A5084580547- Natural Language Processing Techniques
- Artificial Intelligence in Law
- Data Quality and Management
- Topic Modeling
- Scientific Computing and Data Management
- Language, Linguistics, Cultural Analysis
- Research Data Management Practices
- Viral Infections and Outbreaks Research
- Interactive and Immersive Displays
- Artificial Intelligence in Healthcare
- Interpreting and Communication in Healthcare
- Language, Discourse, Communication Strategies
- Usability and User Interface Design
- AI in Service Interactions
- COVID-19 epidemiological studies
- Fire Detection and Safety Systems
- COVID-19 diagnosis using AI
- COVID-19 Digital Contact Tracing
- Legal Language and Interpretation
- COVID-19 Pandemic Impacts
- Data-Driven Disease Surveillance
- Multimodal Machine Learning Applications
- Speech and dialogue systems
- Digital Accessibility for Disabilities
- Algorithms and Data Compression
Malawi University of Business and Applied Sciences
2020-2025
University of Malawi
2020-2025
Colorado College
2014
John D. and Catherine T. MacArthur Foundation
2001
Research Network (United States)
2001
Wherry & Sons (United Kingdom)
1992
University of Notre Dame
1964
David Adelani, Graham Neubig, Sebastian Ruder, Shruti Rijhwani, Michael Beukman, Chester Palen-Michel, Constantine Lignos, Jesujoba Alabi, Shamsuddeen Muhammad, Peter Nabende, Cheikh M. Bamba Dione, Andiswa Bukula, Rooweither Mabuya, Bonaventure F. P. Dossou, Blessing Sibanda, Happy Buzaaba, Jonathan Mukiibi, Godson Kalipe, Derguene Mbaye, Amelia Taylor, Fatoumata Kabore, Chris Chinenye Emezue, Anuoluwapo Aremu, Perez Ogayo, Catherine Gitau, Edwin Munkoh-Buabeng, Victoire Memdjokam Koagne,...
The lack of annotated datasets makes the automatic detection skin problems very difficult, which is also case for most other medical applications. outstanding results achieved by deep learning techniques in developing such applications have improved diagnostic accuracy. Nevertheless, performance these models heavily dependent on volume labelled data used training, unfortunately not available. To address this problem, traditional augmentation usually adopted. Recently, emergence a generative...
Cheikh M. Bamba Dione, David Ifeoluwa Adelani, Peter Nabende, Jesujoba Alabi, Thapelo Sindane, Happy Buzaaba, Shamsuddeen Hassan Muhammad, Chris Chinenye Emezue, Perez Ogayo, Anuoluwapo Aremu, Catherine Gitau, Derguene Mbaye, Jonathan Mukiibi, Blessing Sibanda, Bonaventure F. P. Dossou, Andiswa Bukula, Rooweither Mabuya, Allahsera Auguste Tapo, Edwin Munkoh-Buabeng, Victoire Memdjokam Koagne, Fatoumata Ouoba Kabore, Amelia Taylor, Godson Kalipe, Tebogo Macucwa, Vukosi Marivate, Tajuddeen...
Background Health Passports (HPs) are paper-based, patient-held records used in Malawi to document key details about the health condition of a patient and care provided during medical visits. Aim This paper assessed their use effectiveness within data ecosystem, impact on care. Setting The study setting was facilities Zomba District, Malawi. Methods We undertook descriptive exploratory qualitative determine practices for recording by professionals importance placed HPs patients...
Many of the stochastic models used in inference phylogenetic trees from biological sequence data have polynomial parameterization maps. The image such a map---the collection joint distributions for all parameter choices---forms model space. Since is polynomial, Zariski closure space an algebraic variety which typically much larger than but amenable to study with methods. Of practical interest, however, not full subset formed by Here we develop complete semialgebraic descriptions arising...
The COVID-19 pandemic has spurred the use of AI and DS innovations in data collection aggregation. Extensive on many aspects been collected used to optimize public health response manage recovery patients Sub-Saharan Africa. However, there is no standard mechanism for collecting, documenting disseminating related or metadata, which makes reuse a challenge. INSPIRE utilizes Observational Medical Outcomes Partnership (OMOP) as Common Data Model (CDM) implemented cloud Platform Service (PaaS)...
Introduction Population health data integration remains a critical challenge in low- and middle-income countries (LMIC), hindering the generation of actionable insights to inform policy decision-making. This paper proposes pan-African, Findable, Accessible, Interoperable, Reusable (FAIR) research architecture infrastructure named INSPIRE datahub. cloud-based Platform-as-a-Service (PaaS) on-premises setup aims enhance discovery, integration, analysis clinical, population-based surveys, other...
Advances in speech and language technologies enable tools such as voice-search, text-to-speech, recognition machine translation. These are however only available for high resource languages like English, French or Chinese. Without foundational digital resources African languages, which considered low-resource the context, these advanced remain out of reach. This work details AI4D - Language Program, a 3-part project that 1) incentivised crowd-sourcing, collection curation datasets through an...
Background: The completion of case-based surveillance forms is vital for obtaining case findings during COVID-19 in Malawi. These capture data spanning many attributes used to inform epidemiological analyses, disease descriptions, and indicators. objectives this study were (1) investigate the processes collecting reporting data, (2) explore health workers' perceptions understandings collection tools (3) identify factors contributing discrepancies national data.Methods: A total 75 healthcare...
Background Metadata describe and provide context for other data, playing a pivotal role in enabling findability, accessibility, interoperability, reusability (FAIR) data principles. By providing comprehensive machine-readable descriptions of digital resources, metadata empower both machines human users to seamlessly discover, access, integrate, reuse or content across diverse platforms applications. However, the limited accessibility machine-interpretability existing population health hinder...
<ns3:p>Background The completion of case-based surveillance forms was vital for case identification during COVID-19 in Malawi. Despite significant efforts, the resulting national data suffered from gaps and inconsistencies which affected its optimal usability. objectives this study were to investigate processes collecting reporting data, explore health workers’ perceptions understanding collection tools processes, identify factors contributing quality. Methods A total 75 healthcare...
<sec> <title>BACKGROUND</title> Metadata describe and provide context for other data, playing a pivotal role in enabling findability, accessibility, interoperability, reusability (FAIR) data principles. By providing comprehensive machine-readable descriptions of digital resources, metadata empower both machines human users to seamlessly discover, access, integrate, reuse or content across diverse platforms applications. However, the limited accessibility machine-interpretability existing...
<sec> <title>UNSTRUCTURED</title> </sec>
Abstract Language barriers in healthcare, prevalent globally, often lead to miscommunication between professionals and patients, reducing the quality of care. Recognition, training, formal study these during medical communication African countries is severely limited. Our focused on language healthcare facilities Zomba, an important district Malawi where several local languages are spoken. We employed a qualitative approach conducted questionnaire-based study. Data was gathered from 79 312...
Language barriers in healthcare lead to miscommunication between professionals and patients, thereby reducing the quality of equitable access healthcare. In African countries, recognition formal study these is severely limited despite Africa having more languages than any other continent. Our investigates language facilities Zomba district Malawi, where three major local are spoken.
Over recent years, video‐surveillance systems have seen extensive adoption, largely driven by security imperatives, with radar‐based speed detection being a common feature in traffic monitoring. Despite its prevalence, broader anomaly patterns has not received equivalent focus. This research develops sophisticated deep learning framework, drawing architectural inspiration from MobileNet, ResNet50, and VGG19, to only detect track vehicles but also analyze trajectory data identify nonstandard...