Cesilia Mambile

Cesilia Mambile

Auto-generated profile, based on publicly available information

Cesilia Mambile

Department of Information Systems and Technology, University of Dodoma, Dodoma, Tanzania

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 December 2024 in IEEE Access
Verified
Journal Article
Open access
CMCesilia Mambile
Cesilia Mambile
Shubi Felix KaijageShubi Felix Kaijage
Shubi Felix Kaijage
JLJudith Leo
Judith Leo
Forests are among the world’s most valuable ecological resources. However, they face significant threats from Forest Fires (FFs), causing ecological damage and impacting wildlife and economies. The increasing global occurrence of FFs has created an urgent need for more accurate prediction methods. Traditional FF prediction approaches, reliant on meteorological data and human expertise, are often limited in accuracy and scalability. Deep Learning (DL) offers a promising solution for enhancing prediction capabilities. This systematic review evaluated various DL techniques for FF prediction, analyzing their methodologies, effectiveness, and challenges. Covering studies published between January 2017 and July 2024, 55 of 656 papers were selected for detailed analysis. The study revealed that Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) models are the most frequently used, with most datasets being publicly available. These findings indicate that classification models and simulation-based studies dominate the field. Commonly used metrics include accuracy, precision, recall, F1 score, and Area Under the Receiver Operating Characteristic Curve (AUC-ROC). Key meteorological features, such as Temperature, Humidity, and Wind speed, have been extensively studied using the Normalized Difference Vegetation Index (NDVI), Land Surface Temperature (LST), and Normalized Difference Moisture Index (NDMI), the most commonly used satellite-derived features. However, integrating human activity data remains underexplored despite its potential to improve prediction accuracy significantly. Addressing this gap could enhance the practical applicability of DL models for FF predictions. This study provides insights into the most prevalent and effective DL techniques for FF prediction and highlights areas for future research.

1 August 2024 in Technological Sustainability
Verified
Journal Article
Closed access
CMCesilia Mambile
Cesilia Mambile
AMAugustino Jonas Mwogosi
Augustino Jonas Mwogosi
Purpose: This study aims to explore AI’s potential impact on the broader landscape of higher education in Tanzania. This study contributes to the ongoing discussion of AI’s potential to transform higher education and highlights the ethical considerations and challenges that must be addressed to ensure its successful implementation. This study informs future research and policy decisions in education and technology by providing a detailed understanding of AI’s perceived benefits and challenges in higher education.Design/methodology/approach: A mixed-methods approach was used, which involves collecting and analyzing quantitative and qualitative data to understand the research problem comprehensively. This approach allowed data triangulation and led to a more robust and detailed understanding of this study.Findings: In this study, it was discovered that enhanced assessment, time-saving, personalized learning, improved accessibility and detecting cheating are the perceived benefits of AI as a tool for enhancing learning in higher education, while cost and infrastructure, academic misconduct, data privacy and security, bias and ethical concerns and lack of human interaction are the perceived challenges of AI as a tool for enhancing learning in higher education. Further, it was revealed that students are more accepting of using AI tools in the classroom because they think they are more effective and engaging. On the other hand, faculty were more cautious and skeptical about employing AI tools in the classroom because they worried about how it would affect their teaching methods and job security.Research limitations/implications: The data collection was not conducted face-to-face. To fully capture respondents’ emotional responses, feelings, facial expressions, reactions, or body language was challenging. However, a sufficient number of individuals who participated were very cooperative, and their knowledge was very beneficial in understanding the topic.Originality/value: A unique view of this study is a clear understanding of the perceived benefits and challenges of using AI as a tool for enhancing learning in higher education, as well as the variations in these perceptions among students and faculty. By examining the perspectives of both groups, this study provides a comprehensive understanding of the complex role of AI in higher education. Understanding the broader implications of AI in higher education can inform policy decisions and ensure that AI is used responsibly and ethically.

