Related Experiment Video
Updated: Aug 10, 2026
03:22
Extraction of Saliva, Haemolymph, Salivary Glands, and Midgut from Individual Ticks (Acari: Ixodidae)
Published on: October 31, 2025
AfriTickID: An AI-assisted image-based approach for tick species identification in Kenya
Caroline Kioko1, Hao Cheng1, Joy Ekeya2
1ITC Faculty Geo-Information Science and Earth Observation, University of Twente, Enschede, Netherlands.
Acta Tropica
|August 4, 2026
Summary
Artificial Intelligence (AI) can automate tick identification from images, aiding in controlling tick-borne diseases. A study in Kenya showed MobileNetV2 AI models achieved 75% accuracy in identifying seven tick species.
Area of Science:
- Veterinary Entomology
- Computational Biology
- Public Health
Background:
- Tick-borne diseases present significant risks to human and animal health.
- Accurate tick species distribution mapping is crucial for effective disease control strategies.
Purpose of the Study:
- To develop and evaluate Artificial Intelligence (AI) models for automated tick species identification from images.
- To establish a baseline for AI-assisted tick identification in Kenya.
Main Methods:
- A dataset of 864 tick images from seven species in Kenya was created.
- MobileNetV2 and ResNet-50 deep learning models were trained and evaluated.
- Five-fold cross-validation and normalized confusion matrices were used to assess performance.
Main Results:
- The MobileNetV2 model demonstrated superior performance with 75% accuracy, 74% weighted F1-score, and 69% Kappa score.
- MobileNetV2 achieved high identification rates for Rhipicephalus decoloratus (88%) and Rhipicephalus sanguineus (87%).
- ResNet-50 showed better performance for Rhipicephalus sanguineus (94%) but lower rates for other species.
Conclusions:
- AI models, particularly MobileNetV2, show promise for automated tick identification in Kenya.
- Expanding datasets and external validation are necessary for broader applicability and improved accuracy across all tick species.
