Explore Our Work
Check out some of our key projects that demonstrate our skills and creativity. These works underline our dedication to quality and our clients’ aspirations.
Explore Some Of Our Publications
In our research journey within the field of Artificial Intelligence, we have focused on developing innovative solutions that address pressing challenges across various domains. These publications not only demonstrate our commitment to advancing AI technology but also provide insights into the practical applications of our research, reflecting our ongoing dedication to enhancing efficiency and accuracy in real-world scenarios. Through these efforts, we aim to contribute significantly to the growing body of knowledge in AI and its transformative potential.
This research develops a reliable, bias-free system for detecting Covid-19 from chest X-ray images. We propose a semi-supervised learning approach using the FlexMatch algorithm to combine two datasets, effectively utilizing unlabelled data to enhance accuracy and convergence. The method employs the ResNet50 architecture for end-to-end training on both labelled and unlabelled data for multi-class classification.
Authors: Mutyambizi Mabasa Nyasha, Xinzhong Zhu, Rusike Samantha, Khondowe Enock
Electrocardiograms (ECGs) are essential noninvasive tools for diagnosing cardiovascular diseases. However, the rising demand for ECG interpretation has led to a shortage of trained cardiologists. Existing automated classification techniques often lack accuracy in categorizing cardiovascular diseases, making the development of reliable ECG diagnosis systems crucial. This work introduces a One-Dimensional Convolutional Neural Network (1D-CNN) for ECG classification, featuring three 1D-CNN layers, a fully connected layer, and a SoftMax layer for improved accuracy in identifying abnormalities.
Authors: Khondowe Enock, Mutyambizi Mabasa Nyasha, Xinzhong Zhu, Rusike Samantha
This paper presents a web-based automated attendance system for multiple students using deep learning for face recognition. Built with dlib’s face recognition library and the Django framework, the system enables teachers to register students and manage attendance via video, while recording it in an Excel file. It is designed for easy user monitoring and maintenance to ensure consistent functionality. Testing shows the system is a more accurate and efficient method for attendance tracking.
Authors: Mutyambizi Mabasa Nyasha, Khondowe Enock, Rusike Samantha
Project Portfolio
Discover some of our latest projects, showcasing creativity and innovation.
Meet Our Skilled Team
Based in Zimbabwe, our company proudly features a diverse team of talented developers and designers, with additional team members residing in Zambia and China. This unique mix of backgrounds enriches our projects, allowing us to leverage a wide range of expertise. Many of our developers also contribute to notable industries, bringing valuable insights and best practices to our work. Together, we foster a collaborative environment where creativity flourishes, enabling us to deliver innovative solutions tailored to our clients’ needs.
