First validation of a model-based hepatic percutaneous microwave ablation planning on a clinical dataset

A model-based planning tool, integrated in an imaging system, is envisioned for CT-guided percutaneous microwave ablation. This study aims to evaluate the biophysical model performance, by comparing its prediction retrospectively with the actual ablation ground truth from a clinical dataset in liver. The biophysical model uses a simplified formulation of heat deposition on the applicator and a heat sink related to vasculature to solve the bioheat equation. A performance metric is defined to assess how the planned ablation overlaps the actual ground truth. Results demonstrate superiority of this model prediction compared to manufacturer tabulated data and a significant influence of the vasculature cooling effect. Nevertheless, vasculature shortage due to branches occlusion and applicator misalignment due to registration error between scans affects the thermal prediction. With a more accurate vasculature segmentation, occlusion risk can be estimated, whereas branches can be used as liver landmarks to improve the registration accuracy. Overall, this study emphasizes the benefit of a model-based thermal ablation solution in better planning the ablation procedures. Contrast and registration protocols must be adapted to facilitate its integration into the clinical workflow.

Model-based hepatic percutaneous microwaveablation planning. First validation on a clinical dataset

A model-based planning tool, integrated in an imaging system, is envisioned for CT-guided percutaneous microwave ablation. This study aims to evaluate the biophysical model performance, by comparing its prediction retrospectively with the actualablation ground truth from a clinical data set in liver. The biophysical model uses a simplified formulation of heat depositionon the applicator and a heat sink related to vasculature to solve the bioheat equation. A performance metric is defined toassess how the planned ablation overlaps the actual ground truth. Results demonstrate superiority of this model predictioncompared to manufacturer tabulated data and a significant influence of the vasculature cooling effect. Nevertheless, vasculatureshortage due to branches occlusion and applicator misalignment due to registration error between scans affects the thermalprediction. With a more accurate vasculature segmentation, occlusion risk can be estimated, whereas branches can be usedas liver landmarks to improve the registration accuracy. Overall, this study emphasizes the benefit of a model-based thermalablation solution in better planning the ablation procedures. Contrast and registration protocols must be adapted to facilitate itsintegration into the clinical workflow.

Study of flow effects on temperature-controlled radiofrequency ablation using phantom experiments and forward simulations

PURPOSE

Blood flow is known to add variability to hepatic radiofrequency ablation (RFA) treatment outcomes. However, few studies exist on its impact on temperature-controlled RFA. Hence, we investigate large-scale blood flow effects on temperature-controlled RFA in flow channel experiments and numerical simulations.

METHODS

Ablation zones were induced in tissue-mimicking, thermochromic phantoms with a single flow channel, using an RF generator with temperature-controlled power delivery and a monopolar needle electrode. Channels were generated by molding the phantom around a removable rod. Channel radius and saline flow rate were varied to study the impact of flow on (i) the ablated cross-sectional area, (ii) the delivered generator power, and (iii) the occurrence of directional effects on the thermal lesion. Finite volume simulations reproducing the experimental geometry, flow conditions, and generator power input were conducted and compared to the experimental ablation outcomes.

RESULTS

Vessels of different channel radii r affected the ablation outcome in different ways. For r = 0.275  mm, the ablated area decreased with increasing flow rate while the energy input was hardly affected. For r = 0.9  mm and r = 2.3  mm, the energy input increased toward larger flow rates; for these radii, the ablated area decreased and increased toward larger flow rates, respectively, while still being reduced overall as compared to the reference experiment without flow. Directional effects, that is, local shrinking of the lesion upstream of the needle and an extension thereof downstream, were observed only for the smallest radius. The simulations qualitatively confirmed these observations. As compared to performing the simulations without flow, including flow effects in the simulations reduced the mean absolute error between experimental and simulated ablated areas from 0.23 to 0.12.

CONCLUSION

While the temperature control mechanism did not detect the heat sink effect in the case of the smallest channel radius, it counteracted the heat sink effect in the case of the larger channel radii with an increased energy input; this explains the increase in ablated area toward high flow rates (for r = 2.3  mm). The experiments in a simple phantom setup, thus, contribute to a good understanding of the phenomenon and are suitable for model validation.

