39th Annual Student Research Conference
Automated Fungal Growth Analysis Application - Early Access Release
Mohammed Ayan Mahmood
Dr. Hajeewaka C. Mendis and Dr. Kafi M. Rahman, Faculty Mentors
MoldEZ, version 4.0, is an automated fungal growth analysis application that replaces slow, error prone manual measurements with deep learning based semantic segmentation trained on 200+ annotated images (petri dish mIoU: 99.2%, mold mIoU: 99.4%). The Python application integrates a Tkinter GUI, CLAHE preprocessing, and SciPy/Matplotlib visualization, exporting standardized PDF reports for reproducible analysis. A Raspberry Pi pipeline enables fully automated hourly image capture and cloud logging, providing an end to end solution for accurate, high throughput fungal growth quantification. With empirical validation currently being conducted, the dissertation is progressing steadily. The early access release will help us prepare an advanced version 5.0 of the application.
Keywords: Fungal Growth Analysis, Deep Learning, Petri Dish, Lab Automation, Python, Colonies, Raspberry Pi, MoldEZ
Topic(s):Biology
Computer Science
Presentation Type: Oral Presentation
Session: -4
Location: MG 2007
Time: 10:45