Accepted for/Published in: JMIR Dermatology
Date Submitted: Jan 15, 2026
Date Accepted: May 20, 2026
Image-Based Diagnostic Model for Skin Neglected Tropical Diseases: Development of Benchmark Deep Learning Model through Harmonized Dual Model Architectures Based on the Funnel Framework for Model Architecture Screening
ABSTRACT
Background:
While deep learning (DL)-based methods are the potential technological solutions for the diagnosis of skin Neglected Tropical Diseases (skin NTDs), limited efforts were seen towards the use of such tools in Ethiopia. Data scarcity, methods and models selection issues created further challenges in an attempt to close the previous gap.
Objective:
This study attempts to design image-based diagnostic model for skin NTDs through a synergistic combination of feature extraction, CNN model training on extracted features, and data augmentation.
Methods:
For this study, a new skin images dataset is created using skin photos collected by a team of researchers from NTDs research center of Arba Minch University medical college. The new dataset contains 1495 images in 3 classes having severe class imbalance. Extensive experiments were conducted to find the optimal DL approach by designing a new CNN model, applying transfer learning, using the two-stage approach by training the new CNN model using pretrained models for features extraction, and applying data augmentation based on the two-stage approach. For model selection, the study proposed a novel approach, the funnel framework with cascaded selection of methods and models.
Results:
After hyperparameter tuning, the model trained using DenseNet121 feature extractor scored the highest accuracy of 96.6%, F1-score of 95%, as well as sensitivity of 95%, while the MNv2-based model scored comparable results of 95.6% accuracy, 90% F1-score, and 90% sensitivity. This study finally selected the DenseNet121 and MNv2 models for feature extraction to build the final model for skin NTDs classification.
Conclusions:
The two-stage approach significantly boosted the models’ performance compared to other methods, while the data augmentation method further enhanced the performance of the selected models. Finally, this study suggests further studies using advanced class balancing methods with more data, and a possible integration of other clinical data types.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.