Dataset Information

Exploiting deep learning and volunteered geographic information for mapping buildings in Kano, Nigeria.

ABSTRACT: Buildings in the developing world are inadequately mapped. Lack of such critical geospatial data adds unnecessary challenges to locating and reaching a large segment of the world's most vulnerable population, impeding sustainability goals ranging from disaster relief to poverty reduction. Use of volunteered geographic information (VGI) has emerged as a widely accepted source to fill such voids. Despite its promise, availability of building maps for developing countries significantly lags behind demand. We present a new approach, coupling deep convolutional neural networks (CNNs) with VGI for automating building map generation from high-resolution satellite images for Kano state, Nigeria. Specifically, we trained a CNN with VGI building outlines of limited quality and quantity and generated building maps for a 50,000 km² area. Resulting maps are in strong agreement with existing settlement maps and require a fraction of the manual input needed for the latter. The VGI-based maps will provide support across multiple facets of socioeconomic development in Kano state, and demonstrates potential advancements in current mapping capabilities in resource constrained countries.

SUBMITTER: Yuan J

PROVIDER: S-EPMC6198754 | biostudies-literature | 2018 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Exploiting deep learning and volunteered geographic information for mapping buildings in Kano, Nigeria.

Yuan Jiangye J Roy Chowdhury Pranab K PK McKee Jacob J Yang Hsiuhan Lexie HL Weaver Jeanette J Bhaduri Budhendra B

Scientific data 20181023

Buildings in the developing world are inadequately mapped. Lack of such critical geospatial data adds unnecessary challenges to locating and reaching a large segment of the world's most vulnerable population, impeding sustainability goals ranging from disaster relief to poverty reduction. Use of volunteered geographic information (VGI) has emerged as a widely accepted source to fill such voids. Despite its promise, availability of building maps for developing countries significantly lags behind ...[more]

PMID: 30351298

Similar Datasets

Project description:Nigeria is endemic for cholera since 1970, and Kano State report outbreaks annually with high case fatality ratios ranging from 4.98%/2010 to 5.10%/2018 over the last decade. However, interventions focused on cholera prevention and control have been hampered by a lack of understanding of hotspot Local Government Areas (LGAs) that trigger and sustain yearly outbreaks. The goal of this study was to identify and categorize cholera hotspots in Kano State to inform a national plan for disease control and elimination in the State. We obtained LGA level confirmed and suspected cholera data from 2010 to 2019 from the Nigeria Centre for Disease Control (NCDC) and Kano State Ministry of Health. Data on inland waterbodies and population numbers were obtained from online sources and NCDC, respectively. Clusters (hotspots) were identified using SaTScan through a retrospective analysis of the data for the ten-year period using a Poisson discrete space-time scan statistic. We also used a method newly proposed by the Global Task Force on Cholera Control (GTFCC) to identify and rank hotspots based on two epidemiological indicators including mean annual incidence per 100 000 population of reported cases and the persistence of cholera for the study period. In the ten-year period, 16,461 cholera cases were reported with a case fatality ratio of 3.32% and a mean annual incidence rate of 13.4 cases per 100 000 population. Between 2010 and 2019, the most severe cholera exacerbations occurred in 2014 and 2018 with annual incidence rates of 58.01 and 21.52 cases per 100 000 inhabitants, respectively. Compared to 2017, reported cases and deaths increased by 214.56% and 406.67% in 2018. The geographic distribution of outbreaks revealed considerable spatial heterogeneity with the widest in 2014. Space-time clustering analysis identified 18 out of 44 LGAs as high risk for cholera (hotspots) involving both urban and rural LGAs. Cholera clustered around water bodies, and the relative risk of having cholera inside the hotspot LGA were 1.02 to 3.30 times higher than elsewhere in the State. A total of 4,894,144 inhabitants were in these hotspots LGAs. Of these, six LGAs with a total population of 1.665 million had a relative risk greater than 2 compared to the state as a whole. The SaTScan (statistical) and GTFCC methods were in agreement in hotspots identification. This study identified cholera hotspots LGAs in Kano State from 2010-2019. Hotspots appeared in both urban and rural settings. Focusing control strategies on these hotspots will facilitate control and eliminate cholera from the State.

