Faculty of Science & Technology (FST)
http://erepository.uonbi.ac.ke/handle/11295/8015
2024-03-29T04:35:33ZGovernance of the Practice of Traditional Medicine in Selected Markets of Western Kenya.
http://erepository.uonbi.ac.ke/handle/11295/164317
Governance of the Practice of Traditional Medicine in Selected Markets of Western Kenya.
Chebii, Willy K.
Traditional medicine (TM) is a conglomeration of natural products that are used in the treatment of diseases and sociocultural syndromes. Folk medicine employs a wider array of natural products that primarily include medicinal plants and in some instances the use of minerals or animal products. The practice of TM is gaining immense popularity among rural, urban and peri-urban populations, this is highly associated with the increasing challenge in the treatment of modern lifestyle diseases using allopathic or biomedical drugs. TM therefore provides a justifiable alternative medical system where uses find a lasting remedy or a complement therapy. The study attempted to evaluate the current governance practices of the TM industry, trade and uses, and finally assessed the sociocultural aspects that promote the conservation and preservation of medicinal plant resources. The survey used a purposive sampling technique incorporating snowball methods where knowledgeable and willing respondents were selected for oral interviews and allowed to recruit other competent respondents into the survey. Face to face oral interviews were conducted using a pre-tested and a liberal questionnaire after procuring prior informed consents. Herbarium techniques and standard Flora of Kenya literature sources were used to process and identify frequently sold medicinal plants in the informal markets of Western Kenya. The collected data was organized and presented in MS excel spreadsheets and subjected to descriptive statistics (frequencies, means, percentages). Concise literature and desktop reviews was conducted to address the recent sociocultural conservation strategies of TM. The data was presented on tables, column or bar graphs and pie charts. Frequency of Citation (FC) and Relative Frequency of Citation (RFC) ethnobotanical indices were key in the identification of useful and frequently traded medicinal plant species. Women practitioners’ dominated the medicine markets whereas men dominated the TM leadership circles. Majority of the TM market practitioners (TMPs) were older with a mean age of 64 years and a mean practicing age of 24 years.
Slightly more than half of the TM traders (54%) were willing to be interviewed and only fifteen per cent (15%) of the practitioners had acquired a certificate of registration or recognition. From the market survey, 90 plant species belonging to 79 genera and 46 plant families were identified, with T. emetica (RFC = 0.37) registering the highest RFC followed by D. schimperi (RFC = 0.27), C. spinarum (RFC = 0.23) and Aloe spp. (RFC = 0.23) in that order. The exotic neem tree (A. indica) was commonly sought by buyers whereas trees and roots were the frequently traded plant habit and plant part respectively. Formal governance practices continue to attract a lot of interest and unrivalled attention as compared to community, cultural and societal driven informal governance practices. All practitioners followed the laid down formal governance practices whereas the traditional governance practices varied from market to market as their cultures, ethnic affiliations and herbal remedies vary in use and application. An all-inclusive effort must be put to place to ensure effective mainstreaming and long-lasting integration of TM into the general healthcare. Finally, most commonly traded medicinal plants are threatened daily with uncontrolled extraction must be conserved in the wild, cultivated or domesticated to preclude decline or extinction in the wild.
2023-01-01T00:00:00ZAssessment of Community Institutions' Contribution to Forest and Water Governance in the Kilungu Catchment, Kenya
http://erepository.uonbi.ac.ke/handle/11295/164316
Assessment of Community Institutions' Contribution to Forest and Water Governance in the Kilungu Catchment, Kenya
Wambua, Mumbi P.
The roles and responsibilities of Community Forest Associations (CFAs) and Associations of Water Resource Users (WRUAs) in forest and water resource management have been investigated. However, their comparative contributions to the management of water and forest resources, particularly in the Kilungu catchment, are unknown, as are their effectiveness and activities, organizational and structural design, opportunities and challenges, and capacity to manage catchment sustainably. As a result, a descriptive cross-sectional study design using cluster, simple random, and purposive sampling was carried out in Kenya's Kilungu catchment area. As a result, the seven clusters of the Kilungu catchment were chosen for the study: Kilungu, Kitumbuuni, Ndaatai, Kiongwani, Kenze, Nduluni, and Kiio. The study's main goal was to assess the contributions of the Kenze CFA and Upper Kaiti WRUA to Kilungu catchment management, with specific goals to examine the CFA and WRUA's activities and effectiveness, investigate the structure and functions of the CFA and WRUA, analyze the institutions' challenges and opportunities, and evaluate the CFA and WRUA's capacity in managing the catchment sustainably.