Modeling the interference between shear and longitudinal waves under high intensity focused ultrasound propagation in bone

Magnetic resonance-guided high intensity focused ultrasound (MR-HIFU) is a noninvasive thermal technique that enables rapid heating of a specific area in the human body. Its clinical relevance has been proven for the treatments of soft tissue tumors, like uterine fibroids, and for the treatments of solid tumors in bone. In MR-HIFU treatment, MR-thermometry is used to monitor the temperature evolution in soft tissue. However, this technique is currently unavailable for bone tissue. Computer models can play a key role in the accurate prediction and monitoring of temperature. Here, we present a computer ray tracing model that calculates the heat production density in the focal region. This model accounts for both the propagation of shear waves and the interference between longitudinal and shear waves. The model was first compared with a finite element approach which solves the Helmholtz equation in soft tissue and the frequency-domain wave equation in bone. To obtain the temperature evolution in the focal region, the heat equation was solved using the heat production density generated by the raytracer as a heat source. Then, we investigated the role of the interaction between shear and longitudinal waves in terms of dissipated power and temperature output. The results of our model were in agreement with the results obtained by solving the Helmholtz equation and the frequency-domain wave equation, both in soft tissue and bone. Our results suggest that it is imperative to include both shear waves and their interference with longitudinal waves in the model when simulating high intensity focused ultrasound propagation in solids. In fact, when modeling HIFU treatments, omitting the interference between shear and longitudinal waves leads to an over-estimation of the temperature increase in the tissues.

Modelling the temperature evolution of bone under high intensity focused ultrasound

Magnetic resonance-guided high intensity focused ultrasound (MR-HIFU) has been clinically shown to be effective for palliative pain management in patients suffering from skeletal metastasis. The underlying mechanism is supposed to be periosteal denervation caused by ablative temperatures reached through ultrasound heating of the cortex. The challenge is exact temperature control during sonication as MR-based thermometry approaches for bone tissue are currently not available. Thus, in contrast to the MR-HIFU ablation of soft tissue, a thermometry feedback to the HIFU is lacking, and the treatment of bone metastasis is entirely based on temperature information acquired in the soft tissue adjacent to the bone surface. However, heating of the adjacent tissue depends on the exact sonication protocol and requires extensive modelling to estimate the actual temperature of the cortex. Here we develop a computational model to calculate the spatial temperature evolution in bone and the adjacent tissue during sonication. First, a ray-tracing technique is used to compute the heat production in each spatial point serving as a source term for the second part, where the actual temperature is calculated as a function of space and time by solving the Pennes bio-heat equation. Importantly, our model includes shear waves that arise at the bone interface as well as all geometrical considerations of transducer and bone geometry. The model was compared with a theoretical approach based on the far field approximation and an MR-HIFU experiment using a bone phantom. Furthermore, we investigated the contribution of shear waves to the heat production and resulting temperatures in bone. The temperature evolution predicted by our model was in accordance with the far field approximation and agreed well with the experimental data obtained in phantoms. Our model allows the simulation of the HIFU treatments of bone metastasis in patients and can be extended to a planning tool prior to MR-HIFU treatments.

Bone metastasis treatment using magnetic resonance-guided high intensity focused ultrasound

OBJECTIVES

Bone pain resulting from cancer metastases reduces a patient's quality of life. Magnetic Resonance-guided High Intensity Focused Ultrasound (MR-HIFU) is a promising alternative palliative thermal treatment technique for bone metastases that has been tested in a few clinical studies. Here, we describe a comprehensive pre-clinical study to investigate the effects, and efficacy of MR-HIFU ablation for the palliative treatment of osteoblastic bone metastases in rats.

MATERIALS AND METHODS

Prostate cancer cells (MATLyLu) were injected intra-osseously in Copenhagen rats. Upon detection of pain, as determined with a dynamic weight bearing (DWB) system, a MR-HIFU system was used to thermally ablate the bone region with tumor. Treatment effect and efficacy were assessed using magnetic resonance imaging (MRI), single-photon emission computed tomography (SPECT) with technetium-99m medronate ((99m)Tc-MDP), micro-computed tomography (μCT) and histology.

