P456 Defungi: direct mycological examination of microscopic fungi images

POSTER SESSION 3, SEPTEMBER 23, 2022, 12:30 PM - 1:30 PM: OBJECTIVE: To classify five fungi types using two different deep learning approaches and three different convolutional neural network models, VGG16, Inception V3, and ResNet50. METHOD: A mycological laboratory in Colombia donated the images used for the development of this research work. They were manually labeled into five classes and curated with subject matter expert assistance. The images were later cropped and modified with automated coding routines to produce the final dataset. RESULTS: We present experimental results classifying five types of fungi using two different deep learning approaches and three different convolutional neural network models, VGG16, Inception V3, and ResNet50. The first approach benchmarks the classification performance for the models trained from scratch, while the second approach benchmarks the classification performance using pre-trained models based on the ImageNet dataset. Using k-fold cross-validation testing on the 5-class dataset, the best performing model trained from scratch was Inception V3, reporting 73.2% accuracy. Likewise, the best performing model using transfer learning was VGG16, with 85.04% accuracy. CONCLUSION: The statistics provided by the two approaches create an initial benchmark to encourage future research work to improve classification performance. Furthermore, the dataset built is published on Kaggle and GitHub to encourage future research.