Focus group discussions, key informant interviews, and household questionnaires were used to collect quantitative data. Field notes, interview transcripts, and observation skills were used to collect qualitative data. The data was examined using ANOVA. Cross-tabulations and a Pearson Correlation analysis were used to see if any relationships existed between the study parameters. Even though more respondents were aware of the CFA (47%) than the WRUA (43%), focus group discussions revealed that the WRUA far outperformed the CFA in involving its members in active participation in all aspects of catchment management. A person's (p) correlation analysis of CFA and WRUA structure awareness, functioning, and efficiency in catchment management revealed a strong positive relationship (p=0.76). Furthermore, the upper Kaiti WRUA was more efficient at moderately (31%) and highly effective (37%) in catchment management. The Pearson correlations study on the relationship between education level and statutory function execution was found to be statistically significant to a low degree (p= 0.2 49) for both the CFA and the WRUA.
Kenze CFA faced more challenges than Upper Kaiti WRUA, receiving ratings of Very high (58%) compared to Upper Kaiti WRUA’s rating of Very high (22%), respectively. The findings revealed that the challenges had a significant impact on the outcomes of the Kenze CFA operation. The challenges level affecting CFA and WRUA performance and catchment degradation were positively correlated and statistically significant (p=0.72), according to Pearson correlations analysis. According to the institution's evaluations of aspects of continuously improving functioning, both the CFA and the WRUA were working well in terms of inclusion and equity, with extremely good ratings on accountability, effectiveness, and efficiency.
Finally, in terms of gender inclusivity, equity, and catchment management activity implementation, the Upper Kaiti WRUA structure outperformed the Kenze CFA. Upper Kaiti WRUA was far more effective in involving its members in all aspects of catchment
management. Furthermore, neither institution was fully addressing catchment degradation, with Kenze CFA facing greater difficulties in catchment management than Upper Kaiti WRUA. The study concludes by recommending new governance structures and operating frameworks for the Kenze CFA and the Upper Kaiti WRUA to ensure that these organizations effectively carry out their statutory functions. It also provides empirical evidence that can be used to develop strategies for reviewing the activities of these organizations.
The CFA, in particular, should be given the authority to carry out all of its legal obligations, with a focus on the timely preparation and implementation of management plans, whereas the WRUA should be continuously upgraded to ensure its effectiveness in carrying out its duties. These findings also indicate that to halt the catchment's degradation, it is critical to address the CFA difficulties, improve the adoption of their opportunities, and address all of their sustainability challenges.
The study recommends creating a new, open, and transparent governance strategy for CFA and WRUA structures to ensure institutional capacity, as well as developing new initiatives to improve CFA and WRUA activities and operations, addressing CFA and WRUA challenges, and capitalizing on opportunities to ensure long-term catchment management.
2023-01-01T00:00:00ZUsing Transfer Learning to Leverage Large Un-labelled Datasets to Improve Classification Models in Cases With Small- Labelled Datasets: Application to Paediatric Diagnostic and Prognostic Models
http://erepository.uonbi.ac.ke/handle/11295/164302
Using Transfer Learning to Leverage Large Un-labelled Datasets to Improve Classification Models in Cases With Small- Labelled Datasets: Application to Paediatric Diagnostic and Prognostic Models
Mwaniki, Paul M.
Diagnostic and prognostic models based on machine learning models can improve diagnosis and
identification of patients at risk of adverse health outcomes. Healthcare delivery can thus be
improved in low and middle income countries (LMIC) settings, where making accurate diagnosis
remains a challenge because of lack of essential laboratory tests and trained medical staff. Training
machine learning models requires large labelled datasets which are often unavailable in LMICs.
Moreover, models developed in say high-income settings/countries may not be generalizable to
LMICs because of differences in setting/context where underlying data was collected. Transfer
learning, which stores knowledge gained in solving one problem and incorporates that knowledge
while solving a different but related problem, can overcome the challenges of training machine
learning models using small labelled datasets. Transfer learning can extract knowledge from large
un-labelled datasets or dataset from a different setting and incorporate that knowledge when training
models using small labelled datasets, making it potentially applicable to settings with sparse or unlabelled
data such as in LMIC. Transfer learning has been applied to natural images and natural
language processing, but performance on healthcare data such as medical images, bio-signals and
tabular datasets (e.g. clinical signs and symptoms) has not been evaluated.
This study evaluates the use of transfer learning in improving the performance of diagnostic and
prognostic models fitted using small labelled datasets. Three types of datasets were evaluated.
Firstly, paediatric chest x-rays were classified into WHO standardized categories for diagnosis of
pneumonia. Secondly, physiological signals from a pulse oximeter were used to predict
hospitalization status, and lastly, tabular data comprising clinical signs and symptoms were used to
predict positive blood culture results (bacteremia). The performance of models fitted with and
without transfer learning were compared for each dataset.
Transfer learning approaches using multi-task learning and pre-trained models (supervised and
unsupervised pre-training) were used to leverage a large chest x-ray dataset from a high income
setting to improve performance of models trained on a small chest x-ray dataset from seven LMICs.
