Feasibility of non-invasive temperature estimation by the assessment of the average gray-level content of B-mode images

Thermometry and ablation monitoring with ultrasound

“…Alvarenga, Teixeira and collaborators have studied the possibility of using the grey-level co-occurrence matrix (GLCM), the GLCM-derived entropy, and the average gray-level content (AVGL) to estimate temperature in vivo (51–53). Since there is excellent correlation between temperature and these B-mode image-derived indices up to 44° C, these easy to implement methods could be utilized for quantitative monitoring of hyperthermia.…”

Uncertainty evaluation from non-invasive estimation of temperature variation using B-mode ultrasonic images from a plastic phantom

“…However, despite the non-linearity, the relation between AVGL and temperature rise was reproducible, Table 1 Values of a for the 10 tests performed and respective combined uncertainties, and also the correlation between AVGL and temperature variations. allowing the computation of consistent two-dimensional temperature rise maps in homogeneous media [10]. Finally, it is worth emphasising that this is the unique study presenting a model to uncertainty assessment of temperature rise estimations based on AVGL.…”

“…Other authors show that microwave thermotherapy applied to pig and bovine livers [13,14] also produce linear relations between AVGL and temperature rise. On the other hand, literature shows that therapeutic ultrasound (TUS) heating tends to produce non-linear relations between AVGL and temperature rise, once temperature increases in a fast way forced by the TUS pressure field (which is a spatially inhomogeneous source) [10]. However, despite the non-linearity, the relation between AVGL and temperature rise was reproducible, Table 1 Values of a for the 10 tests performed and respective combined uncertainties, and also the correlation between AVGL and temperature variations.…”

“…The US scanner was set to reproduce B-mode ultrasonic images visually compatible to clinical practice, and then the applied parameters were left unchanged during the test. This procedure enables the assessment of differential AVGL changes, minimizing the effect of variations on the instrumentation settings, and discarding the need of normalization [10]. The experimental setup was developed at Labus/Inmetro and is presented in Fig.…”

“…The main effects observed are apparent image motion due to changes in speed-of-sound, and changes in image intensity due to variations in the amplitude of the echo signal caused by modifications in the attenuation and backscattered properties. Literature reports changes in the intensity of B-mode images mainly caused by temperature-related changes in backscattered energy (CBE) from tissue inhomogeneity [9][10][11][12][13][14]. The analysis of the temperature-dependent average grey-level feature (AVGL) was described in different papers, including images collected from ex-vivo porcine, bovine, and chicken samples heated by microwave radiation, therapeutic ultrasound and water-bath immersion, and correlations between AVGL and temperature were reported to be above 70% [10,[12][13][14][15][16].…”