongly Powered by AbleSci AI