Alice Bagyiereyele Lakyiere
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Alice Bagyiereyele Lakyiere has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Nutifafa Yao Agbenor-Efunam
Department of Physics, Kwame Nkrumah University of Science and Technology , Kumasi, Ashanti Region
2 shared publications - Rose-Mary Owusuaa Mensah Gyening
Kwame Nkrumah University of Science and Technology
2 shared publications - Akyana Britwum
Department of Physics, College of Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
1 shared publication - Edmund Ilimoan Yamba
Department of Meteorology and Climate Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
1 shared publication - Franklin Asiedu-Bekoe
Ghana Health Service
1 shared publication Gaddafi Abdul-Salaam
Kwame Nkrumah University of Science and Technology
1 shared publication
Latest Publications
- Open Access
- 2 November 2026
- Published by Oxford University Press (OUP) in Oxford Open Digital Health
MOSQMIXERNET: A CUSTOM DEEP LEARNING ARCHITECTURE FOR EARLY-STAGE MOSQUITO CLASSIFICATION AND GENUS-LEVEL IDENTIFICATION
Mosquito-borne diseases such as malaria, dengue, zika, and chikungunya impose a disproportionate heavy public health burden on Africa. While artificial intelligence (AI) tools have been applied in species identification on the adult mosquitoes, the early aquatic stages have been largely ignored. Yet, these early stages are critical since vector control strategies often target larvae and pupae to break the transmission cycle before mosquitoes mature into biting adults. Moreover, the identification of these specimens is prone to human error by experts due to the striking similarity in morphological features between larval and pupal stages across species. This limitation delays surveillance and increases the risk of misclassification, potentially compromising vector control efforts. In response, an Africa-based curated dataset is introduced, encompassing the larval and pupal stages of Aedes, Anopheles, and Culex species, along with non-mosquito specimens of similar morphology. This dataset helps address the underrepresentation of African samples in global repositories and provides a rare contribution to pupal-stage classification, a category largely overlooked in most classification tasks. MosqMixerNet is introduced as a lightweight deep learning model designed for image-based classification of mosquito early stages. The proposed MosqMixerNet balances compactness and accuracy with only 169 666 trainable parameters and was benchmarked against MobileNetV2, DenseNet121, and NASNetMobile. For genus-level identification, it achieved 99.37% accuracy, surpassing NASNetMobile (97.30%), MobileNetV2 (97.26%), and DenseNet121 (80.39%). The work advances computer vision for public health and tropical disease control, and provides a foundation for integration into digital vector monitoring platforms in endemic settings.
- 2 November 2026
- Published by Oxford University Press (OUP) in Oxford Open Digital Health
Open AccessMOSQMIXERNET: A CUSTOM DEEP LEARNING ARCHITECTURE FOR EARLY-STAGE MOSQUITO CLASSIFICATION AND GENUS-LEVEL IDENTIFICATION
Mosquito-borne diseases such as malaria, dengue, zika, and chikungunya impose a disproportionate heavy public health burden on Africa. While artificial intelligence (AI) tools have been applied in species identification on the adult mosquitoes, the early aquatic stages have been largely ignored. Yet, these early stages are critical since vector control strategies often target larvae and pupae to break the transmission cycle before mosquitoes mature into biting adults. Moreover, the identification of these specimens is prone to human error by experts due to the striking similarity in morphological features between larval and pupal stages across species. This limitation delays surveillance and increases the risk of misclassification, potentially compromising vector control efforts. In response, an Africa-based curated dataset is introduced, encompassing the larval and pupal stages of Aedes, Anopheles, and Culex species, along with non-mosquito specimens of similar morphology. This dataset helps address the underrepresentation of African samples in global repositories and provides a rare contribution to pupal-stage classification, a category largely overlooked in most classification tasks. MosqMixerNet is introduced as a lightweight deep learning model designed for image-based classification of mosquito early stages. The proposed MosqMixerNet balances compactness and accuracy with only 169 666 trainable parameters and was benchmarked against MobileNetV2, DenseNet121, and NASNetMobile. For genus-level identification, it achieved 99.37% accuracy, surpassing NASNetMobile (97.30%), MobileNetV2 (97.26%), and DenseNet121 (80.39%). The work advances computer vision for public health and tropical disease control, and provides a foundation for integration into digital vector monitoring platforms in endemic settings.
- Open Access
- 19 February 2026
- Published by Elsevier in SSRN Electronic Journal
A Low-Cost IoT Acoustic and Environmental Sensing System for Real-Time Mosquito Surveillance in Resource-Limited Regions
Vector-borne diseases such as malaria and dengue cause over 700,000 deaths annually, disproportionately affecting resource-limited regions. Effective mosquito surveillance is critical but often hindered by infrastructure limitations and high equipment costs. This paper presents MosquitoTrack, a low-cost, modular IoT system designed for real-time mosquito monitoring using acoustic and environmental data. The system captures mosquito wingbeat signatures via an INMP441 MEMS microphone and records synchronized temperature, humidity, and GPS coordinates using an ESP32 microcontroller, DHT22 sensor, and A9G module. Audio and environmental data are time-stamped at 5-second intervals to maintain contextual alignment. MosquitoTrack supports configurable sampling rates (8 - 44.1 kHz), multiple operation modes, and functions offline with a flexible input voltage (5 - 30 V). During validation, the device recorded ambient conditions of 29.89°C ± 0.66°C and 49.15% ± 1.62% relative humidity. FFT and spectrogram analyses identified Aedes mosquito wingbeat frequencies between 550 - 900 Hz, consistent with known patterns. Initial recordings yielded an average SNR of 1.80 dB, which improved to 4.68 dB with bandpass filtering (300 - 1100 Hz), and up to 18.02 dB after Wiener noise reduction. These results demonstrate MosquitoTrack’s effectiveness in isolating mosquito acoustic signals in noisy environments. With offline functionality and accessible data storage, the device is well suited for mobile Health initiatives, citizen science, and education, empowering communities to participate in vector control efforts.
