Noninvasive Temperature Estimation Research Articles

On the possibility of non-invasive multilayer temperature estimation using soft-computing methods

Objective and motivation This work reports original results on the possibility of non-invasive temperature estimation (NITE) in a multilayered phantom by applying soft-computing methods. The existence of reliable non-invasive temperature estimator models would improve the security and efficacy of thermal therapies. These points would lead to a broader acceptance of this kind of therapies. Several approaches based on medical imaging technologies were proposed, magnetic resonance imaging (MRI) being appointed as the only one to achieve the acceptable temperature resolutions for hyperthermia purposes. However, MRI intrinsic characteristics (e.g., high instrumentation cost) lead us to use backscattered ultrasound (BSU). Among the different BSU features, temporal echo-shifts have received a major attention. These shifts are due to changes of speed-of-sound and expansion of the medium. Novelty aspects The originality of this work involves two aspects: the estimator model itself is original (based on soft-computing methods) and the application to temperature estimation in a three-layer phantom is also not reported in literature. Materials and methods In this work a three-layer (non-homogeneous) phantom was developed. The two external layers were composed of (in % of weight): 86.5% degassed water, 11% glycerin and 2.5% agar–agar. The intermediate layer was obtained by adding graphite powder in the amount of 2% of the water weight to the above composition. The phantom was developed to have attenuation and speed-of-sound similar to in vivo muscle, according to the literature. BSU signals were collected and cumulative temporal echo-shifts computed. These shifts and the past temperature values were then considered as possible estimators inputs. A soft-computing methodology was applied to look for appropriate multilayered temperature estimators. The methodology involves radial-basis functions neural networks (RBFNN) with structure optimized by the multi-objective genetic algorithm (MOGA). In this work 40 operating conditions were considered, i.e. five 5-mm spaced spatial points and eight therapeutic intensities ( I SATA ) : 0.3, 0.5, 0.7, 1.0, 1.3, 1.5, 1.7 and 2.0 W / cm 2 . Models were trained and selected to estimate temperature at only four intensities, then during the validation phase, the best-fitted models were analyzed in data collected at the eight intensities. This procedure leads to a more realistic evaluation of the generalisation level of the best-obtained structures. Results and discussion At the end of the identification phase, 82 (preferable) estimator models were achieved. The majority of them present an average maximum absolute error (MAE) inferior to 0.5 °C. The best-fitted estimator presents a MAE of only 0.4 °C for both the 40 operating conditions. This means that the gold-standard maximum error (0.5 °C) pointed for hyperthermia was fulfilled independently of the intensity and spatial position considered, showing the improved generalisation capacity of the identified estimator models. As the majority of the preferable estimator models, the best one presents 6 inputs and 11 neurons. In addition to the appropriate error performance, the estimator models present also a reduced computational complexity and then the possibility to be applied in real-time. Conclusions A non-invasive temperature estimation model, based on soft-computing technique, was proposed for a three-layered phantom. The best-achieved estimator models presented an appropriate error performance regardless of the spatial point considered (inside or at the interface of the layers) and of the intensity applied. Other methodologies published so far, estimate temperature only in homogeneous media. The main drawback of the proposed methodology is the necessity of a-priory knowledge of the temperature behavior. Data used for training and optimisation should be representative, i.e., they should cover all possible physical situations of the estimation environment.

A performance analysis of echographic ultrasonic techniques for non-invasive temperature estimation in hyperthermia range using phantoms with scatterers

