Trouble viewing this mail? View in browser

Half Year Newsletter

Hello Broadcast 👋,

Welcome back to DSAIL Mail! I hope you get to enjoy the read on what we have been up to in the last couple of months.

Data Science Africa (DSA) 2025

We are pleased to report on the highly valuable experience that the DSAIL team had of attending the Data Science Africa (DSA) 2025 conference hosted at the University of Ibadan, Nigeria. The conference ran from Monday, June 2nd, to Thursday, June 5th, 2025, concluding a day early to accommodate the Eid al-Fitr holiday. It provided an excellent opportunity to engage with cutting-edge research, practical applications, and the vibrant African and international data science community. The theme of this year's event was “Responsible Data Science: A Practical Path for Sustainable Development.”

IndabaX Kenya AI Summit

Dr. Joseph Muguro, Research Affiliate and Mary Kariuki, Research Intern at Center for Data Science and Artificial Intelligence (DSAIL), attended IndabaX Kenya AI summit held at Maseno University from 18th June to 20th June 2025, The summit brought together AI enthusiasts, practitioners and researchers from different regions in Kenya. The event offered an enriching experience for networking, learning, and exchanging ideas with AI professionals and researchers. Read more here

During the summit, Mary had an opportunity to present her work under the African research showcase on Building a Health Dataset for Low-Resource Kenyan Languages, an ongoing project at DSAIL. Her project stood out among the best and earned her the position of second runner-up. Dr Joseph Muguro led an engaging workshop titled Futuristic AI: Innovation for Safety and Development, which explored the use of Natural Language Processing (NLP) for safer roads, drawing insights from Kenya's transportation studies.

African-Next-Voices Data Collection Workshop

On 10th April 2025, we hosted a workshop for the African Next Voices project, the project aims to establish a corpus of text and voice data for five kenyan languages Dholuo, Gikuyu,Somali, Kalenjin and maasai. The workshop featured three language groups: Maasai, Kikuyu, and Somali. The participants were trained on transcription techniques using the ThiNK app . This dataset will support the development of language-based technologies such as translation tools, speech recognition and digital assistants.

DSAIL AI in Health Workshop

On 30th May 2025, Center for Data Science and Artificial Intelligence(DSAIL) held a workshop on the application of AI in health, that brought together health workers, language experts, researchers and the general public.The workshop focused on effective use of AI technology in healthcare including ethical and privacy concerns in handling health data and Kenya’s readiness in adopting AI In healthcare. The insights and contribution shared during the workshop by the participants will play a significant role in ongoing work of Building a Health Dataset for Low-Resource Kenyan Languages.

TAHMO Workshop

The Centre for Data Science and Artificial Intelligence (DSAIL) at Dedan Kimathi University of Technology played a pivotal role in the two-day TAHMO co-design workshop held in Nairobi in March 2025. This workshop, organized by the Trans-African Hydro-Meteorological Observatory (TAHMO) in collaboration with the Kenya Meteorological Department, aimed to evaluate and enhance quality assurance and control (QA/QC) practices across Automated Weather Station (AWS) networks. DSAIL's participation underscored its commitment to leveraging data science and artificial intelligence to improve the reliability and accuracy of meteorological data, which is crucial for climate resilience and economic development across Africa. The collaborative nature of the workshop allowed DSAIL to engage with other stakeholders in co-creating innovative solutions tailored to the African context

DSAIL at IST-Africa Conference

We had the opportunity to attend IST-Africa 2025 at the Edge Convention Center in Nairobi. It is a pan-African conference that brings together policymakers, researchers, innovators, and industry leaders to explore how ICT, research, and digital innovation are shaping sustainable development across the continent. The sessions offered insights into strengthening Africa’s digital innovation ecosystems, with themes ranging from online education and AI to energy systems and applied research.

The team representing DSAIL had presentations. Here are a few presentation shots taken during their presentations.

•

Samuel Mbatia Gachana — Using Deep Learning techniques to monitor reforested stands in Kenya

•

Teofilo Ligawa - Leveraging AI Models for Regional Weather Prediction: A Data Pipeline for Africa

•

Victor Ruto — Camera Trap Data Analysis For Grevy’s Zebra Monitoring

•

Paul Bett — Deployment and Evaluation of the Autonomous Monitoring of Insects (AMI) System for Ecosystem Health Assessment in Kenya

Open Hardware Summit 2025

Gabriel Kiarie attended the 2025 Open Hardware Summit that was held at the University of Edinburgh, Edinburgh, Scotland, from 30th to 31st May 2025. Gabriel was a 2023 Open Hardware Summit Fellow. During the 2025 Open Hardware Summit, Gabriel gave a talk on Developing Low-Cost, Open-Source Hardware for Ecosystems Monitoring in the African Context. The talk described some of the open-source hardware developed at DSAIL.

EXCLAIM Symposium and Swiss National Supercomputing Centre (CSCS) visit

Between 2nd and 4th June 2025, Austin and Teofilo had the privilege of attending the 2025 EXCLAIM Symposium that was held at ETH Zürich, Switzerland. This conference focused on the intersection of AI and climate science. The symposium brought together researchers, data scientists, climate modellers, and technologists all working toward one goal: making Earth system science more accurate, equitable, and actionable using artificial intelligence. They also had the privilege of visiting the Swiss National Supercomputing Centre (CSCS) offices in Lugano, where we toured the machine room and learnt how the machines work from the server side and how the supercomputers are maintained—including their innovative lake water cooling system. During the symposium, they had the opportunity to present a poster with some of our preliminary work and got some feedback on it. Read more here

DSAIL Publications 📚

Self-Supervised Multi-Task Learning for the Detection and Classification of RHD-Induced Valvular Pathology

We are excited to share our latest publication, which presents a self-supervised multi-task learning approach for automated echocardiographic analysis in the context of rheumatic heart disease (RHD). Our method enables simultaneous learning of view classification and RHD pathology detection tasks by leveraging unlabelled echo data to overcome annotation scarcity. The results demonstrate strong performance in identifying standard echocardiographic views and detecting RHD-related abnormalities, with accuracy and F1 scores comparable to expert cardiologists. This multi-task framework enables scalable RHD detection with minimal supervision while maintaining diagnostic precision. For the full paper and detailed findings, visit https://doi.org/10.3390/jimaging11040097

Comparing point counts, passive acoustic monitoring, citizen science and machine learning for bird species monitoring in the Mount Kenya ecosystem

A recent paper, Comparing point counts, passive acoustic monitoring, citizen science and machine learning for bird species monitoring in the Mount Kenya ecosystem, by Prof Ciira Maina from DSAIL and Dr Peter Njoroge from the ornithology section at the National Museums of Kenya has been published in the Philosophical Transactions of the Royal Society B. An accompanying dataset, Ndege Zetu, of over 20 hours of audio recordings of bird species obtained from the Mt Kenya ecosystem and annotated by expert ornithologists has been released on Dryad.

It was great catching up 😊.

Subscribe to our newsletter here to stay updated on what we do.

Dedan Kimathi University of Technology - Nyeri, Kenya