ETH LAB

ETH LABORATORY

MUHAS EMERGING TECHNOLOGIES FOR HEALTH RESEARCH AND DEVELOPMENT LABORATORY

The MUHAS Emerging Technologies for Health (ETH) research and development laboratory is a facility that operates under the Biomedical Engineering unit, Physiology Department at MUHAS School of biomedical sciences. The lab is focus on research and development of healthcare solutions by using emerging technologies with the aim of improving patient care and outcomes. These technologies include Artificial Intelligence, Virtual reality, Augmented Reality, Blockchain etc.

The lab also serves as a learning and training environment in biomedical engineering and related fields by providing short trainings, workshops and mentorships.

MISSION

To research and develop solutions for healthcare by using emerging technologies. Including Artificial Intelligence, Virtual reality, Augmented reality, Blockchain

VISION

Leading research and development laboratory for healthcare technologies in the country, region and on the continent.

Researchers

Three members of the lab have PhD’s in Artificial Intelligence, Signal Processing and Mathematical modelling obtained from the UK, Norway and Japan. Currently other three members are pursuing their PhDs in the same fields in China and German.

The other four members have masters and degrees in Biomedical engineering, with other members having degrees in prosthetic and Orthotic.

Some of the researchers of this lab have won different awards and competitions including “The International Innovation Competition for Automation and UAV 2014”, a competition held in China and IEEE Outstanding Young Investigator Research Visit award. Researchers in this lab have published several papers in peer-reviewed journals including Nature and IEEE. Some of the published works have been featured in global media houses including Forbes and the BBC.
Here is the list of publications that researchers from our Lab have either led or participated in.

Fully portable and wireless universal brain–machine interfaces enabled by flexible scalp electronics and deep learning algorithm.

Decoding movement intent patterns based on spatiotemporal and adaptive filtering method towards active motor training in stroke rehabilitation systems

Towards image-based cancer cell lines authentication using deep neural networks

A new technique for the prediction of heart failure risk driven by hierarchical neighborhood component-based learning and adaptive multi-layer networks

A channel selection approach based on convolutional neural network for multi-channel EEG motor imagery decoding

Efficient channel selection approach for motor imaginary classification based on convolutional neural network

The use of artificial intelligence-based innovations in the health sector in Tanzania: A scoping review

Effect of window conditioning parameters on the classification performance and stability of EMG-based feature extraction methods

Ascertaining the optimal myoelectric signal recording duration for pattern recognition based prostheses control

Determining the optimal window parameters for accurate and reliable decoding of multiple classes of upper limb motor imagery tasks

HIV Viral Dynamics in the Presence of Antiretroviral Treatment

GUBS: Graph-Based Unsupervised Brain Segmentation in MRI Images

Brain tumor segmentation based on minimum spanning tree

Threshold estimation based on local minima for nucleus and cytoplasm segmentation

AI for Improving ZHSF Enrolment

ETH Lab, in partnership with PharmAccess, is implementing an AI-driven initiative to increase enrolment of informal sector workers into the Zanzibar Health Services Fund (ZHSF). The project integrates ZHSF, Matibabu ID, and social media data to identify enrolment gaps and target high-potential groups.

Our team is developing predictive models to map underserved populations and deploying a multilingual WhatsApp and web-based AI chatbot that supports users through registration, payments, renewals, and real-time assistance. The system also provides dashboards that track enrolment trends and digital engagement.

This ongoing project aims to boost ZHSF enrolment, enhance digital service delivery, and provide the Ministry of Health with data-driven insights—advancing scalable and sustainable health financing solutions in Zanzibar.

Kikohozi App – AI Cough Analysis for Respiratory Diseases

ETH Lab is working with Aga Khan University to develop the Kikohozi platform, an AI-powered system that uses cough sounds to help detect TB, asthma, and COPD. Our team is leading the development of the AI models, data pipelines, and mobile/web applications. The project is building Tanzania’s largest cough sound database and creating real-time diagnostic tools to support respiratory disease screening in low-resource settings.

Ongoing Project | Led by Aga Khan University
Technical Lead: MUHAS – ETH Lab

AI Algorithm for Early Detection of Cervical Cancer (Pap Smear Images)

Funder: NIH — Status: Ongoing
Deep-learning methods for classifying cervical cancer lesions in Pap smear images, focusing on women living with HIV.

Virus Image: Electron microscopic image of rabies virus. Courtesy: U.S. Centers for Disease Control and Prevention / Dr. Fred Murphy

Artificial intelligence for Rabies screening and early prediction of its outbreak

We are using artificial intelligence (AI) to detect rabies epidemics which involves the application of machine learning algorithms and techniques to analyze and process data related to rabies outbreaks. The goal is to identify patterns and trends in the data that can help predict, monitor, and respond to rabies epidemics more effectively. This is a joint project between MUHAS and Ifakara Health Institute.
Funder: Lacuna fund

Artificial Intelligence for screening of TB among people living with HIV
Tuberculosis (TB) is a serious and potentially life-threatening disease that can be particularly challenging for people living with human immunodeficiency virus (HIV). People with HIV are at a higher risk of developing TB, and TB can also accelerate the progression of HIV. Therefore, screening and treatment for TB are critical components of care for people living with HIV. At ETH, we are developing an Artificial Intelligence based tool for screening of TB among people living with HIV.

Funder:
IDRC through HASH (Hub for Artificial Intelligence in Maternal, Sexual and Reproductive Health)

PREVIOUS PROJECTS

Zanzibar Matibabu ID – AI Health Intelligence Project

ETH Lab partnered with PharmAccess to build an AI-powered health analytics platform for Zanzibar, using Matibabu ID data covering over 1.7 million people. We integrated clinical, demographic, and facility-quality datasets and developed predictive models for disease trends, medicine needs, chronic conditions, and high-risk groups.

The platform includes an interactive dashboard that equips the Ministry of Health with real-time insights for planning and procurement. The project strengthens evidence-based decision-making, reduces stockouts, and improves health outcomes—showcasing ETH Lab’s capability to deliver impactful AI solutions in public health.

Determining Validity, Optimality and Feasibility of introducing Socio-Economic Status Assessment Questionnaire in Routine Immunization Services in Tanzania.

Using Artificial Intelligence to tackle ‘infodemic’ in Tanzania.
Using Artificial Intelligence to tackle challenges of tourism in Zanzibar

Artificial intelligence for early detection of breast cancer

Following cervical cancer, breast cancer is the second most common cancer and the second leading cause of cancer mortality among women in Tanzania, this is largely due the absence of screening programs and limited access to diagnosis. In this project, we are creating Artificial Intelligence based breast cancer diagnostic system using breast ultrasound and mammogram.

Funders:
• Google
• IDRC Through Villgro Africa
• SIDA through MUHAS