MRSaiFE: An AI-based Approach Towards the Real-Time Prediction of Specific Absorption Rate

Artificial Intelligence for Neuroimaging in Pediatric Cancer

BACKGROUND/OBJECTIVES

Artificial intelligence (AI) is transforming neuroimaging by enhancing diagnostic precision and treatment planning. However, its applications in pediatric cancer neuroimaging remain limited. This review assesses the current state, potential applications, and challenges of AI in pediatric neuroimaging for cancer, emphasizing the unique needs of the pediatric population.

METHODS

A comprehensive literature review was conducted, focusing on AI's impact on pediatric neuroimaging through accelerated image acquisition, reduced radiation, and improved tumor detection. Key methods include convolutional neural networks for tumor segmentation, radiomics for tumor characterization, and several tools for functional imaging. Challenges such as limited pediatric datasets, developmental variability, ethical concerns, and the need for explainable models were analyzed.

RESULTS

AI has shown significant potential to improve imaging quality, reduce scan times, and enhance diagnostic accuracy in pediatric neuroimaging, resulting in improved accuracy in tumor segmentation and outcome prediction for treatment. However, progress is hindered by the scarcity of pediatric datasets, issues with data sharing, and the ethical implications of applying AI in vulnerable populations.

CONCLUSIONS

To overcome current limitations, future research should focus on building robust pediatric datasets, fostering multi-institutional collaborations for data sharing, and developing interpretable AI models that align with clinical practice and ethical standards. These efforts are essential in harnessing the full potential of AI in pediatric neuroimaging and improving outcomes for children with cancer.

Investigation of alternative RF power limit control methods for 0.5T, 1.5T, and 3T parallel transmission cardiac imaging: A simulation study

PURPOSE

To investigate safety and performance aspects of parallel-transmit (pTx) RF control-modes for a body coil atB 0 ≤ 3 T $$ {B}_0\le 3\mathrm{T} $$ .

METHODS

Electromagnetic simulations of 11 human voxel models in cardiac imaging position were conducted forB 0 = 0.5 T $$ {B}_0=0.5\mathrm{T} $$ ,1.5 T $$ 1.5\mathrm{T} $$ and3 T $$ 3\mathrm{T} $$ and a body coil with a configurable number of transmit channels (1, 2, 4, 8, 16). Three safety modes were considered: the 'SAR-controlled mode' (SCM), where specific absorption rate (SAR) is limited directly, a 'phase agnostic SAR-controlled mode' (PASCM), where phase information is neglected, and a 'power-controlled mode' (PCM), where the voltage amplitude for each channel is limited. For either mode, safety limits were established based on a set of 'anchor' simulations and then evaluated in 'target' simulations on previously unseen models. The comparison allowed to derive safety factors accounting for varying patient anatomies. All control modes were compared in terms of theB 1 + $$ {B}_1^{+} $$ amplitude and homogeneity they permit under their respective safety requirements.

RESULTS

Large safety factors (approximately five) are needed if only one or two anchor models are investigated but they shrink with increasing number of anchors. The achievableB 1 + $$ {B}_1^{+} $$ is highest for SCM but this advantage is reduced when the safety factor is included. PCM appears to be more robust against variations of subjects. PASCM performance is mostly in between SCM and PCM. Compared to standard circularly polarized (CP) excitation, pTx offers minorB 1 + $$ {B}_1^{+} $$ improvements if local SAR limits are always enforced.

CONCLUSION

PTx body coils can safely be used atB 0 ≤ 3 T $$ {B}_0\le 3\mathrm{T} $$ . Uncertainties in patient anatomy must be accounted for, however, by simulating many models.

Rapid prediction of MRI-induced RF heating of active implantable medical devices using machine learning