Optimization of efficiency in hyperthermia requires a precise and non-invasive estimation of internal distribution of temperature. Although there are several research trends for ultrasonic temperature estimation, efficient equipments for its use in the clinical practice are not still available. The main objective of this work was to research about the limitations and potential improvements of previously reported signal processing options in order to identify research efforts to facilitate their future clinical use as a thermal estimator. In this document, we have a critical analysis of potential performance of previous ultrasonic research trends for temperature estimation inside materials, using different processing techniques proposed in frequency, time and phase domains. It was carried out in phantom with scatterers, assessing at their specific applicability, linearity and limitations in hyperthermia range. Three complementary evaluation indexes: technique robustness, Mat–lab processing time and temperature resolution, with specific application protocols, were defined and employed for a comparative quantification of the behavior of the techniques. The average increment per °C and mm was identified for each technique (3 KHz/°C in the frequency analysis, 0.02 rad/°C in the phase domain, while increments in the time domain of only 1.6 ns/°C were found). Their linearity with temperature rising was measured using linear and quadratic regressions and they were correlated with the obtained data. New improvements in time and frequency signal processing in order to reveal the potential thermal and spatial resolutions of these techniques are proposed and their subsequent improved estimation results are shown for simulated and measured A-scans registers. As an example of these processing novelties, an excellent potential resolution of 0.12 °C into hyperthermia range, with near-to-linear frequency dependence, could be achieved. Specifically defined “numerical” and physical multi-scatter phantoms are described, which mimic ultrasound velocity in tissues of about 1560 m/s @ 35 °C and have a quasi-uniform internal scattering structure designed to assure standard signal patterns adequate for processing comparisons in the same time and sound velocity conditions for all the techniques analyzed, and to obtain easily repeatable multi-pulse echo-patterns. A perfect lineal dependence (100% of correlation coefficient) between the unitary average increment measured by each technique and temperature rising was observed while working with simulated A-scan registers, where all the parameters are under an accurate control. Nevertheless a very small quadratic tendency appeared in the results obtained from experimental echo registers, which are more similar to a real tissues case. It would be an interesting future work to analyze the behavior of these techniques in real tissues in order to confirm or reject this light quadratic tendency. Finally, new methods were detailed and applied in order to precisely quantify the advantages of each estimation technique; their respective intrinsic limitations were also underlined.

Neuro-genetic non-invasive temperature estimation: Intensity and spatial prediction

The existence of proper non-invasive temperature estimators is an essential aspect when thermal therapy applications are envisaged. These estimators must be good predictors to enable temperature estimation at different operational situations, providing better control of the therapeutic instrumentation. In this work, radial basis functions artificial neural networks were constructed to access temperature evolution on an ultrasound insonated medium. The employed models were radial basis functions neural networks with external dynamics induced by their inputs. Both the most suited set of model inputs and number of neurons in the network were found using the multi-objective genetic algorithm. The neural models were validated in two situations: the operating ones, as used in the construction of the network; and in 11 unseen situations. The new data addressed two new spatial locations and a new intensity level, assessing the intensity and space prediction capacity of the proposed model. Good performance was obtained during the validation process both in terms of the spatial points considered and whenever the new intensity level was within the range of applied intensities. A maximum absolute error of 0.5 degrees C+/-10% (0.5 degrees C is the gold-standard threshold in hyperthermia/diathermia) was attained with low computationally complex models. The results confirm that the proposed neuro-genetic approach enables foreseeing temperature propagation, in connection to intensity and space parameters, thus enabling the assessment of different operating situations with proper temperature resolution.

Multilayered non-invasive temperature estimation from backscattered ultrasound

A major drawback on the application of thermal therapies is the lack of reliable knowledge about temperature in the region under treatment. The most attractive approach is non-invasive temperature estimation (NITE) strategies and, among them, techniques based on backscattered ultrasound (BSU) are a convenient choice for its portability and simplicity. Using BSU some methods have been proposed for temperature estimation basically for homogeneous media, and simple operating situations. More complex scenarios can be explored by non-linear approaches. In this work, a methodology based on neural networks is presented for temperature estimation in a multilayer phantom. A three-layer gel-based phantom was heated at eight intensities, between 0.30 and 2.0 W/cm2 (at 1MHz) for 15 minutes. Temperature and BSU signals were acquired from five spatial-points at each 10 seconds. Temporal echo-shifts, induced by temperature change, were computed from the BSU signals, and used as model's input information. Neural models were developed to estimate temperature at four intensities. Then models were validated in data including all the intensities. The best model presents a maximum absolute error of 0.43 °C for untrained situations. To the best of our knowledge this is the first NITE approach in multilayered media.

Evaluation of performance in some ultrasonic procedures for non-invasive thermal estimation into hyperthermia phantoms

To dispose of a precise and non-invasive temperature measurement over the treated area becomes very important during hyperthermia treatments, in order to make possible an optimization of their healing effects. A performance analysis of some previously reported ultrasonic techniques, which were proposed for non-invasive temperature estimation, is made in this work over phantoms mimicking human tissues. A first technique is based on discrete scattering modeling for tissue characterization and spectral analysis of frequency distributions related to average scatters spacing [R. Seip & E.S. Ebbini]. Other technique uses certain relations between tissue temperature changes and time-shifts in echoes due to thermally induced sound velocity changes and expansions [R. Maass-Moreno & C.A. Damianou]. And finally, the third procedure analyzed here, is a recently proposed alternative, based on phase demodulation processing, to estimate the indirect effects of echoes time-shifts in the phase domain [M. Vázquez, A. Ramos, et al.]. The three options are analyzed for the same simulated & measured multi-pulse echo patterns, looking for detecting possible advantages and inconveniences in each case. Finally, some improvement paths over the basic estimation methods are also explored.

