# A Complete Transfer Learning-Based Pipeline for Discriminating Between Select Pathogenic Yeasts from Microscopy Photographs

## TL;DR

This paper introduces a deep learning pipeline that accurately identifies pathogenic yeasts from microscope images, offering a fast and affordable diagnostic solution.

## Contribution

A novel CNN-based pipeline using transfer learning for rapid and accurate classification of six pathogenic yeast species from microscopy images.

## Key findings

- The model achieved 93.91% accuracy at the patch level and 99.09% at the whole image level.
- The pipeline reduces reliance on costly molecular methods for yeast identification in clinical settings.

## Abstract

Pathogenic yeasts are an increasing concern in healthcare, with species like Candida auris often displaying drug resistance and causing high mortality in immunocompromised patients. The need for rapid and accessible diagnostic methods for accurate yeast identification is critical, especially in resource-limited settings. This study presents a convolutional neural network (CNN)-based approach for classifying pathogenic yeast species from microscopy images. Using transfer learning, we trained the model to identify six yeast species from simple micrographs, achieving high classification accuracy (93.91% at the patch level, 99.09% at the whole image level) and low misclassification rates across species, with the best performing model. Our pipeline offers a streamlined, cost-effective diagnostic tool for yeast identification, enabling faster response times in clinical environments and reducing reliance on costly and complex molecular methods.