Quantitative ultrasound classification of healthy and chemically
degraded ex-vivo cartilage
Abstract
In this study, we explored the potential of seventeen quantitative
ultrasound parameters (radiofrequency-based) in assessing the
progressive loss of collagen and proteoglycans (mimicking an
osteoarthritis condition) in ex-vivo bovine cartilage samples. The
majority of the analyzed metrics showed significant changes as the
degradation progressed due to trypsin and collagenase treatment. For the
first time, we employed a combination of these ultrasound parameters to
create machine learning models for the automated detection of a model of
healthy and degraded cartilage samples. A logistic regression model
exhibited a remarkable capability of distinguishing between healthy and
collagenase-treated cartilage, achieving accuracy and an
area under the curve values of 93% and 90%, respectively. When
comparing healthy and trypsin-treated cartilage, an ensemble model
yielded accuracy and an area under the curve values of
83% and 75%, respectively. Histological and mechanical analyses
further confirmed the ultrasound findings, as collagenase had more
pronounced impact on both mechanical and histological properties
compared to trypsin. These metrics were obtained using an ultrasound
probe, with a transmission frequency of 15 MHz, typically used for the
diagnosis of musculoskeletal diseases. As a perspective, the proposed
quantitative ultrasound assessment could become a new standard for
monitoring cartilage health, aiding in the early detection of cartilage
pathologies and enabling prompt interventions.