| 1. |
- Tran, Quoc Cuong, et al.
(författare)
-
Novel Ensemble Landslide Predictive Models Based on the Hyperpipes Algorithm : A Case Study in the Nam Dam Commune, Vietnam
- 2020
-
Ingår i: Applied Sciences. - Switzerland : MDPI. - 2076-3417. ; 10:11
-
Tidskriftsartikel (refereegranskat)abstract
- Development of landslide predictive models with strong prediction power has become a major focus of many researchers. This study describes the first application of the Hyperpipes (HP) algorithm for the development of the five novel ensemble models that combine the HP algorithm and the AdaBoost (AB), Bagging (B), Dagging, Decorate, and Real AdaBoost (RAB) ensemble techniques for mapping the spatial variability of landslide susceptibility in the Nam Dan commune, Ha Giang province, Vietnam. Information on 76 historical landslides and ten geo-environmental factors (slope degree, slope aspect, elevation, topographic wetness index, curvature, weathering crust, geology, river density, fault density, and distance from roads) were used for the construction of the training and validation datasets that are the prerequisites for building and testing the proposed models. Using different performance metrics (i.e., the area under the receiver operating characteristic curve (AUC), negative predictive value, positive predictive value, accuracy, sensitivity, specificity, root mean square error, and Kappa), we verified the proficiency of all five ensemble learning techniques in increasing the fitness and predictive powers of the base HP model. Based on the AUC values derived from the models, the ensemble ABHP model that yielded an AUC value of 0.922 was identified as the most efficient model for mapping the landslide susceptibility in the Nam Dan commune, followed by RABHP (AUC = 0.919), BHP (AUC = 0.909), Dagging-HP (AUC = 0.897), Decorate-HP (AUC = 0.865), and the single HP model (AUC = 0.856), respectively. The novel ensemble models proposed for the Nam Dan commune and the resultant susceptibility maps can aid land-use planners in the development of efficient mitigation strategies in response to destructive landslides.
|
|
| 2. |
- Huong, Dao Lan, et al.
(författare)
-
Burden of premature mortality in rural Vietnam from 1999 - 2003 : analyses from a Demographic Surveillance Site
- 2006
-
Ingår i: Population Health Metrics. - : BioMed Central (BMC). - 1478-7954. ; 4, s. Article nr 9-
-
Tidskriftsartikel (refereegranskat)abstract
- Background: Assessing the burden of disease contributes towards evidence-based allocation of limited health resources. However, such measures are not yet commonly available in Vietnam. Taking advantage of the FilaBavi Demographic Surveillance Site (FilaBavi DSS) in Vietnam, this study aimed to establish the feasibility of applying the Years of Life Lost (YLL) technique in the context of a defined DSS, and to estimate the importance of the principal causes of premature mortality in a rural area of Vietnam between 1999 and 2003.Methods: Global Burden of Disease methods were applied. Causes of death were ascertained by verbal autopsy.Results: In five years, 1,240 deaths occurred and for 1,220 cases cause of death information from verbal autopsy was available. Life expectancy at birth was 71.0 (95% confidence interval 69.9–72.1) in males and 80.9 (79.9–81.9) in females. The discounted, but not age weighted YLL per 1,000 population was 85 and 55 for males and females, respectively. The leading causes of YLL and death counts were cardiovascular diseases, malignant neoplasms, unintentional injuries, and neonatal causes. Males contributed 54% of total deaths and 59% of YLL. Males experienced higher YLL than women across all causes. Filabavi mortality estimates are considerably lower than 2002 WHO country estimates for Vietnam. Also the FilaBavi cause distribution varies considerably from the WHO result.Conclusion: The combination of localised demographic surveillance, verbal autopsy and the application of YLL methods enable new insights into the magnitude and importance of significant public health issues in settings where evidence for planning is otherwise scarce. Local mortality data vary considerably from the WHO model-based estimates.
|
|
| 3. |
|
|
| 4. |
- Byass, Peter, et al.
(författare)
-
Refining a probabilistic model for interpreting verbal autopsy data.