Noninvasive temperature estimation by detecting echo-strain change including thermal expansion

This article studies the feasibility of noninvasive temperature estimation by detecting echo-strain including thermal expansion in therapeutic ultrasound treatment. This technique evaluates distributions of echo-strain and temperature inside the tissue by detecting echo signals pre- and post-heating, in combination with the temperature dependence of sound speed and thermal expansion. In the computer simulation and experimental study, echo signals pre- and post-heating are acquired and then the temperature elevation is evaluated by correlation analysis. Results demonstrate that this technique can effectively extend the measured temperature range up to 75°C with an accuracy of ±2°C.

MO‐E‐L100F‐03: Noninvasive Temperature Imaging Using Diagnostic Pulse‐Echo Ultrasound

Several ultrasonic parameters are known to have temperature dependence including the speed of sound, attenuation, nonlinearity parameter, etc. Local changes in tissue temperature can be estimated noninvasively using diagnostic pulse‐echo ultrasound. A number of temperature estimation methods have been proposed since the early 1990s with varying levels of success. One method that was investigated by numerous groups is based on minute changes in echo locations due to temperature‐dependent changes in the speed of sound and thermal expansion. Both in vitro and in vivo results have been obtained to demonstrate the feasibility of the method. However, several limitations have been shown to limit the use of this method in practice. In this paper, we describe a true two‐dimensional displacement tracking algorithm with robust temperature estimation from real‐time 2D pulse‐echo ultrasound. The method employs a physics‐based Kalman filter derived from the transient bioheat transfer equation (BHTE). The filter is shown to be effective in rejecting displacement artifacts from tissue motion as well as those resulting from the thermal lens effects.This lecture will provide a description of the BHTE and the system approach leading to the design of the Kalman filter for tracking the temperature data based on the observed displacements in echo location in the region of interest. The tracking properties of the filter and its ability to reject artifacts from the observed displacements will be illustrated. One‐dimensional and two‐dimensional versions of the filter will be presented and contrasted in terms of their ability to reject displacement artifacts. Comparisons with previously published algorithms will be given using in vivo and in vitro data.We will also present results on using the Kamlan filter methodology in tracking changes in tissue absorption associated with lesion formation using high intensity focused ultrasound (HIFU). This technique may lead to quantitative parameter imaging for monitoring and control of HIFU treatments under ultrasound image guidance.The Educational Objectives:1. Understand the origin of temperature‐dependence of relevant acoustic parameter and their effects on observed pulse‐echo data.2. Understand the limitations of noninvasive temperature estimation using pulse‐echo data.3. Understand the connection between the physical model for temperature evolution and robust temperature and parameter tracking in tissue media.

Noninvasive Temperature Estimation in Oncology Hyperthermia Using Phase Changes in Pulse–Echo Ultrasonic Signals

An accurate and noninvasive temperature measurement inside phantoms and tissues is important to improve hyperthermia treatment. The use of ultrasonic echo-pulses for this purpose is a promising line of research. In this paper, we propose an alternative method for processing echo-signals received from inner regions in phantoms. Ultrasonic estimation of temperature is carried out using a phase demodulation technique, which gauges the indirect effects in the phase domain of the previously reported time-shifts in the echo position. The potential performance and high sensitivity of this procedure is explained and verified with several simulated examples. Finally, the proposed procedure is applied to the experimental assessment of phase changes with temperature in ultrasonic echoes successively acquired from different milieus. In all cases, the results show good performance and high sensitivity. This is an interesting alternative to measurements in waveforms of the very small delays (nanoseconds) associated with this ultrasonic temperature estimation.

Non-invasive temperature prediction of in vitro therapeutic ultrasound signals using neural networks

In this paper, a novel black-box modelling scheme applied to non-invasive temperature prediction in a homogeneous medium subjected to therapeutic ultrasound is presented. It is assumed that the temperature in a point of the medium is non-linearly related to some spectral features and one temporal feature, extracted from the collected RF-lines. The black-box models used are radial basis functions neural networks (RBFNNs), where the best-fitted models were selected from the space of model structures using a genetic multi-objective strategy. The best-fitted predictive model presents a maximum absolute error less than 0.4 degrees C in a prediction horizon of approximately 2 h, in an unseen data sequence. This work demonstrates that this type of black-box model is well-suited for punctual and non-invasive temperature estimation, achieving, for a single point estimation, better results than the ones presented in the literature, encouraging research on multi-point non-invasive temperature estimation.