The interaction between an active implantable medical device and magnetic resonance imaging (MRI) radiofrequency (RF) fields can cause excessive tissue heating. Existing methods for predicting RF heating in the presence of an implant rely on either extensive phantom experiments or electromagnetic (EM) simulations with varying degrees of approximation of the MR environment, the patient, or the implant. On the contrary, fast MR thermometry techniques can provide a reliable real-time map of temperature rise in the tissue in the vicinity of conductive implants. In this proof-of-concept study, we examined whether a machine learning (ML) based model could predict the temperature increase in the tissue near the tip of an implanted lead after several minutes of RF exposure based on only a few seconds of experimentally measured temperature values. We performed phantom experiments with a commercial deep brain stimulation (DBS) system to train a fully connected feedforward neural network (NN) to predict temperature rise after ~3 minutes of scanning at a 3 T scanner using only data from the first 5 seconds. The NN effectively predicted ΔTmax-R2 = 0.99 for predictions in the test dataset. Our model also showed potential in predicting RF heating for other various scenarios, including a DBS system at a different field strength (1.5 T MRI, R2 = 0.87), different field polarization (1.2 T vertical MRI, R2 = 0.79), and an unseen implant (cardiac leads at 1.5 T MRI, R2 = 0.91). Our results indicate great potential for the application of ML in combination with fast MR thermometry techniques for rapid prediction of RF heating for implants in various MR environments.Clinical Relevance- Machine learning-based algorithms can potentially enable rapid prediction of MRI-induced RF heating in the presence of unknown AIMDs in various MR environments.

A pathway towards a two-dimensional, bore-mounted, volume body coil concept for ultra high-field magnetic resonance imaging

Lack of a body-sized, bore-mounted, radiofrequency (RF) body coil for ultrahigh field (UHF) magnetic resonance imaging (MRI) is one of the major drawbacks of UHF, hampering the clinical potential of the technology. Transmit field (B₁ ) nonuniformity and low specific absorption rate (SAR) efficiencies in UHF MRI are two challenges to be overcome. To address these problems, and ultimately provide a pathway for the full clinical potential of the modality, we have designed and simulated two-dimensional cylindrical high-pass ladder (2D c-HPL) architectures for clinical bore-size dimensions, and demonstrated a simplified proof of concept with a head-sized prototype at 7 T. A new dispersion relation has been derived and electromagnetic simulations were used to verify coil modes. The coefficient of variation (CV) for brain, cerebellum, heart, and prostate tissues after B₁ ⁺ shimming in silico is reported and compared with previous works. Three prototypes were designed in simulation: a head-sized, body-sized, and long body-sized coil. The head-sized coil showed a CV of 12.3%, a B₁ ⁺ efficiency of 1.33 μT/√W, and a SAR efficiency of 2.14 μT/√(W/kg) for brain simulations. The body-sized 2D c-HPL coil was compared with same-sized transverse electromagnetic (TEM) and birdcage coils in silico with a four-port circularly polarized mode excitation. Improved B₁ ⁺ uniformity (26.9%) and SAR efficiency (16% and 50% better than birdcage and TEM coils, respectively) in spherical phantoms was observed. We achieved a CV of 12.3%, 4.9%, 16.7%, and 2.8% for the brain, cerebellum, heart, and prostate, respectively. Preliminary imaging results for the head-sized coil show good agreement between simulation and experiment. Extending the 1D birdcage coil concept to 2D c-HPLs provides improved B₁ ⁺ uniformity and SAR efficiency.

Personalized local SAR prediction for parallel transmit neuroimaging at 7T from a single T1-weighted dataset

Purpose

Parallel RF transmission (PTx) is one of the key technologies enabling high quality imaging at ultra‐high fields (≥7T). Compliance with regulatory limits on the local specific absorption rate (SAR) typically involves over‐conservative safety margins to account for intersubject variability, which negatively affect the utilization of ultra‐high field MR. In this work, we present a method to generate a subject‐specific body model from a single T1‐weighted dataset for personalized local SAR prediction in PTx neuroimaging at 7T.

Methods

Multi‐contrast data were acquired at 7T (N = 10) to establish ground truth segmentations in eight tissue types. A 2.5D convolutional neural network was trained using the T1‐weighted data as input in a leave‐one‐out cross‐validation study. The segmentation accuracy was evaluated through local SAR simulations in a quadrature birdcage as well as a PTx coil model.

Results

The network‐generated segmentations reached Dice coefficients of 86.7% ± 6.7% (mean ± SD) and showed to successfully address the severe intensity bias and contrast variations typical to 7T. Errors in peak local SAR obtained were below 3.0% in the quadrature birdcage. Results obtained in the PTx configuration indicated that a safety margin of 6.3% ensures conservative local SAR estimates in 95% of the random RF shims, compared to an average overestimation of 34% in the generic “one‐size‐fits‐all” approach.

Conclusion

A subject‐specific body model can be automatically generated from a single T1‐weighted dataset by means of deep learning, providing the necessary inputs for accurate and personalized local SAR predictions in PTx neuroimaging at 7T.