Exploring convolutional neural networks with transfer learning for diagnosing Lyme disease from skin lesion images

BACKGROUND AND OBJECTIVE

Lyme disease which is one of the most common infectious vector-borne diseases manifests itself in most cases with erythema migrans (EM) skin lesions. Recent studies show that convolutional neural networks (CNNs) perform well to identify skin lesions from images. Lightweight CNN based pre-scanner applications for resource-constrained mobile devices can help users with early diagnosis of Lyme disease and prevent the transition to a severe late form thanks to appropriate antibiotic therapy. Also, resource-intensive CNN based robust computer applications can assist non-expert practitioners with an accurate diagnosis. The main objective of this study is to extensively analyze the effectiveness of CNNs for diagnosing Lyme disease from images and to find out the best CNN architectures considering resource constraints.

METHODS

First, we created an EM dataset with the help of expert dermatologists from Clermont-Ferrand University Hospital Center of France. Second, we benchmarked this dataset for twenty-three CNN architectures customized from VGG, ResNet, DenseNet, MobileNet, Xception, NASNet, and EfficientNet architectures in terms of predictive performance, computational complexity, and statistical significance. Third, to improve the performance of the CNNs, we used custom transfer learning from ImageNet pre-trained models as well as pre-trained the CNNs with the skin lesion dataset HAM10000. Fourth, for model explainability, we utilized Gradient-weighted Class Activation Mapping to visualize the regions of input that are significant to the CNNs for making predictions. Fifth, we provided guidelines for model selection based on predictive performance and computational complexity.

RESULTS

Customized ResNet50 architecture gave the best classification accuracy of 84.42% ±1.36, AUC of 0.9189±0.0115, precision of 83.1%±2.49, sensitivity of 87.93%±1.47, and specificity of 80.65%±3.59. A lightweight model customized from EfficientNetB0 also performed well with an accuracy of 83.13%±1.2, AUC of 0.9094±0.0129, precision of 82.83%±1.75, sensitivity of 85.21% ±3.91, and specificity of 80.89%±2.95. All the trained models are publicly available at https://dappem.limos.fr/download.html, which can be used by others for transfer learning and building pre-scanners for Lyme disease.

CONCLUSION

Our study confirmed the effectiveness of even some lightweight CNNs for building Lyme disease pre-scanner mobile applications to assist people with an initial self-assessment and referring them to expert dermatologist for further diagnosis.

[The open labs Tous Chercheurs]

L’enjeu d’une culture scientifique pour tous est d’importance face à la difficulté des citoyens à critiquer les données de la science avec des arguments rationnels, notamment en ce qui concerne la biologie et la santé (vaccination, procréation médicalement assistée, etc.), car le grand public ne veut plus croire sur parole ce que disent les experts. Dans ce contexte, rapprocher les citoyens de la recherche scientifique représente un réel défi pour l’avenir, que les laboratoires ouverts Tous Chercheurs ont voulu relever.

A gene repertoire for nitrogen transporters in Laccaria bicolor

Ectomycorrhizal interactions established between the root systems of terrestrial plants and hyphae from soil-borne fungi are the most ecologically widespread plant symbioses. The efficient uptake of a broad range of nitrogen (N) compounds by the fungal symbiont and their further transfer to the host plant is a major feature of this symbiosis. Nevertheless, we far from understand which N form is preferentially transferred and what are the key molecular determinants required for this transfer. Exhaustive in silico analysis of N-compound transporter families were performed within the genome of the ectomycorrhizal model fungus Laccaria bicolor. A broad phylogenetic approach was undertaken for all families and gene regulation was investigated using whole-genome expression arrays. A repertoire of proteins involved in the transport of N compounds in L. bicolor was established that revealed the presence of at least 128 gene models in the genome of L. bicolor. Phylogenetic comparisons with other basidiomycete genomes highlighted the remarkable expansion of some families. Whole-genome expression arrays indicate that 92% of these gene models showed detectable transcript levels. This work represents a major advance in the establishment of a transportome blueprint at a symbiotic interface, which will guide future experiments.