Purpose: This study aimed to use the multiscale blob feature (MBF) method to quantitatively evaluate porcine striated muscle injuries.
Methods: A porcine striated muscle injury model was induced by microwave ablation and anhydrous acetic acid injection, respectively. Then, both 2D sonographic and histological features of the lesions were recorded and compared. Later, MBF was used to quantitatively evaluate the porcine striated muscle injuries by extracting the texture features from the 2D ultrasonogram via measuring eight textural parameters (Mean, SDev, NOB, [Formula: see text], [Formula: see text], HOD, DOD, and POD).
Results: Microwave ablation produced oval or round-like lesions, which had a pale gray color, with an echo attenuation detected at lesion center due to carbonization; anhydrous acetic acid injection produced long, stripped lesions, which had a slate-gray color, with a gas-like intense echo at lesion center. There were significant differences in Mean, [Formula: see text] and POD between the muscle samples treated by microwave ablation and the control samples, as well as significant differences in NOB, [Formula: see text] and POD between the muscle samples treated by anhydrous acetic acid injection and the control. There were significant differences in Mean, [Formula: see text], NOB, and [Formula: see text] between the muscle samples treated by microwave ablation and those treated by anhydrous acetic acid injection.
Conclusion: Quantitative evaluation of striated muscle injuries using the MBF method was able to differentiate the muscle injuries caused by microwave ablation and anhydrous acetic acid injection, suggesting that this method may be a potential and reliable tool for quantitative evaluation of muscle injuries.
Keywords: Anhydrous acetic acid injection; Microwave ablation; Multiscale blob features; Quantitative ultrasonography; Striated muscle injury.