SINGLE BLACK-BOX MODELS FOR TWO-POINT NON-INVASIVE TEMPERATURE PREDICTION

In this paper the performance of a genetically selected radial basis functions neural network is evaluated for non-invasive two-point temperature estimation in a homogeneous medium, irradiated by therapeutic ultrasound at physiotherapeutic levels. In this work a single neural network was assigned to estimate the temperature profile at the two considered points, and more consistent results were obtained than when considering one model for each point. This result was possible by increasing the model complexity. The best model predicts the temperature from two unseen data sequences during approximately 2 hours, with a maximum absolute error less than 0.5 °C, as desired for a therapeutic temperature estimator.

Noninvasive Estimation of Tissue Temperature Via High-Resolution Spectral Analysis Techniques

We address the noninvasive temperature estimation from pulse-echo radio frequency signals from standard diagnostic ultrasound imaging equipment. In particular, we investigate the use of a high-resolution spectral estimation method for tracking frequency shifts at two or more harmonic frequencies associated with temperature change. The new approach, employing generalized second-order statistics, is shown to produce superior frequency shift estimates when compared to conventional high-resolution spectral estimation methods Seip and Ebbini (1995). Furthermore, temperature estimates from the new algorithm are compared with results from the more commonly used echo shift method described in Simon et al. (1998).

Relationship between the temperature and the acoustic nonlinearity parameter in biological tissues

Recently with the rapid development of the high-intensity focused ultrasound (HIFU) in biomedical ultrasound, much attention has been paid to the noninvasive temperature estimation in biological tissue in order to determine the region and degree of the ultrasound-induced lesions. In ultrasound hyperthermal therapy it is highly desirable to study the real-time noninvasive monitoring of temperature distribution in biological tissue. In this paper, the relationship between the nonlinearity parameterB/A and the temperature in biological tissue is studied and compared with the theoretical model as well as the experimental results from the thermocouple. Results indicated thatB/A could be used as an effective tool to monitor the temperature distribution in biological media.

Model-Based Optimization of Phased Arrays for Electromagnetic Hyperthermia

A summary of recent progress in model-based optimization of phased arrays for electromagnetic hyperthermia is reported. The electromagnetic phased array has the potential to overcome many of the difficulties associated with noninvasive hyperthermia, and is more effective if the driving amplitudes and phases of the array are carefully selected. A computationally efficient method for the optimization of the steady-state temperature distribution, a major driver of therapeutic response, has been developed. By employing a dual set of superposition principles, the technique minimizes the number of computationally expensive forward problems that must be solved in the course of an optimization. Additionally, a scheme that employs emerging noninvasive tomographic temperature estimation techniques, such as magnetic resonance thermometry, to perform optimization of a phased array has been developed and demonstrated experimentally. Conclusions about the potential value of each of the developed techniques are reached and directions for further research are indicated.

A model-based noninvasive temperature estimation technique for monitoring HIFU therapy using backscattered ultrasound

A new noninvasive temperature estimation technique using the bioheat transfer equation (BHTE) to constrain inversion of ultrasonic travel time data has been developed for HIFU therapy monitoring. Initial estimates for BHTE parameters and ultrasonic travel time temperature dependence δτ(T) are noninvasively obtained in situ using ultrasound backscatter signals acquired during two probe heating exposures performed prior to therapy: one to obtain medium thermal parameters, the other to estimate the magnitude of the local HIFU heat source. These estimates are used in BHTE simulations and compared to local travel time changes measured periodically during therapy using a medical ultrasound imager. Quantitative 2-D temperature maps are obtained by constrained data fitting. Preliminary results obtained from in vitro experiments in tissue mimicking phantoms will be presented. Temperature maps are computed throughout therapy delivery and posttreatment cooling periods, and validated against independent temperature measurements obtained using thermocouples placed close to but not at the HIFU focus. This model-based technique permits temperature estimation throughout the entire therapeutic range and is completely noninvasive, and is thus a departure from previously reported techniques. This ultrasound-based approach could serve as an alternative to MRI-based temperature estimation methods in a clinical setting.

