Automated Colony-Forming Unit (CFU) Counting of Bacteria Using Digital Image Analysis Through Computer Vision with Python - Supplementary Data
Description
This supplementary material corresponds to an Excel file (.xlsx) that documents key information associated with the development and validation of an automated bacterial colony counter based on computer vision, implemented in Python. The file contains two main sheets. Sheet 1 (Table S1) describes the photographic devices used to acquire images from the database, including device type, model, and technical specifications relevant to image analysis. Sheet 2 presents the results of automatic image processing, including the number of colonies detected by the algorithm, the manual reference count, true positives, false positives, and false negatives. Based on these data, the system performance metrics (precision, recall, and F-measure) are reported, as well as the processing time per image. This dataset supports the quantitative evaluation of the performance and computational efficiency of the proposed colony counter.
Files
Steps to reproduce
The data in Table S2 were obtained from images of Petri dishes acquired in a photographic booth built to ensure controlled lighting and capture conditions. The images were processed using a script developed in Python, which applies computer vision techniques for the automatic detection and counting of colony-forming units (CFUs). The CFUs were cultured in nutrient agar and R2A media from water samples taken from a glacial lake located in the Llaca ravine. The performance of the system was evaluated by comparing the automated count with the manual reference count, from which true positives, false positives, false negatives, and the metrics of precision, recall, and F-measure were calculated, as well as the processing time.
Institutions
- Universidad Nacional Santiago Antunez de Mayolo