Data science in public health: A review of predictive analytics for disease control in the USA and Africa

World Journal of Advanced Research and Reviews, Journal Year: 2024, Volume and Issue: 21(1), P. 2753 - 2769

Published: Jan. 30, 2024

This scholarly paper delves into the realm of data science in public health, with a specific focus on transformative role predictive analytics disease control across United States and Africa. Set against backdrop rapidly evolving healthcare challenges, study aims to dissect synthesize advancements, applications, hurdles associated data-driven health strategies these diverse geographical contexts. Employing qualitative analysis peer-reviewed literature, meticulously examines evolution analytics, comparing structures, scrutinizing key diseases challenges prevalent both regions. The scope extends exploring ethical considerations technological advancements utilization, offering panoramic view current potential landscape health. findings reveal significant surge application particularly USA for chronic management Africa infectious control. highlights successes implementing policies, emphasizing need balanced approach that addresses technological, ethical, cultural barriers. future AI machine learning is identified as promising domain, further innovation integration policy. Conclusively, recommends continued investment applications advocating collaborative efforts overcome implementation considerations. underscores enhancing delivery, more effective, efficient, equitable systems globally.

Language: Английский

---

A review of deep learning models to detect malware in Android applications

Cyber Security and Applications, Journal Year: 2023, Volume and Issue: 1, P. 100014 - 100014

Published: Feb. 12, 2023

Android applications are indispensable resources that facilitate communication, health monitoring, planning, data sharing and synchronization, social interaction, business financial transactions. However, the rapid increase in smartphone penetration rate has consequently led to an cyberattacks. Smartphone use permissions allow users utilize different functionalities, making them susceptible malicious software (malware). Despite rise applications' usage cyberattacks, of deep learning (DL) models detect emerging malware is still nascent. Therefore, this review sought explain DL applied applications, explore their performance as well identify research gaps present recommendations for future work. This study adopted preferred reporting items systematic reviews meta-analyses (PRISMA) guidelines guide review. The revealed convolutional neural networks, gated recurrent bidirectional long short-term memory, memory (LSTM) cubic-LSTM most prominent learning-based detection applications. findings show increasingly becoming effective technique real-time. monitoring tracking information flow behavior a daunting task because evolving nature human behavior. training mobile application updated datasets paramount developing models. There also need before downloading improve security smartphones.

Language: Английский

---

A Deep Learning-Based Chatbot to Enhance Maternal Health Education

Published: March 7, 2024

Maternal mortality remains a global concern, with resource-constrained countries disproportionately affected due to inherent challenges in such countries, like underfunding, distant health facilities, lack of access maternal education and inequitable services. Though medical chatbots are gaining popularity, lag, there is dearth specific local languages. Therefore, this study utilised natural language processing develop chatbot using feedforward deep neural network. The model was trained three African languages (Sesotho, Shona Ndebele) English, the deployed Flask server through web app present friendly interface users. training evaluation losses reached zero, while accuracies 100%.

Language: Английский

---

Emerging technologies’ role in reducing under-five mortality in a low-resource setting: Challenges and perceived opportunities by public health workers in Makonde District, Zimbabwe

Journal of Child Health Care, Journal Year: 2023, Volume and Issue: unknown

Published: July 18, 2023

Under-five mortality (U5M) remains a global challenge, with Sub-Saharan Africa being the hardest hit. The coronavirus disease 2019 (COVID-19) has strained healthcare systems, threatening to reverse current gains in U5M health outcomes. It threatened progress made towards achieving United Nations Sustainable Development Goal 3 due its strain on resource reassignment and prioritisation by authorities globally. Low-resource settings inherently face unique challenges fighting providing quality under-fives, like understaffing, drug shortages, underfunding, skills gaps lack of specialised equipment, contributing high rates. This study explored public facilities’ reducing low-resource setting Zimbabwe workers’ perceptions emerging technologies’ role addressing those challenges. Twenty workers participated interviews focus group. They perceived technologies (ETs) as panacea supporting data-driven healthcare, improving follow-up outcomes through automated reminders medication clinic visits, aiding diagnosis, continuous monitoring, education, supply critical supplies delivery development. In this paper, technology is any information communication that not been utilised full potential Zimbabwe’s domain. Findings indicate Makonde would welcome ETs improve under-five well-being.