Project description:BACKGROUND:Malaria surveillance system strengthening is essential in the progress towards malaria elimination. In Nigeria, more attention is being given to this recently as the country is striving towards achieving elimination. However, the surveillance system performance is fraught with challenges including poor data quality with varying magnitude by state. This study evaluated the operation of the Kano State malaria surveillance system and assessed its key attributes. METHODS:An observational study design comprising a survey, record review and secondary data analysis, and mixed methods data collection approach were used. Four key stakeholders' and 35 Roll Back Malaria Focal Persons (RBMs) semi-structured interviews on operation of the system and attributes of the surveillance system, were conducted. We analyzed the abstracted 2013-2016 National Health Management Information System web-based malaria datasets. The surveillance system was evaluated using the "2001 United States Centers for Disease Control's updated guidelines for Evaluating Public Health Surveillance Systems". Data were described using means, standard deviation, frequencies and proportions. Chi-squared for linear trends was used. RESULTS:Overall, 24 RBMs (68.6%) had ≤ 15-year experience on malaria surveillance, 29 (82.9%) had formal training on malaria surveillance; 32 RBMs (91.4%) reported case definitions were easy-to-use, reporting forms were easy-to-fill and data flow channels were clearly defined. Twenty-seven respondents (69.2%) reported data tools could accommodate changes and all RBMs understood malaria case definitions. All respondents (4 stakeholders and 34 RBMs [97.1%]) expressed willingness to continue using the system and 33 (84.6%) reported analyzed data were used for decision-making. Public health facilities constituted the main data source. Overall, 65.0% of funding were from partner agencies. Trend of malaria cases showed significant decline (χ2trend = 7.49; P = 0.0006). Timeliness of reporting was below the target (≥ 80%), except being 82% in 2012. CONCLUSIONS:Malaria surveillance system in Kano State was simple, flexible, acceptable, useful and donor-driven but the data were not representative of all health facilities. Timeliness of reporting was suboptimal. We recommended reporting from private health facilities, strengthening human resource capacity for supportive supervision and ensuring adequate government funding to enhance the system's representativeness and improve data quality.

Project description:BackgroundStillbirths are a poignant representation of global inequality. Nigeria is documented to have the second highest rate; yet, the reporting system is inadequate in most Nigerian healthcare facilities. The aim was to identify the determinants of stillbirth among deliveries in the Murtala Muhammad Specialist Hospital (MMSH), Kano, Nigeria.MethodsTwo study designs were used: a case-control study (S1) and a prospective cohort study (S2). Both studies were carried out at the MMSH. For S1, stillbirths were retrospectively matched to a livebirth by time (target of 24 hours' time variation) to establish a case-control study with a 1:1 ratio. Eligibility into S2 included all mothers who were presented at the MMSH in labour regardless of birth outcome. Both were based on recruitment durations, not sample sizes (3 months and 2 months, respectively, 2017-2018). The demographic and clinical data were collected through paper-based questionnaires. Univariable logistic regression was used. Multivariable logistic regression was used to explore relationships between area type and other specific factors.FindingsStillbirth incidence in S2 was 180/1,000 births. Stillbirth was associated with the following factors; no maternal education, previous stillbirth(s), prematurity, living in both semi-rural and rural settings, and having extended time periods between rupture of membranes and delivery. Findings of the multivariable analysis (S1 and S2) indicated that the odds of stillbirth, for those living in a rural area, were further exacerbated in those mothers who had no education, lived in a shack, or had any maternal disease.InterpretationThis research identifies the gravity of this situation in this area and highlights the need for action. Further understanding of some of the findings and exploration into associations are required to inform intervention development.FundingThis collaboration was partially supported by funding from Health and Care Research Wales.

Project description:IntroductionThe knowledge, attitude, and practice of emergency neonatal resuscitation are critical requirements in any facility that offers obstetric and neonatal services. This study aims to conduct a needs assessment survey and obtain individual and facility-level data on expertise and readiness for neonatal resuscitation. We hypothesize that neonatal emergency preparedness among healthcare providers in Kano, Nigeria is associated with the level of knowledge, attitudinal disposition, practice and equipment availability at the facility level.MethodsA semi-structured, self-administered questionnaire was administered to a cross-section of health providers directly involved with neonatal care (n = 112) and attending a neonatal resuscitation workshop in Kano state. Information regarding knowledge, attitude, practice and facility preparedness for neonatal resuscitation was obtained. Bloom's cut-off score and a validated basic emergency obstetric and neonatal care assessment tool were adopted to categorize outcomes. Multivariable logistic regression was employed to determine independent predictors of knowledge and practice.ResultsAlmost half (48% and 42% respectively) of the respondents reported average level of self-assessed knowledge and comfort during resuscitation. Only 7% (95% CI:3.2-13.7) and 5% (95% CI:2.0-11.4) of health providers demonstrated good knowledge and practice scores respectively, with an overall facility preparedness of 46%. Respondents' profession as a physician compared to nurses and midwives predicted good knowledge (aOR = 0.08, 95% CI: 0.01-0.69; p = 0.01), but not practice.ConclusionHealthcare provider's knowledge and practice including facility preparedness for emergency neonatal resuscitation were suboptimal, despite the respondents' relatively high self-assessed attitudinal perception. Physicians demonstrated higher knowledge compared to other health professionals. The low level of respondents' awareness, practice, and facility readiness suggest the current weak state of secondary health systems in Kano.

Dataset Information

Exploiting deep learning and volunteered geographic information for mapping buildings in Kano, Nigeria.

Publications

Exploiting deep learning and volunteered geographic information for mapping buildings in Kano, Nigeria.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets