Development, validation, and pilot MRI safety study of a high-resolution, open source, whole body pediatric numerical simulation model

Modifying the trajectory of epicardial leads can substantially reduce MRI-induced RF heating in pediatric patients with a cardiac implantable electronic device at 1.5T

PURPOSE

After epicardial cardiac implantable electronic devices are implanted in pediatric patients, they become ineligible to receive MRI exams due to an elevated risk of RF heating. We investigated whether simple modifications in the trajectories of epicardial leads could substantially and reliably reduce RF heating during MRI at 1.5 T, with benefits extending to abandoned leads.

METHODS

Electromagnetic simulations were performed to assess RF heating of two common 35-cm epicardial lead trajectories exhibiting different degrees of coupling with MRI incident electric fields. Experiments in anthropomorphic phantoms implanted with commercial cardiac implantable electronic devices confirmed the findings. Both electromagnetic simulations and experimental measurements were performed using head-first and feet-first positioning and various landmarks. Transfer function approach was used to assess the performance of suggested modifications in realistic body models.

RESULTS

Simulations (head-first, chest landmark) of a 35-cm epicardial lead with a trajectory where the excess length of the lead was looped and placed on the inferior surface of the heart showed an 87-fold reduction in the 0.1 g-averaged specific absorption rate compared with the lead where the excess length was looped on the anterior surface. Repeated experiments with a commercial epicardial device confirmed this. For fully implanted systems following low-specific absorption rate trajectories, there was a 16-fold reduction in the average temperature rise and a 28-fold reduction for abandoned leads. The transfer function method predicted a 7-fold reduction in the RF heating in 336 realistic scenarios.

CONCLUSION

Surgical modification of epicardial lead trajectory can substantially reduce RF heating at 1.5 T, with benefits extending to abandoned leads.

Investigation and benchmarking of U-Nets on prostate segmentation tasks

In healthcare, a growing number of physicians and support staff are striving to facilitate personalized radiotherapy regimens for patients with prostate cancer. This is because individual patient biology is unique, and employing a single approach for all is inefficient. A crucial step for customizing radiotherapy planning and gaining fundamental information about the disease, is the identification and delineation of targeted structures. However, accurate biomedical image segmentation is time-consuming, requires considerable experience and is prone to observer variability. In the past decade, the use of deep learning models has significantly increased in the field of medical image segmentation. At present, a vast number of anatomical structures can be demarcated on a clinician's level with deep learning models. These models would not only unload work, but they can offer unbiased characterization of the disease. The main architectures used in segmentation are the U-Net and its variants, that exhibit outstanding performances. However, reproducing results or directly comparing methods is often limited by closed source of data and the large heterogeneity among medical images. With this in mind, our intention is to provide a reliable source for assessing deep learning models. As an example, we chose the challenging task of delineating the prostate gland in multi-modal images. First, this paper provides a comprehensive review of current state-of-the-art convolutional neural networks for 3D prostate segmentation. Second, utilizing public and in-house CT and MR datasets of varying properties, we created a framework for an objective comparison of automatic prostate segmentation algorithms. The framework was used for rigorous evaluations of the models, highlighting their strengths and weaknesses.

Effect of subject-specific head morphometry on specific absorption rate estimates in parallel-transmit MRI at 7 T

PURPOSE

To assess the accuracy of morphing an established reference electromagnetic head model to a subject-specific morphometry for the estimation of specific absorption rate (SAR) in 7T parallel-transmit (pTx) MRI.

METHODS

Synthetic T1 -weighted MR images were created from three high-resolution open-source electromagnetic head voxel models. The accuracy of morphing a "reference" (multimodal image-based detailed anatomical [MIDA]) electromagnetic model into a different subject's native space (Duke and Ella) was compared. Both linear and nonlinear registration methods were evaluated. Maximum 10-g averaged SAR was estimated for circularly polarized mode and for 5000 random RF shim sets in an eight-channel transmit head coil, and comparison made between the morphed MIDA electromagnetic models and the native Duke and Ella electromagnetic models, respectively.

RESULTS

The averaged error in maximum 10-g averaged SAR estimation across pTx MRI shim sets between the MIDA and the Duke target model was reduced from 17.5% with only rigid-body registration, to 11.8% when affine linear registration was used, and further reduced to 10.7% when nonlinear registration was used. The corresponding figures for the Ella model were 16.7%, 11.2%, and 10.1%.

CONCLUSION

We found that morphometry accounts for up to half of the subject-specific differences in pTx SAR. Both linear and nonlinear morphing of an electromagnetic model into a target subject improved SAR agreement by better matching head size, morphometry, and position. However, differences remained, likely arising from details in tissue composition estimation. Thus, the uncertainty of the head morphometry and tissue composition may need to be considered separately to achieve personalized SAR estimation.

Aluminum Thin Film Nanostructure Traces in Pediatric EEG Net for MRI and CT Artifact Reduction

Magnetic resonance imaging (MRI) and continuous electroencephalogram (EEG) monitoring are essential in the clinical management of neonatal seizures. EEG electrodes, however, can significantly degrade the image quality of both MRI and CT due to substantial metallic artifacts and distortions. Thus, we developed a novel thin film trace EEG net ("NeoNet") for improved MRI and CT image quality without compromising the EEG signal quality. The aluminum thin film traces were fabricated with an ultra-high-aspect ratio (up to 17,000:1, with dimensions 30 nm × 50.8 cm × 100 µm), resulting in a low density for reducing CT artifacts and a low conductivity for reducing MRI artifacts. We also used numerical simulation to investigate the effects of EEG nets on the B1 transmit field distortion in 3 T MRI. Specifically, the simulations predicted a 65% and 138% B1 transmit field distortion higher for the commercially available copper-based EEG net ("CuNet", with and without current limiting resistors, respectively) than with NeoNet. Additionally, two board-certified neuroradiologists, blinded to the presence or absence of NeoNet, compared the image quality of MRI images obtained in an adult and two children with and without the NeoNet device and found no significant difference in the degree of artifact or image distortion. Additionally, the use of NeoNet did not cause either: (i) CT scan artifacts or (ii) impact the quality of EEG recording. Finally, MRI safety testing confirmed a maximum temperature rise associated with the NeoNet device in a child head-phantom to be 0.84 °C after 30 min of high-power scanning, which is within the acceptance criteria for the temperature for 1 h of normal operating mode scanning as per the FDA guidelines. Therefore, the proposed NeoNet device has the potential to allow for concurrent EEG acquisition and MRI or CT scanning without significant image artifacts, facilitating clinical care and EEG/fMRI pediatric research.

A high-resolution pediatric female whole-body numerical model with comparison to a male model

Objective. Numerical models are central in designing and testing novel medical devices and in studying how different anatomical changes may affect physiology. Despite the numerous adult models available, there are only a few whole-body pediatric numerical models with significant limitations. In addition, there is a limited representation of both male and female biological sexes in the available pediatric models despite the fact that sex significantly affects body development, especially in a highly dynamic population. As a result, we developed Athena, a realistic female whole-body pediatric numerical model with high-resolution and anatomical detail.Approach. We segmented different body tissues through Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) images of a healthy 3.5 year-old female child using 3D Slicer. We validated the high anatomical accuracy segmentation through two experienced sub-specialty-certified neuro-radiologists and the inter and intra-operator variability of the segmentation results comparing sex differences in organ metrics with physiologic values. Finally, we compared Athena with Martin, a similar male model, showing differences in anatomy, organ metrics, and MRI dosimetric exposure.Main results. We segmented 267 tissue compartments, which included 50 brain tissue labels. The tissue metrics of Athena displayed no deviation from the literature value of healthy children. We show the variability of brain metrics in the male and female models. Finally, we offer an example of computing Specific Absorption Rate and Joule heating in a toddler/preschooler at 7 T MRI.Significance. This study introduces a female realistic high-resolution numerical model using MRI and CT scans of a 3.5 year-old female child, the use of which includes but is not limited to radiofrequency safety studies for medical devices (e.g. an implantable medical device safety in MRI), neurostimulation studies, and radiation dosimetry studies. This model will be open source and available on the Athinoula A. Martinos Center for Biomedical Imaging website.

Computational dosimetry in MRI in presence of hip, knee or shoulder implants: do we need accurate surgery models?

Objective.To quantify the effects of different levels of realism in the description of the anatomy around hip, knee or shoulder implants when simulating, numerically, radiofrequency and gradient-induced heating in magnetic resonance imaging. This quantification is needed to define how precise the digital human model modified with the implant should be to get realistic dosimetric assessments.Approach. The analysis is based on a large number of numerical simulations where four 'levels of realism' have been adopted in modelling human bodies carrying orthopaedic implants.Main results. Results show that the quantification of the heating due to switched gradient fields does not strictly require a detailed local anatomical description when preparing the digital human model carrying an implant. In this case, a simple overlapping of the implant CAD with the body anatomy is sufficient to provide a quite good and conservative estimation of the heating. On the contrary, the evaluation of the electromagnetic field distribution and heating caused by the radiofrequency field requires an accurate description of the tissues around the prosthesis.Significance. The results of this paper provide hints for selecting the 'level of realism' in the definition of the anatomical models with embedded passive implants when performing simulations that should reproduce, as closely as possible, thein vivoscenarios of patients carrying orthopaedic implants.

A comparative study of MRI-induced RF heating in pediatric and adult populations with epicardial and endocardial implantable electronic devices

Patients with congenital heart defects, inherited arrhythmia syndromes, and congenital disorders of cardiac conduction often receive a cardiac implantable electronic device (CIED). At least 75% of patients with CIEDs will need magnetic resonance imaging (MRI) during their lifetime. In 2011, the US Food and Drug Administration approved the first MR-conditional CIEDs for patients with endocardial systems, in which leads are passed through the vein and affixed to the endocardium. The majority of children, however, receive an epicardial CIED, where leads are directly sewn to the epicardium. Unfortunately, an epicardial CIED is a relative contraindication to MRI due to the unknown risk of RF heating. In this work, we performed anthropomorphic phantom experiments to investigate differences in RF heating between endocardial and epicardial leads in both pediatric and adult-sized phantoms, where adult endocardial CIED was the control. Clinical Relevance-This work provides a quantitative comparison of MRI RF heating of epicardial and endocardial leads in pediatric and adult populations.

The Position and Orientation of the Pulse Generator Affects MRI RF Heating of Epicardial Leads in Children

Infants and children with congenital heart defects often receive a cardiac implantable electronic device (CIED). Because transvenous access to the heart is difficult in patients with small veins, the majority of young children receive epicardial CIEDs. Unfortunately, however, once an epicardial CIED is placed, patients are no longer eligible to receive magnetic resonance imaging (MRI) exams due to the unknown risk of MRI-induced radiofrequency (RF) heating of the device. Although many studies have assessed the role of device configuration in RF heating of endocardial CIEDs in adults, such case for epicardial devices in pediatric patients is relatively unexplored. In this study, we evaluated the variation in RF heating of an epicardial lead due to changes in the lateral position and orientation of the implantable pulse generator (IPG). We found that changing the orientation and position of the IPG resulted in a five-fold variation in the RF heating at the lead's tip. Maximum heating was observed when the IPG was moved to a left lateral abdominal position of patient, and minimum heating was observed when the IPG was positioned directly under the heart. Clinical Relevance- This study examines the role of device configuration on MRI-induced RF heating of an epicardial CIED in a pediatric phantom. Results could help pediatric cardiac surgeons to modify device implantation to reduce future risks of MRI in patients.