Language: Английский

---

Artificial intelligence for healthcare in Africa: a scientometric analysis

Health and Technology, Journal Year: 2023, Volume and Issue: 13(6), P. 947 - 955

Published: Nov. 1, 2023

Language: Английский

---

Application of machine learning techniques for predicting child mortality and identifying associated risk factors

Published: March 1, 2023

Despite continuous persistent efforts to enhance child health through, among other things, universal access care, mortality remains a significant public concern on global scale. Child is attributed several factors including birth asphyxia/trauma, demographic and socioeconomic factors, preterm intrapartum-related complications, pneumonia, preventable treatable diseases, congenital anomalies, poor quality healthcare, hygiene nutrition, sanitation others. In many sub-Saharan African nations, Zimbabwe, the use of machine learning techniques predict still in its infancy. Therefore, this study applied algorithms decision trees, random forest, logistic regression XGBoost develop predictive models that utilize nationally representative survey data. The classifier achieved an accuracy 74 ^% , forest 72%, Decision tree high 81%. All under-five precision 95 %. However, recall 76%, 84%. Logistic Regression F1-score 84%, 83%, 83% 89% for XGBoost. outperformed models. Integrating such into information systems can significantly assist policymakers healthcare professionals improve status children, care most importantly, preventive measures, immunization programmes, policies, decision-making health. Understanding risk designing intervention programmes aimed at while reducing mortality.

Language: Английский

---

A Framework for the Adoption of Emerging Technologies to Reduce Under-Five Mortality in Zimbabwe

Published: March 1, 2023

Under-five mortality remains a global health concern as many countries have failed to achieve the United Nations Millennium Development Goal 4 (MDG 4). Children under five (under-fives) continue perish preventable deaths globally. Zimbabwe is amongst Sub-Saharan African that MDG on under-five mortality. Regardless of evidence from other regions emerging technologies help eliminate among under-fives, Zimbabwe's adoption such in public facilities nascent. The country has introduced some digital facilities, but they are not specific paediatric care. Likewise, research paid little attention Therefore, this study proposes framework for adopting and utilizing reduce facilities. pragmatism philosophy guided study. It employed sequential exploratory mixed-methods design explore factors affecting perceived role technologies, with aim designing technology framework. Future studies could focus integrating existing systems harness data generated enhance care through information systems.

Language: Английский

---

A Deep Learning Model for Predicting Under-Five Mortality in Zimbabwe

Published: Aug. 3, 2023

The death of children before they reach five years old (under-five mortality or U5M) is a global scourge that has attracted the attention many governments, including World Health Organisation and United Nations. Children under-five in Sub-Saharan Africa are disproportionately susceptible to death, with fifteen-fold likelihood compared their counterparts developed countries. Regardless numerous efforts by Zimbabwean Government improve child health, such as free access care, provision nutritional supplements, immunisation programmes prevention mother-to-child transmission, country still high rates (U5MRs). Zimbabwe's failure reduce U5MRs acceptable levels suggests current methods must be complemented. Identifying contextual risk factors at could help paediatricians make timely targeted interventions policymakers review existing craft new policies save children's lives. Therefore, this study applied deep learning 2019 Multiple Indicator Cluster Survey data predict identify its associated factors. used neural network four hidden layers, k-fold cross-validation stochastic gradient descent (SGD) optimiser. All layers Rectified Linear Unit activation function except output layer, which sigmoid for binary classification. model produced 90.04% accuracy, 92.39% precision, 87.30% recall 95.04% area under curve. Though predicts mortality, it does not prescribe appropriate lives, gap future studies fill.

Language: Английский

---

Machine Intelligence in Africa: a survey

Published: Jan. 18, 2024

In the last 5 years, availability of large audio datasets in African countries has opened unlimited opportunities to build machine intelligence (MI) technologies that are closer people and speak, learn, understand, do businesses local languages, including for those who cannot read write. Unfortunately, these not fully exploited by current MI tools, leaving several Africans out business opportunities. Additionally, many state-of-the-art models culture-aware, ethics their adoption indexes questionable. The lack thereof is a major drawback applications Africa. This paper summarizes recent developments Africa from multi-layer multiscale culture-aware perspective, showcasing use cases 54 through 400 articles on research, industry, government actions, as well uses art, music, informal economy, small survey also opens discussions reliability rankings continent algorithmic definitions unclear terms used MI.

Language: Английский

---

Towards Data-Driven Artificial Intelligence Models for Monitoring, Modelling and Predicting Illicit Substance Use

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 361 - 379

Published: Jan. 1, 2024

Language: Английский

---