INTRODUCTION. Osteoarthritis (OA) is a growing societal burden, due to the ageing population. Less invasive, less damaging, and cheaper methods for diagnosis are needed, and sound technology is an emerging tool in this field. AIMS. The aim of the current research was to: 1) investigate the potential of visual scalogram analysis of Acoustic Emission (AE) frequencies within the human audible range (20–20000 Hz) to diagnose knee OA, 2) correlate the qualitative visual scalogram analysis of the AE with OA symptoms, and 3) to do this based on information gathered during gait. METHODS. The analysis was carried out on a database collected during a prospective sound study on healthy and osteoarthritic knees. Sound recordings obtained with a contact microphone mounted on the patella and attached to a digital pre-amplifier, whilst patients were walking on a treadmill, were visualised, manually sampled, and transformed into scalograms. Features of the scalograms were described and qualitatively analysed through chi-squared tests for association with healthy or OA knees (knee status), and with severity of OA pain and functional symptoms and impact on quality of life (QoL), activities of daily living (ADL) and sports using the Knee Injury and Osteoarthritis Outcome Score (KOOS) subscales. RESULTS. 28 patients (56 knees) were included in the analysis. Our method provides a wide variety of different scalogram features: if no events were recorded, the scalogram was classified as “quiet” (Fig 1). In case of abnormal recordings, data analysis evaluated association with the total count of the three most common events that appeared: 1. Peak (Fig 2), 2. Scattered (Fig 3) or 3. Island (localized noise but not presenting as a peak) (Fig 4) – “scalogram features”. No association was found between global scalogram characteristics (quiet versus “any noise”) and knee status (healthy or OA) (χ. 2. =3.163, p=0.075), but was found between knee status and three specific scalogram features (χ. 2. =9.743, p=0.008). The strongest association was a higher frequency of the “scattered” feature in the OA group (χ. 2. =9.06, p=0.01). Scalogram characteristics had no significant association with the sports and recreation (χ. 2. =1.74, p=0.419) nor the activities of daily living (χ. 2. =1.80, p=0.406) KOOS subscales. Significant association was found between scalogram characteristic and the pain (χ. 2. =10.34, p=0.006), quality of life (χ. 2. =6.58, p=0.037), and symptoms (χ. 2. =7.54, p=0.023) subscales. CONCLUSION. Promising results from analysis of individual features and of KOOS subscales establish the potential of acoustic analysis in evaluation of OA knees. More analysis of the data is needed to better define the variety of scalogram features. The future consequences of this research would be the development of a fast and affordable, non-invasive, radiation-free and potentially portable approach to evaluation, diagnosis and longitudinal monitoring of knee disorders