Deep Learning Model for Identifying Sandfly Species and Gender
Author Information
Author(s): Fraiwan Mohammad
Primary Institution: Jordan University of Science and Technology
Hypothesis
Can a convolutional neural network improve the accuracy and efficiency of identifying the sex and species of sandflies?
Conclusion
The study developed a deep learning model that achieved over 95% accuracy in classifying the sex and species of three sandfly species.
Supporting Evidence
- The model achieved classification accuracy exceeding 95%.
- Using both genital and pharyngeal images improved classification performance.
- The study addresses the limitations of traditional sandfly identification methods.
Takeaway
This study created a smart computer program that can quickly tell the difference between male and female sandflies, which helps in controlling diseases they spread.
Methodology
The study used a convolutional neural network to analyze pharyngeal and genital images of sandfly samples collected over two years.
Limitations
The dataset size could be increased for better model robustness, and some irrelevant regions were highlighted in the Grad-CAM maps, suggesting potential data leakage.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website