Explore the vast range of titles available.

Background It is unknown whether Female Community Health Volunteers’ (FCHVs) can counsel for hyperlipidemia in rural Nepal. Methods Using the Health Belief Model, we evaluated FCHV’s knowledge, self-efficacy, and barriers to counsel for hyperlipidemia in two phases eleven months apart among 28 FCHVs from rural mid-Western Nepal. In each phase, we conducted four Focused Group Discussions (FGDs), hyperlipidemia-related training and two similar surveys before and after the training. We used inductive and deductive codes for thematic analysis and descriptive statistics for quantitative analysis. We integrated the results for complementarity and convergence using concurrent embedded design (Qual + quan). Results FCHVs’ mean age was 48 years and 21 out of 28 had worked for > 10 years. We found four themes in FGDs. In Phase 1, despite having interest, FCHVs had limited knowledge and confidence in counseling for hyperlipidemia. However, with sufficient training, they believed they could counsel. In Phase 2, FCHVs conveyed improved knowledge and self-efficacy. They expressed community might be concerned about their expertise, which improved in Phase 2. Quantitatively, FCHVs’ knowledge improved immediately after the initial training, which was stable in Phase 2. Inadequate training was identified less as a barrier in Phase 2, but inadequate time and incentive were identified more often, and community’s perception of FCHVs’ skills remained unchanged. Conclusion FCHVs want to provide hyperlipidemia counseling. Despite our trainings and FCHV’s perceived self-efficacy, knowledge gap persisted. FCHVs’ workload, inadequate incentives and knowledge were important barriers. Balanced workload, regular trainings and adequate incentives are important to engage FCHVs in hyperlipidemia management.

The project aims to develop differentially private deep learning models for image classification on CIFAR-10 datasets \\cite{cifar10} and analyze the impact of various privacy parameters on model accuracy. We have implemented five different deep learning models, namely ConvNet, ResNet18, EfficientNet, ViT, and DenseNet121 and three supervised classifiers namely K-Nearest Neighbors, Naive Bayes Classifier and Support Vector Machine. We evaluated the performance of these models under varying settings. Our best performing model to date is EfficientNet with test accuracy of \\(59.63\\%\\) with the following parameters (Adam optimizer, batch size 256, epoch size 100, epsilon value 5.0, learning rate \\(1e-3\\), clipping threshold 1.0, and noise multiplier 0.912).