Mean scatterer spacing estimation from pellets using cepstral analysis: A preliminary study

Ultrasonic backscattered signals from biological tissues contain information regarding their structures, more specifically, their scatterer structures. This work investigates the use of cepstral analysis in characterizing periodicities in ultrasound Alines due to uniformity in the scatterer distribution. The A-line is simulated as a convolution between a generated radio-frequency (RF) pulse and a scattering medium containing uniformly distributed scatterers along with randomly situated scatterers. The cepstral analysis is tested by varying the regularity of the scatterers. Simulation results indicate that the mean scatterer spacing can be estimated using both power and complex cepstrum, where it manifests itself as a peak. Experimental results conducted on cell pellets imaged at 1 and 56 hrs revealed a peak at 0.15 mm. In conclusion, a general agreement between simulation and experimental results was found. Future work include the further investigation of the use of mean scatterer spacing as a new biomarker for cancer treatment monitoring.