Recent Applications of Artificial Intelligence in Radiotherapy: Where We Are and Beyond

In recent decades, artificial intelligence (AI) tools have been applied in many medical fields, opening the possibility of finding novel solutions for managing very complex and multifactorial problems, such as those commonly encountered in radiotherapy (RT). We conducted a PubMed and Scopus search to identify the AI application field in RT limited to the last four years. In total, 1824 original papers were identified, and 921 were analyzed by considering the phase of the RT workflow according to the applied AI approaches. AI permits the processing of large quantities of information, data, and images stored in RT oncology information systems, a process that is not manageable for individuals or groups. AI allows the iterative application of complex tasks in large datasets (e.g., delineating normal tissues or finding optimal planning solutions) and might support the entire community working in the various sectors of RT, as summarized in this overview. AI-based tools are now on the roadmap for RT and have been applied to the entire workflow, mainly for segmentation, the generation of synthetic images, and outcome prediction. Several concerns were raised, including the need for harmonization while overcoming ethical, legal, and skill barriers.

Numerical estimation of the B1 transmit field distortion in a copper EEG trace comparison with the thin-film based resistive trace "NeoNet"

This study investigates the effects of EEG traces in B1 transmit field distortion in a 3T MRI. EEG is a non-invasive method to monitor brain activities. Although EEG monitors brain activities with a high temporal resolution, it has trouble localizing the signal source. The EEG-fMRI is the multimodal imaging method, but care is needed to use EEG while in MRI as EEG traces create the signal distortion to the MRI. To tackle this problem, resistive traces are developed using thin-film technology to reduce the signal distortion during MRI. Numerical simulation was used to estimate the amount of B1 transmit field distortion of NeoNet and copper-based EEG nets (CuNet - with and without current limiting resistors) compared with the case without EEG net (NoNet). The reduced B1 transmit field distortion is estimated in the case of NeoNet compared to the CuNets. NeoNet is an MR-compatible high-density EEG net designed for pediatric subjects. The proposed NeoNet traces will facilitate/enable such EEG/fMRI pediatric studies with mitigated artifacts, which in turn will help to move the pediatric EEG/fMRI field forward.Clinical Relevance-This study estimates the benefit of the thin-film based EEG net with reduced B1 transmit artifact for the multimodal study of EEG-fMRI. The results are compared with commercial EEG trace made with copper metal with current limiting resistors. It is reported that about 470,000 children are suffering from Epilepsy. The MR-compatible resistive EEG traces se EEG-fMRI has potential to be a valuable tool to help understand pediatric Epilepsy and move the pediatric EEG-fMRI field forward.

Numerical simulation of the radiofrequency safety of 128-channel hd-EEG nets on a 29-month-old whole-body model in a 3 Tesla MRI

This study investigates the radiofrequency (RF) induced heating in a pediatric whole-body voxel model with a high-density electroencephalogram (hd-EEG) net during magnetic resonance imaging (MRI) at 3 Tesla. A total of three cases were studied: no net (NoNet), a resistive hd-EEG (NeoNet), and a copper (CuNet) net. The maximum values of specific absorption rate averaged over 10g-mass (10gSAR) in the head were calculated with the NeoNet was 12.51 W/kg and in the case of the NoNet was 12.40 W/kg. In contrast, the CuNet case was 17.04 W/Kg. Temperature simulations were conducted to determine the RF-induced heating without and with hd-EEG nets (NeoNet and CuNet) during an MRI scan using an age-corrected and thermoregulated perfusion for the child model. The results showed that the maximum temperature estimated in the child's head was 38.38 °C for the NoNet, 38.43 °C for the NeoNet, and 43.05 °C for the CuNet. In the case of NeoNet, the maximum temperature estimated in the child's head remained compliant with IEC 60601 for the MRI RF safety limit. However, the case of CuNet estimated to exceed the RF safety limit, which may require an appropriate cooling period or a hardware design to suppress the RF-induced heating.