1 June 2025 in Natural Hazards Research
Verified
Journal Article
Open access
CMCesilia Mambile
Cesilia Mambile
SKShubi Kaijage
Shubi Kaijage
JLJudith Leo
Judith Leo

1 May 2025 in Digital Health
Verified
Journal Article
Open access
AMAugustino Jonas Mwogosi
Augustino Jonas Mwogosi
CMCesilia Mambile
Cesilia Mambile
Objective: Digital ecosystems integrate interconnected tools, platforms, and stakeholders to support healthcare delivery. While prior research has examined their technological architecture, limited attention has been paid to their role in enhancing communication and collaboration, particularly in low-resource settings. This scoping review explores how digital ecosystems facilitate these functions, identifying key components, applications, barriers, and enablers. Methods: Following the Arksey and O’Malley framework, enhanced by Levac et al., a scoping review was conducted using Scopus, PubMed, JSTOR, and Emerald. The review included studies published between 2009 and 2024 and adhered to PRISMA-ScR reporting standards. A total of 15 eligible studies were analysed thematically. Results: Five core ecosystem components emerged: interoperability, stakeholder diversity, patient-centred tools, innovative platforms, and knowledge sharing. Ecosystems supported real-time communication, improved service coordination, and enhanced care accessibility through tools such as telemedicine and mHealth. In low-resource settings, barriers included fragmented digital infrastructure, privacy regulations, and provider resistance. Enablers included strong stakeholder engagement, supportive policies, and low-bandwidth digital solutions like SMS-based platforms. Conclusion: This review highlights the communicative and collaborative functions of digital ecosystems as central to their value in healthcare not just their technological capacity. It contributes a contextual perspective by detailing implementation challenges and adaptations in low- and middle-income countries. To advance the field, future research should adopt standardized frameworks, expand empirical studies in underrepresented settings, and emphasize inclusive, scalable designs that respond to local healthcare realities.

25 November 2025 in BMC Medical Education
Verified
Journal Article
Open access
PDF file
AMAugustino Jonas Mwogosi
Augustino Jonas Mwogosi
CMCesilia Mambile
Cesilia Mambile
RSRajabu Simba
Rajabu Simba
HHHussein Haruna
Hussein Haruna
Integrating digital technologies in clinical dentistry training is increasingly recognized as essential for modernizing competency-based education. However

21 April 2025 in Digital Health
Verified
Journal Article
Open access
AMAugustino Jonas Mwogosi
Augustino Jonas Mwogosi
CMCesilia Mambile
Cesilia Mambile
Background: Adopting electronic health records (EHR) is critical for improving healthcare efficiency, yet implementation remains inconsistent, particularly in developing countries. This study applies a scoping review approach to evaluate the macro-environmental factors influencing EHR adoption in Tanzania using the PEST (Political, Economic, Social, and Technological) analysis framework. Objectives: This study explores the adoption of EHR systems in Tanzanian healthcare using the PEST analysis framework to identify macro-environmental factors influencing implementation. Methods: A scoping review methodology was employed, examining 35 studies published between 2010 and 2024. A comprehensive search was conducted in Scopus, PubMed, JSTOR, and Emerald. Reference lists of relevant studies were also reviewed for additional sources. Data were systematically extracted using a standardised framework categorising studies under political, economic, social, and technological domains. Thematic synthesis was applied to identify key patterns. Results: The review included 35 studies, highlighting that political inconsistencies, economic constraints, social resistance, and technological limitations hinder EHR adoption. Key barriers include weak policy enforcement, unsustainable funding, workforce resistance, and poor ICT infrastructure. However, potential solutions such as strengthening governance frameworks, leveraging public–private partnerships, targeted digital literacy training, and improving ICT infrastructure were identified. Conclusion: This study provides valuable insights into the macro-environmental challenges affecting the adoption of EHR systems in Tanzania. Addressing these barriers through a comprehensive policy and infrastructure development strategy is critical to ensuring successful digital health transformation.