Handheld Briefcase Optical Coherence Tomography with Real-Time Machine Learning Classifier for Middle Ear Infections

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Otolaryngology

Background:

• Middle ear infections are common, especially in children, with significant financial costs.
• Current diagnostic methods rely on subjective visual inspection via otoscope.
• Non-invasive quantitative assessment is needed for accurate diagnosis.

Purpose of the Study:

• To develop a portable optical coherence tomography (OCT) system for diagnosing middle ear infections.
• To integrate a machine learning platform for automated image analysis.
• To improve the accuracy and usability of OCT for middle ear condition diagnosis.

Main Methods:

• A briefcase-housed OCT system was developed with a handheld imaging probe.
• A real-time machine learning platform, utilizing a random forest classifier, was implemented.
• The system was tested for its ability to categorize middle ear images for effusions and biofilms.

Main Results:

• The OCT system successfully provided non-invasive, quantitative assessment of middle ear conditions.
• The machine learning classifier accurately categorized images based on the presence of effusions and biofilms.
• The system demonstrated user-invariant classification results, reducing reliance on expert interpretation.

Conclusions:

• The briefcase OCT system integrated with machine learning offers a promising tool for diagnosing middle ear infections.
• This technology can enhance diagnostic accuracy and accessibility in clinical settings.
• It has the potential to improve the management of middle ear infections, particularly in pediatric populations.