Urinary Cell-Free DNA detection platform for bacterial cystitis diagnosis via spot-based machine learning

Area of Science:

• Biomedical Engineering
• Infectious Disease Diagnostics
• Molecular Diagnostics

Background:

• Bacterial cystitis, often caused by E. coli, requires rapid identification for effective treatment.
• Urinary cell-free DNA (cfDNA) shows promise as a biomarker but distinguishing bacterial from non-bacterial causes is challenging.

Purpose of the Study:

• To develop an image-based, machine learning-assisted fluorescent detection system for simultaneous quantitative and qualitative analysis of urinary cfDNA.
• To enable rapid and accurate diagnosis of bacterial cystitis.

Main Methods:

• Utilized polyethyleneimine (PEI)-functionalized glass slides for cfDNA enrichment.
• Employed fluorescence imaging with propidium iodide labeling and YOLOv5 object detection algorithm for automated analysis.
• Validated the system using a bacterial cystitis mouse model and 148 human urine specimens.

Main Results:

• The platform quantifies cfDNA within 20 minutes, with a detection limit of 20.29 ng/mL.
• Automated analysis of cfDNA spot patterns differentiated bacterial (77.12% smaller spots) from non-bacterial cystitis (13.87% smaller spots).
• Achieved 94.43% diagnostic accuracy, 94.45% sensitivity, and 94.35% specificity for bacterial cystitis in clinical samples, significantly outperforming conventional methods.

Conclusions:

• This rapid, on-site platform requires no urine pretreatment, advancing point-of-care testing (POCT) for bacterial cystitis.
• The system has potential applications for digital biomarker detection in various inflammatory and infectious diseases.