A novel method incorporating annotation from multiple human readers/annotators of chest x-rays is
proposed and evaluated. Self-supervised learning (SSL) methods were used to extract features from
pulse oximeter signals and to initialize end-to-end deep learning models for predicting
hospitalization status (unsupervised pre-training). Features extracted using SSL were used to predict
hospitalization using logistic regression. Finally, deep learning models for predicting bacteremia
using clinical signs and symptoms were compared with logistic regression models. The deep
learning models were either initialized randomly or using weights from auto-encoders
(unsupervised pre-training).
Supervised and unsupervised pre-training improved classification performance of chest x-rays
marginally (accuracy 0.61 vs 0.59 and 0.60 vs 0.59, respectively). Multi-task learning did not
improve classification of chest x-rays, while incorporating annotations from multiple human readers
had higher performance (accuracy 0.62 vs 0.61). Features extracted from pulse oximeter signals
using SSL models were predictive of hospitalization. The AUCs of logistic regression model
trained on features extracted using SSL models were 0.83 and 0.80 for SSL model trained using
labelled data only and SSL model trained using both labelled and unlabelled data, respectively.
End-to-end deep learning models had AUCs of 0.73 when initialized randomly, 0.77 when
initialized using SSL model trained using labelled data only, and 0.80 when initialized using both
labelled and unlabelled pulse oximeter signals. Logistic regression models for predicting positive
blood cultures performed better than deep learning for small training datasets (AUC 0.67 vs 0.62)
and marginally worse for large datasets (AUC 0.70 vs 0.71). Initializing deep learning models using
weights from auto-encoders did not have any effect on performance on models for predicting
bacteremia.
Our results suggest that transfer learning can improve performance of models trained on
homogenous data types such as medical images and bio-signals but may have no effect on a
heterogeneous tabular data. SSL can be is an effective technique for extracting features from biosignals
that could be used to predict various physiological parameters such as respiratory rate. Deep
learning models perform worse than logistic regression in predicting bacteraemia using clinical
signs and symptoms when the dataset is small.
2023-01-01T00:00:00ZEcology of Immature Stages of the Dengue Fever Vector Aedes Aegypti (L.) (Diptera: Culicidae) in Rural and Urban Sites of the Southern Coast of Kenya
http://erepository.uonbi.ac.ke/handle/11295/164298
Ecology of Immature Stages of the Dengue Fever Vector Aedes Aegypti (L.) (Diptera: Culicidae) in Rural and Urban Sites of the Southern Coast of Kenya
Ngugi, Harun N.
Aedes aegypti is the most important vector of dengue fever and several other arboviruses of
public health such as Zika and Chikungunya. Currently vector management is the only available
option for disease control. Efficient vector control and development of meaningful surveillance
methods depends on a good understanding of vector ecology of which little is known in Kenya.
The objectives of this study were to characterize breeding habitats of Ae. aegypti, determine
seasonal distribution and abundance of Aedes aegypti larvae and pupae in rural and urban sites in
coastal Kenya, identify households that are consistently productive for Ae. aegypti pupae and to
determine susceptibility of Aedes aegypti larvae to the biological larvicide Bacillus thuringiensis
var israelensis (Bti). Entomological, demographic and environmental data was collected from
twenty sentinel households once a month for 24 months (June 2014 to May 2016) in the rural
and urban sites of southern coast of Kenya. All water holding containers in and around houses
were inspected for Ae aegypti larvae and pupae and oviposition traps set weekly in the study
households. Susceptibility of Ae aegypti larvae to a biocontrol agent Bacillus thuringiensis var
israelensis (Bti) was evaluated. Of the 6,566 container visits, only 5.11%, were found positive
for Ae aegypti immatures in the study sites. In both sites significantly more Ae aegypti positive
wet containers were found outdoors than indoors. The most important containers were buckets,
drums and tyres which produced over 70% of all the immatures in both sites. The median
number of months in which pupae were observed in households was 4 and ranged from 0 to
15.The strongest risk factor for pupal abundance was presence of high habitat counts (OR = 1.27,
95% CI 1.00-1.60). Initial efficacy results showed that Bacillus thuringiensis AM65-52 WG
formulation eliminated 100% of larvae in 24 hours. The results of this study indicate that
breeding habitats of Ae. aegypti are abundant outdoors , but, only a few containers are
productive. Further, Ae. aegypti pupal persistence at the household level in urban and rural sites
was observed. High counts of breeding containers was associated with increased risk of pupal
abundance.in households. Targeting productive containers and households that exhibit high
pupal abundance and persistence in vector control interventions may result into costeffective
management of the dengue vector and arboviral transmission in this region.
2023-01-01T00:00:00Z