Research Article
Gender-Specific Mental Health Outcomes in Central America: A Natural Experiment
Hugo Moises Montesinos-Yufa*
,
Thea Nagasuru-McKeever*
Issue:
Volume 10, Issue 3, September 2024
Pages:
45-50
Received:
5 July 2024
Accepted:
25 July 2024
Published:
27 August 2024
Abstract: The COVID-19 pandemic and subsequent restrictions have had profound impacts on mental health worldwide, with varying effects across different demographics and regions. Specifically, COVID lockdown measures are known to have had a disparate impact on women. This study aims to better understand this phenomenon by investigating the effect of COVID-19 stringency measures on depression rates among men and women in the Republic of Nicaragua and the Republic of Honduras. The two neighboring countries serve as a natural experiment: the former noted for its relaxed approach to the pandemic, and the latter implementing stricter lockdown measures. Using a Bayesian structural time series model, yearly depression rates were analyzed in both countries, utilizing various weather indicators as predictors, including yearly rainfall and average ground temperature data. In both countries, rates of depression among women were historically higher than among men. The difference in depression rates between women and men increased during the intervention period in both countries (p < 0.001). However, the absolute effect of the intervention in Honduras was significantly higher (p < 0.001) at 0.39 (95% CI: 0.37, 0.41) compared to Nicaragua, which was 0.26 (95% CI: 0.21, 0.31). These findings suggest that the higher stringency measures in Honduras, including prolonged lockdowns and restrictions on movement, may have disproportionately affected women's mental health. These results highlight the importance of considering women’s wellbeing when designing and implementing public health policies, particularly during crises like the COVID-19 pandemic.
Abstract: The COVID-19 pandemic and subsequent restrictions have had profound impacts on mental health worldwide, with varying effects across different demographics and regions. Specifically, COVID lockdown measures are known to have had a disparate impact on women. This study aims to better understand this phenomenon by investigating the effect of COVID-19 ...
Show More
Research Article
An Active Learning Semantic Segmentation Model Based on an Improved Double Deep Q-Network
Yan Yu*
Issue:
Volume 10, Issue 3, September 2024
Pages:
51-61
Received:
14 August 2024
Accepted:
22 August 2024
Published:
27 August 2024
Abstract: Image semantic segmentation is essential in fields such as computer vision, autonomous driving, and human-computer interaction due to its ability to accurately identify and classify each pixel in an image. However, this task is fraught with challenges, including the difficulty of obtaining detailed pixel labels and the problem of class imbalance in segmentation datasets. These challenges can hinder the effectiveness and efficiency of segmentation models. To address these issues, we propose an active learning semantic segmentation model named CG_D3QN, which is designed and implemented based on an enhanced Double Deep Q-Network (D3QN). The proposed CG_D3QN model incorporates a hybrid network structure that combines a dueling network architecture with Gated Recurrent Units (GRUs). This novel approach improves policy evaluation accuracy and computational efficiency by mitigating a Q-value overestimation and making better use of historical state information. Our experiments, conducted on the CamVid and Cityscapes datasets, reveal that the CG_D3QN model significantly reduces the number of required sample annotations by 65.0% compared to traditional methods. Additionally, it enhances the mean Intersection over Union (IoU) for underrepresented categories by approximately 1% to 3%. These results highlight the model’s effectiveness in lowering annotation costs, addressing class imbalance, and its versatility across different segmentation networks.
Abstract: Image semantic segmentation is essential in fields such as computer vision, autonomous driving, and human-computer interaction due to its ability to accurately identify and classify each pixel in an image. However, this task is fraught with challenges, including the difficulty of obtaining detailed pixel labels and the problem of class imbalance in...
Show More