RESULTS

DWB analysis demonstrated that MR-HIFU-treated animals retained 58.6 ± 20.4% of limb usage as compared to 2.6 ± 6.3% in untreated animals (P=0.003). MR-HIFU delayed tumor specific growth rates (SGR) from 29 ± 6 to 13 ± 5%/day (P<0.001). Untreated animals (316.5 ± 78.9 mm(3)) had a greater accumulation of (99m)Tc-MDP than HIFU-treated animals (127.0 ± 42.7 mm(3), P=0.004). The total bone volume increase for untreated and HIFU-treated animals was 15.6 ± 9.6% and 3.0 ± 4.1% (P=0.004), respectively. Histological analysis showed ablation of nerve fibers, tumor, inflammatory and bone cells.

CONCLUSIONS

Our study provides a detailed characterization of the effects of MR-HIFU treatment on bone metastases, and provides fundamental data, which may motivate and advance its use in the clinical treatment of painful bone metastases with MR-HIFU.

The magnetic susceptibility effect of gadolinium-based contrast agents on PRFS-based MR thermometry during thermal interventions

Background

Proton resonance frequency shift (PRFS) magnetic resonance (MR) thermometry exploits the local magnetic field changes induced by the temperature dependence of the electron screening constant of water protons. Any other local magnetic field changes will therefore translate into incorrect temperature readings and need to be considered accordingly. Here, we investigated the susceptibility changes induced by the inflow and presence of a paramagnetic MR contrast agent and their implications on PRFS thermometry.

Methods

Phantom measurements were performed to demonstrate the effect of sudden gadopentetate dimeglumine (Gd-DTPA) inflow on the phase shift measured using a PRFS thermometry sequence on a clinical 3 T magnetic resonance-guided high-intensity focused ultrasound (MR-HIFU) system. By proton nuclear magnetic resonance spectroscopy, the temperature dependence of the Gd-DTPA susceptibility was measured, as well as the effect of liposomal encapsulation and release on the bulk magnetic susceptibility of Gd-DTPA. In vivo studies were carried out to measure the temperature error induced in a rat hind leg muscle upon intravenous Gd-DTPA injection.

Results

The phantom study showed a significant phase shift inside the phantom of 0.6 ± 0.2 radians (mean ± standard deviation) upon Gd-DTPA injection (1.0 mM, clinically relevant amount). A Gd-DTPA-induced magnetic susceptibility shift of Δχ_Gd-DTPA = 0.109 ppm/mM was measured in a cylinder parallel to the main magnetic field at 37°C. The temperature dependence of the susceptibility shift showed dΔχ_Gd-DTPA/dT = -0.00038 ± 0.00008 ppm/mM/°C. No additional susceptibility effect was measured upon Gd release from paramagnetic liposomes. In vivo, intravenous Gd-DTPA injection resulted in a perceived temperature change of 2.0°C ± 0.1°C at the center of the hind leg muscle.

Conclusions

The use of a paramagnetic MR contrast agent prior to MR-HIFU treatment may influence the accuracy of the PRFS MR thermometry. Depending on the treatment workflow, Gd-induced temperature errors ranging between -4°C and +3°C can be expected. Longer waiting time between contrast agent injection and treatment, as well as shortening the ablation duration by increasing the sonication power, will minimize the Gd influence. Compensation for the phase changes induced by the changing Gd presence is difficult as the magnetic field changes are arising nonlocally in the surroundings of the susceptibility change.

Comparison of inversion-recovery gradient- and spin-echo and fast spin-echo techniques in the detection and characterization of liver lesions

PURPOSE

To compare respiratory-triggered inversion-recovery (IR) gradient- and spin-echo (GRASE) magnetic resonance (MR) imaging with respiratory-triggered T2-weighted fast spin-echo (SE) imaging in the diagnosis of liver metastases.