- 2006
-
Ingår i: Scandinavian journal of public health. - : SAGE Publications. - 1403-4948 .- 1651-1905. ; 34:1, s. 26-31
-
Tidskriftsartikel (refereegranskat)abstract
- Objective: To build on the previously reported development of a Bayesian probabilistic model for interpreting verbal autopsy (VA) data, attempting to improve the model's performance in determining cause of death and to reassess it. Design: An expert group of clinicians, coming from a wide range geographically and in terms of specialization, was convened. Over a four-day period the content of the previous probabilistic model was reviewed in detail and adjusted as necessary to reflect the group consensus. The revised model was tested with the same 189 VA cases from Vietnam, assessed by two local clinicians, that were used to test the preliminary model. Results: The revised model contained a total of 104 indicators that could be derived from VA data and 34 possible causes of death. When applied to the 189 Vietnamese cases, 142 (75.1%) achieved concordance between the model's output and the previous clinical consensus. The remaining 47 cases (24.9%) were presented to a further independent clinician for reassessment. As a result, consensus between clinical reassessment and the model's output was achieved in 28 cases (14.8%); clinical reassessment and the original clinical opinion agreed in 8 cases (4.2%), and in the remaining 11 cases (5.8%) clinical reassessment, the model, and the original clinical opinion all differed. Thus overall the model was considered to have performed well in 170 cases (89.9%). Conclusions: This approach to interpreting VA data continues to show promise. The next steps will be to evaluate it against other sources of VA data. The expert group approach to determining the required probability base seems to have been a productive one in improving the performance of the model.
|
|
| 5. |
- Huong, Dao Lan, 1974-
(författare)
-
Mortality in transitional Vietnam
- 2006
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Understanding mortality patterns is an essential pre-requisite for guiding public health action and for supporting development of evidence-based policy. However, such information is not sufficiently available in Vietnam. Mortality statistics and causes of death are solely collected from health facilities while most deaths occur at home without the presence of health professionals. Facility-based data cannot represent what happened in the wider community. This thesis studies the patterns and burdens of mortality as well as their relationships with socio-economic status in rural Vietnam. The overall aim is to contribute to the improvement of the current system of mortality data collection in the country for the purposes of public health planning and priority setting. The study was carried out within the framework of an ongoing Demographic Surveillance System (DSS) in Bavi district, Hatay province, northern rural Vietnam. This study used a verbal autopsy (VA) approach to identify cause of death in a cohort of approximately 250,000 person- years over a five-year period from 1999 to 2003. During the five year study, a total of 1,240 deaths were recorded and VA was successfully completed for 1,220 cases. Results revealed that VA was an appropriate and useful method for ascertaining cause of death in this rural Vietnamese community where specific data were otherwise scarce. The mortality pattern reflected a transitional pattern of disease in which the leading cause of death was cardiovascular diseases (CVD), followed by neoplasms, infectious and parasitic diseases, and external causes, accounting for 28.9%, 14.5%, 11.2%, and 9.8%, respectively. In terms of premature mortality, there were 85 and 55 Years of Life Lost (YLL) per 1,000 population for males and females respectively. The largest contributions to YLL were CVDs, malignant neoplasms, unintentional injuries, and perinatal and neonatal causes. In general, men had higher mortality rates than women for all mortality categories. In adults of 20 years and above, mortality rates increased substantially with age, and showed similar age effects for all mortality categories with the strongest association for non-communicable diseases (NCD). Education was an important factor for survival in general, and high economic status seemed to benefit men more than women. Compared with cancer and other NCD causes, higher CVD rates were observed among males, the elderly, and those without formal education, using a Cox proportional hazards model. This study is an initial effort to provide information on mortality patterns in a community using longitudinal follow-up of a dynamic cohort. Continuing the study using the VA approach as part of routine data collection in the setting will help to show trends in mortality patterns for the community over time, which may be useful for priority setting and health planning purposes, not only locally but also at the national level. Further analyses are needed to understand mortality inequality across all ages to have a comprehensive picture of mortality burdens in the setting. Validation studies and further standardization of VA methods should be carried out whenever possible to improve the performance and extension of the technique.
|
|