- 19 February 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessA Low-Cost IoT Acoustic and Environmental Sensing System for Real-Time Mosquito Surveillance in Resource-Limited Regions
Vector-borne diseases such as malaria and dengue cause over 700,000 deaths annually, disproportionately affecting resource-limited regions. Effective mosquito surveillance is critical but often hindered by infrastructure limitations and high equipment costs. This paper presents MosquitoTrack, a low-cost, modular IoT system designed for real-time mosquito monitoring using acoustic and environmental data. The system captures mosquito wingbeat signatures via an INMP441 MEMS microphone and records synchronized temperature, humidity, and GPS coordinates using an ESP32 microcontroller, DHT22 sensor, and A9G module. Audio and environmental data are time-stamped at 5-second intervals to maintain contextual alignment. MosquitoTrack supports configurable sampling rates (8 - 44.1 kHz), multiple operation modes, and functions offline with a flexible input voltage (5 - 30 V). During validation, the device recorded ambient conditions of 29.89°C ± 0.66°C and 49.15% ± 1.62% relative humidity. FFT and spectrogram analyses identified Aedes mosquito wingbeat frequencies between 550 - 900 Hz, consistent with known patterns. Initial recordings yielded an average SNR of 1.80 dB, which improved to 4.68 dB with bandpass filtering (300 - 1100 Hz), and up to 18.02 dB after Wiener noise reduction. These results demonstrate MosquitoTrack’s effectiveness in isolating mosquito acoustic signals in noisy environments. With offline functionality and accessible data storage, the device is well suited for mobile Health initiatives, citizen science, and education, empowering communities to participate in vector control efforts.
- Open Access
- 1 February 2026
- Published by SAGE Publications in Digital Health
- Vol. 12
Artificial intelligence in uterine fibroid imaging: A review of trends, performance, and generalizability
The uterine fibroid is the most prevalent benign gynecologic tumor, affecting up to 80% of women by the age of 50 and serving as the major reason for hysterectomies, infertility, and reproductive morbidity globally. While extensive literature exists on its management, the disease still poses challenges due to its heterogeneous diagnostic criteria, unpredictable growth pattern, and varying treatment efficacy, prompting increased focus on AI techniques for clinical decision-making. This review highlights the recent advancements in AI applications for fibroid management through a narrative review of 39 publications from 2021 to 2026 in five categories: detection and segmentation, differentiation, prediction of high-intensity focused ultrasound (HIFU) treatment results, surgery planning and intraoperative guidance, and monitoring growth and recurrence. Radiomic analysis, machine learning, and deep learning methods have shown strong performance in lesion segmentation and treatment outcome prediction across several retrospective, single-center studies. Nonetheless, most of the research was conducted retrospectively, from single centers, with poor external validation, little emphasis on intraoperative assistance, long-term follow-up, and model explainability. The geographic imbalance in favor of developed regions casts severe ethical concerns. Of the 39 studies reviewed, only one originated from an African institution, a critical gap given that women of African descent bear a disproportionately higher fibroid burden globally. This near-total absence of African clinical data risks producing AI models that perform inequitably in the populations that need them most. Realising AI’s potential in fibroid care demands multicenter prospective trials, standardised methodologies, and deliberate investment in underrepresented regions, particularly sub-Saharan Africa.
- 1 February 2026
- Published by SAGE Publications in Digital Health
- Vol. 12
Open AccessArtificial intelligence in uterine fibroid imaging: A review of trends, performance, and generalizability
The uterine fibroid is the most prevalent benign gynecologic tumor, affecting up to 80% of women by the age of 50 and serving as the major reason for hysterectomies, infertility, and reproductive morbidity globally. While extensive literature exists on its management, the disease still poses challenges due to its heterogeneous diagnostic criteria, unpredictable growth pattern, and varying treatment efficacy, prompting increased focus on AI techniques for clinical decision-making. This review highlights the recent advancements in AI applications for fibroid management through a narrative review of 39 publications from 2021 to 2026 in five categories: detection and segmentation, differentiation, prediction of high-intensity focused ultrasound (HIFU) treatment results, surgery planning and intraoperative guidance, and monitoring growth and recurrence. Radiomic analysis, machine learning, and deep learning methods have shown strong performance in lesion segmentation and treatment outcome prediction across several retrospective, single-center studies. Nonetheless, most of the research was conducted retrospectively, from single centers, with poor external validation, little emphasis on intraoperative assistance, long-term follow-up, and model explainability. The geographic imbalance in favor of developed regions casts severe ethical concerns. Of the 39 studies reviewed, only one originated from an African institution, a critical gap given that women of African descent bear a disproportionately higher fibroid burden globally. This near-total absence of African clinical data risks producing AI models that perform inequitably in the populations that need them most. Realising AI’s potential in fibroid care demands multicenter prospective trials, standardised methodologies, and deliberate investment in underrepresented regions, particularly sub-Saharan Africa.