MATERIALS AND METHODS

In this prospective study, two radiologists independently identified focal hepatic lesions on respiratory-triggered IR GRASE and respiratory-triggered fast SE MR images in 28 consecutive patients with 186 (135 malignant and 51 benign) proved lesions. A combination of findings at surgery, intraoperative ultrasonography (US), and histologic examination served as the standard of reference. Contrast-to-noise ratios (CNRs) were obtained from 86 lesions larger than 10 mm.

RESULTS

The sensitivity in the detection of liver metastases was, independent of lesion size and observer, higher for IR GRASE imaging (55%) than for fast SE imaging (44%-50%) (observer 1, P = .014; observer 2, P = .21). Confidence levels with IR GRASE imaging were higher, but not significantly so, than those with fast SE imaging (P < .098). Both observers characterized liver lesions better with IR GRASE than with fast SE imaging (observer 1, P = .04; observer 2, P = .48). The metastasis-liver CNR was significantly higher (P = .012) with IR GRASE imaging.

CONCLUSION

The respiratory-triggered IR GRASE sequence is a fast alternative to the respiratory-triggered fast SE sequence in the evaluation of suspected liver metastases.

[Detection of focal lung lesions with magnetic resonance tomography using T2-weighted ultrashort turbo-spin-echo-sequence in comparison with spiral computerized tomography]

PURPOSE

To compare spiral CT and MRT for the detection of focal pulmonary lesions.

PATIENTS AND METHODS

50 patients with focal pulmonary lesions confirmed by spiral CT were examined using a T2-weighted UTSE sequence (TE: 90 ms, TR: 1500-3000 ms, echo interval 9 ms, 8 mm slice thickness, diastolic triggering, expiratory breath gating). Image quality was compared using a 4-stage scale. Lesions with a minimum size of 2 mm were counted and measured in the CT image. The results were compared with the MRT images.

RESULTS

The image quality in CT examinations with an average value of 1.22 better than that in MRT (1.78). In total 163 pulmonary lesions with a size of 2-115 mm were found by CT. MRT found 151/163 lesions (92.6%). Of the 12 lesions not detected, 9 were smaller than 4 mm, 1 corresponded to a 12 mm large, completely calcified granuloma. In 2 cases there was a 4 or 5 mm large unspecific scar. Thus, 160/163 (98.1%) of all lesions larger than 3 mm were detected.

CONCLUSIONS

MRT with use of a suitable UTSE sequence is an alternative to CT for the detection of focal pulmonary lesions with a size larger than 3 mm.

Interventional breast MR imaging: clinical use of a stereotactic localization and biopsy device

PURPOSE

To evaluate the usefulness of preoperative magnetic resonance (MR) imaging-guided stereotactic localization and core biopsy of suspicious breast lesions that are visible at breast MR imaging alone (ie, that are clinically, mammographically, and ultrasonographically occult), with the goal of integrating this technique into the diagnostic and therapeutic work-up of MR-suspicious lesions in a clinical setting.

MATERIALS AND METHODS

A stereotactic breast biopsy device was used for needle placement in and guide wire localization of 97 lesions in 66 patients or core biopsy of five lesions in five patients; all lesions were visible at MR imaging. Interventions were performed with MR guidance on a 0.5- or 1.5-T system.

RESULTS

Lesion localization and resection were successful in 95 of the 97 lesions; two of the lesions were not resected in spite of correct guide wire localization. In this series, 53 (55%) of 97 lesions proved malignant (11 [21%] in situ; 42 [79%] invasive). Lesions were 4-19 mm (mean, 8.7 mm); all invasive cancers corresponded to a pT1 tumor stage. Location of the lesion in the parenchyma (retroareolar or prepectoral) did not interfere with accessibility.

CONCLUSION

MR imaging-guided stereotactic hook-wire placement and excisional biopsy are accurate and effective in managing lesions identified at only breast MR imaging. MR imaging-guided core biopsy holds promise for allowing a definite work-up of these lesions.

