Aleksandar Miladinović

Accurate classification of retinal diseases such as dry age-related macular degeneration, wet AMD, epiretinal membrane, full-thickness macular hole (MH), lamellar MH, and central serous chorioretinopathy (CSC) is essential for effective treatment and clinical decision-making. Traditional deep learning models, however, often struggle with imbalanced datasets and lack interpretability, limiting their translational applicability in ophthalmology. We propose a neurosymbolic framework that integrates a convolutional neural network (CNN) with a symbolic reasoning layer based on expert-defined clinical rules. A total of 10,846 optical coherence tomography images were retrospectively collected and categorized into seven diagnostic classes: dry AMD, wet AMD, epiretinal membrane, full-thickness MH, lamellar MH, central serous chorioretinopathy, and healthy retinas. Our neurosymbolic model achieved macro-precision 0.83, recall 0.82, and F1 0.81, on internal dataset, having slightly better performance than the CNN (0.64/0.83/0.68). On the external dataset, it retained superior performance, macro-precision 0.85, recall 0.79, F1 0.78, versus the CNN (0.73/0.64/0.59). Our hybrid neurosymbolic framework introduces a unified paradigm that couples symbolic reasoning with a conventional CNN, improving diagnostic performance while delivering transparent, clinically interpretable decisions. It is particularly effective for rare and complex conditions that often challenge end-to-end deep learning models. By integrating symbolic clinical logic with visual pattern recognition, the neurosymbolic model fosters trust in artificial intelligence-assisted diagnostics and supports precise, explainable decision-making in retinal care.

Purpose: Tympanometry, the current gold standard for assessing middle ear function, has limitations in infants and in cases of tympanic membrane perforation due to its potential invasiveness, as it requires the application of varying pressure in the ear canal. To overcome these issues, the Pressure-Less Acoustic Immittance (PLAI ^TM ) device was recently developed as a non-invasive alternative that does not require pressure application. However, reference values and age-specific norms are not yet established. This study aims to define age-specific reference ranges for key acoustic admittance parameters using PLAI ^TM in healthy individuals and to offer a preliminary comparison with subjects affected by Otitis Media with Effusion (OME), a common middle ear pathology. Methods: A total of 318 healthy ears and 68 OME-affected ears were analyzed, covering an age range from 4 months to 80 years. Parameters extracted from PLAI ^TM admittance curves included resonance frequency, admittance peak, minimum and maximum frequencies of the resonance peak, bandwidth, half peak height, and equivalent volume. Results: Results indicate a strong age dependence for all parameters, best modeled with a logarithmic fit. Statistical analysis supports categorizing ages into three groups (0–3, 3–12, 12+ years), reflecting stages of ear development. Most parameters stabilized after age 12. Significant differences were observed between healthy and OME ears across several PLAI ^TM metrics. Conclusion: These findings establish essential reference values and demonstrate the clinical potential of PLAI ^TM for diagnosing middle ear dysfunctions, particularly in populations where conventional tympanometry is limited.

Background/Objective: This study explores a novel approach for diagnosing common middle ear pathologies using Pressure-Less Acoustic Immittance (PLAI™), a non-invasive alternative to conventional tympanometry. Methods: A total of 516 ear measurements were collected and stratified into three age groups: 0–3, 3–12, and 12+ years, reflecting key developmental stages. PLAI™-derived acoustic parameters, including resonant frequency, peak admittance, canal volume, and resonance peak frequency boundaries, were analyzed using Random Forest classifiers, with SMOTE addressing class imbalance and SHAP values assessing feature importance. Results: Age-specific models demonstrated superior diagnostic accuracy compared to non-stratified approaches, with macro F1-scores of 0.79, 0.84, and 0.78, respectively. Resonant frequency, ear canal volume, and peak admittance consistently emerged as the most informative features. Notably, age-based stratification significantly reduced false negative rates for conditions such as Otitis Media with Effusion and tympanic membrane retractions, enhancing clinical reliability. These results underscore the relevance of age-aware modeling in pediatric audiology and validate PLAI™ as a promising tool for early, pressure-free middle ear diagnostics. Conclusions: While further validation on larger, balanced cohorts is recommended, this study supports the integration of machine learning and acoustic immittance into more accurate, developmentally informed screening frameworks.