Noninvasive Estimation of Temperature and pH in Human Lower Leg Muscles using 1H Nuclear Magnetic Resonance Spectroscopy

The temperature and pH of human lower leg muscles were estimated noninvasively using 1H‐NMR spectroscopy at 3.0 and 1.5 T on five normal volunteers (21M, 24M, 27M, 34M, 47M). The chemical shifts of water and imidazole protons relative to cholines (–N+CH3) or creatines (–CH3) could be used as the temperature and pH probes, respectively [1–4]. Using the chemical shift, estimated values of the temperature in gastrocnemius (GAS; shell region) and soleus muscles (SOL; core region) under ambient temperature (21–25°C) were 33.6±0.4 and 35.3±0.4°C (mean±SE), respectively (significantly different; P < 0.01). The values of pH in these muscles were estimated to be 6.97 ± 0.01 and 6.96 ± 0.02, respectively. Alternation of the surface temperature of the lower leg from 40 to 10°C significantly changed the temperature in GAS (P < 0.0001) from 35.8 ± 0.4 to 26.2 ± 1.2°C and the pH in GAS rose from 6.95 ± 0.01 to 7.01 ± 0.01 (P <0.01). However, the values of pH and temperature in SOL were not significantly affected by this maneuver. These results indicate that the pH in GAS was moderately changed by muscle temperature (r= –0.59, P <0.01), and its change was estimated to be –0.005 pH units/°C.

Noninvasive pressure and temperature estimation in large-volume apparatus by equation-of-state cross-calibration

The utility of internal calibration of large-volume cell high-pressure (p) and high-temperature (T) conditions by a noninvasive equation of state (EOS) cross-calibration method is assessed. This involved the introduction of simultaneously measured unit-cell volumes, measured by in-situ x-ray diffraction in Paris-Edinburgh large-volume apparatus, for gold and NaCl into their EOS to obtain a simplified solution set for p and T run conditions. These have been gauged from K-type thermocouple measurements and the EOS of hBN to establish an estimate of the combined errors involved in this method. It was found that the total error (in combined p, T and three EOS calculations) results in apparent p miscalculation of 0.2 GPa and with a mean error in T of 13%. This technique alone is valuable in high p and T experimentation above ∼5 GPa where no reliable pressure correction exists for thermocouples and in situations where the use of a thermocouple is not advisable, for example, where thermocouple materials are sample-reactive. Combined with furnace power - p- T curves, this method offers a viable alternative to invasive measurement.

Noninvasive estimation of tissue temperature via ultrasound frequency shift using a narrow-band transducer

When applying the high power focused ultrasound to medical therapy, it is necessary to monitor the temperature change in tissue. Some methods have been proposed to measure this change noninvasively. One of those techniques measures the frequency changes in harmonics which are linear with the temperature changes. The linear coefficient has been defined with the fundamental frequency changes and the temperature changes. This measurement requires at least three harmonics, so that it requires a wideband transducer to estimate the fundamental frequency efficiently. This paper suggests a method to measure the temperature changes noninvasively with only one component of harmonics, under the condition that the proportional constant is defined such as the thermal expansion coefficient. It does not require the wideband transducer. Additionally, it shows the mutual relationship between thermal expansion coefficient, thermal frequency shift coefficient, and thermal speed change coefficient, which are defined in similar form.

Noninvasive temperature estimation in tissue via ultrasound echo-shifts. Part II. In vitro study.

Time shifts in echo signals returning from a heated volume of tissue correlate well with the temperature changes. In this study the relationship between these time shifts (or delays) and the tissue temperature was investigated in excised muscle tissue (turkey breast) as a possible dosimetric method. Heat was induced by the repeated activation of a sharply focused high-intensity ultrasound beam. Pulse echoes were sent and received with a confocal diagnostic transducer during the brief periods when the high-intensity ultrasonic beam was inactive. The change in transit time between echoes collected at different temperatures was estimated using cross-correlation techniques. With spatial-peak temporal-peak intensities (ISPTP) of less than 950W/cm2, the delay versus temperature relationship was fit to a linear equation with highly reproducible coefficients. The results confirmed that for spatial-peak temperature increases of approximately 10 degrees C, temperature-dependent changes in velocity were the single most important factor determining the observed delay, and a linear approximation could produce accurate temperature estimations. Nonlinear phenomena that occurred during the high-intensity irradiation had no significant effect on the measured delay. At ISPTP of 1115-2698 W/cm2, the delay-temperature relationship showed a similar monotonically decreasing pattern, but as the temperature peaked its slope gradually increased. This may reflect the curvilinear nature of the velocity-temperature relationship, but it may also be related to irreversible tissue modifications and to the use of the spatial-peak temperature to experimentally characterize the temperature changes. Overall, the results were consistent with theoretical predictions and encourage further experimental work to validate other aspects of the technique.