Breast neoplasms: T2* susceptibility-contrast, first-pass perfusion MR imaging

PURPOSE

To evaluate the differentiation of benign from malignant breast tumors with T2*-weighted perfusion magnetic resonance (MR) imaging (blood volume imaging) versus that with dynamic T1-weighted contrast agent-enhanced MR imaging.

MATERIALS AND METHODS

Ten healthy adult volunteers and 18 adult patients with benign or malignant lesions underwent both conventional T1-weighted dynamic contrast-enhanced breast MR imaging and repetitive first-pass, single-section, dynamic T2*-weighted perfusion MR imaging. Images were obtained before, during, and after injection of 20 mL of gadopentetate dimeglumine; peak gadopentetate dimeglumine concentrations were calculated from the maximal signal intensity loss on T2*-weighted images.

RESULTS

No perfusion effect was detectable in healthy breast parenchyma. A strong susceptibility-mediated signal intensity loss occurred in malignant breast tumors. No or only minor perfusion effects were seen in fibroadenomas, in spite of their rapid enhancement at T1-weighted dynamic imaging. Perfusion imaging was possible after conventional dynamic contrast-enhanced breast MR imaging.

CONCLUSION

T2*-weighted perfusion imaging exploits the susceptibility-mediated signal intensity loss of a first-pass bolus of gadopentetate dimeglumine within the capillary bed. First-pass perfusion imaging of breast lesions is feasible. It is promising in the differentiation of benign from malignant, rapidly enhancing lesions.

[MR mammography at 0.5 tesla. II. The capacity to differentiate malignant and benign lesions in MR mammography at 0.5 and 1.5 T]

PURPOSE

To determine whether medium field strength (0.5 T) MR mammography is able to differentiate between benign and malignant lesions in the same way as a 1.5 T standard technique.

MATERIAL

In 40 consecutive female patients with nodular lesions, examinations were carried out at 1.5 T (2D-FFE, TR/TE/FA 200/3.9/80) and 0.5 T (3D-FFE, 24/3.4/40).

RESULTS

There was wide spread of the speed of enhancement of malignant tumours for both techniques (44%-145% with signal increase in the first minute of 42%-189%). In 15 out of 17 carcinomas, the rapidity of uptake and final degree of enhancement after contrast was higher during 0.5 T/3D measurements than it was for 1.5 T/2D images. Fibroadenomas and mastopathies showed similar enhancement characteristics for both techniques. Sharply defined "wash-out" was seen only in malignant lesions; it appeared ten times more frequently in the 0.5 T/3D examinations than at 1.5 T/2D.

CONCLUSIONS

The distinction between benign and malignant lesions can be made with more certainty using medium field strength and 3D-FFE sequence than using the 2D high-field standard technique.

[MR mammography at 0.5 tesla. I. Comparison of image quality and sensitivity of MR mammography at 0.5 and 1.5 T]

PURPOSE

To determine whether dynamic MR mammography is possible on midfield systems without loss of diagnostic sensitivity when compared to the standard highfield technique.

MATERIALS AND METHODS

42 consecutive patients were examined twice: Once using the standard dynamic 2D gradient echo technique at 1.5 T; a second examination was performed on a 0.5 T system. For the midfield examinations a 3D sequence with optimized T1 contrast was used to compensate for the shorter T1 relaxation times at 0.5 T. Subtraction images were calculated to improve detectability of enhancing lesions.

RESULTS

Image quality was comparable on both systems. Mean enhancement of lesions was higher at 0.5 T/3D as compared to 1.5 T/2D (161% versus 112%). In malignant lesions, enhancement at 0.5 T/3D surpassed that at 1.5 T/2D in 88% of cases; average maximum signal intensity increase of cancers was significantly higher at 0.5 T/3D as compared to 1.5 T/2D (183% versus 108% relative to baseline). One satellite lesion of a recurrent carcinoma was detected on the 0.5 T/3D images only.

CONCLUSION

A 3D gradient echo pulse sequence can be used to compensate for the T1 shortening effect of the lower field strength. With a 3D sequence, sensitivity of MRM at 0.5 T is even superior to that of the standard 2D highfield technique.