Jianfeng Zheng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jianfeng Zheng has a total of 90 co-authors. Here are the top co-authors by number of shared publications:
- Ji Chen
University of Houston
49 shared publications - Wolfgang Kainz
High Performance Computing for MRI Safety, Jasper, GA, USA
24 shared publications - Qingyan Wang
Department of Electrical and Computer Engineering, University of Houston, Houston, TX, USA
13 shared publications Wolfgang Kainz
High Performance Computing for MRI Safety Jasper Georgia USA
13 shared publications- Yu Wang
Department of Electrical and Computer Engineering, University of Houston, Houston, TX, USA
13 shared publications - Ran Guo
Department of Electrical and Computer Engineering, University of Houston, Houston, TX, USA
7 shared publications
Latest Publications
- 16 January 2026
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 95 (5) , 2863-2872
Unintended Vagus Nerve Stimulation From Cuff Electrode During MRI : Combined Effects of Gradient and Radiofrequency Fields
Purpose: Emissions generated during magnetic resonance imaging (MRI)—including gradient coil induced electric fields and radiofrequency coil induced heating near nerve fiber—may alter neural activation inside patients. This study investigates the combined effects of these emissions on vagus nerve activation in the presence of cuff electrodes. Methods: Electromagnetic, thermal, and neurophysiological simulations were performed to quantify activation thresholds under MRI‐induced fields. The study examined the impact of gradient field exposure and RF‐induced heating, particularly for the trapezoidal waveform of the gradient coil with short pulse duration. Results: The results indicate that the presence of the cuff electrode significantly reduces the activation threshold under gradient field exposure, while RF‐induced heating further decreases the threshold for stimulations with short pulse durations. In some scenarios, the reduced neuron activation threshold can be lower than peripheral nerve stimulation limits defined in IEC 60601‐2‐33. Conclusion: These findings indicate the potential risk of unintended vagus nerve stimulation in MRI environments, emphasizing the need for safety considerations in patients with implantable vagus nerve stimulators.
- 16 January 2026
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 95 (5) , 2863-2872
Unintended Vagus Nerve Stimulation From Cuff Electrode During MRI : Combined Effects of Gradient and Radiofrequency Fields
Purpose: Emissions generated during magnetic resonance imaging (MRI)—including gradient coil induced electric fields and radiofrequency coil induced heating near nerve fiber—may alter neural activation inside patients. This study investigates the combined effects of these emissions on vagus nerve activation in the presence of cuff electrodes. Methods: Electromagnetic, thermal, and neurophysiological simulations were performed to quantify activation thresholds under MRI‐induced fields. The study examined the impact of gradient field exposure and RF‐induced heating, particularly for the trapezoidal waveform of the gradient coil with short pulse duration. Results: The results indicate that the presence of the cuff electrode significantly reduces the activation threshold under gradient field exposure, while RF‐induced heating further decreases the threshold for stimulations with short pulse durations. In some scenarios, the reduced neuron activation threshold can be lower than peripheral nerve stimulation limits defined in IEC 60601‐2‐33. Conclusion: These findings indicate the potential risk of unintended vagus nerve stimulation in MRI environments, emphasizing the need for safety considerations in patients with implantable vagus nerve stimulators.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 172078-172084
Effect of Orthopedic Plate Length on Peripheral Nerve Stimulation During MRI: Combined Gradient-Induced and RF-Thermal Effects
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 172078-172084
Open AccessEffect of Orthopedic Plate Length on Peripheral Nerve Stimulation During MRI: Combined Gradient-Induced and RF-Thermal Effects
- 11 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electromagnetic Compatibility
- Vol. 68 (1) , 21-29
Orthopedic Implants Induced Unintended Nerve Stimulation During Transcutaneous Electrical Nerve Stimulation Treatment
Transcutaneous electrical nerve stimulation (TENS) is a widely used modality for pain relief and neuromuscular rehabilitation. However, the presence of implantable orthopedic devices can significantly distort the surrounding electric field, potentially leading to unintended neuronal activation. This study examines the influence of such implants on nerve stimulation during TENS therapy through a combination of numerical modeling and experimental validation. Electromagnetic simulations and neurophysiological modeling using NEURON, based on an anatomically accurate human body model, demonstrate pronounced electric field distortions at the edges of implantable devices, resulting in a lowered activation threshold for adjacent neurons. These findings are corroborated by phantom experiments, which show substantial alterations in electric field distribution in the presence of orthopedic plates and screws. The results suggest that TENS application should maintain a safe distance from implantable devices to minimize the risk of inadvertent nerve activation.
- 11 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electromagnetic Compatibility
- Vol. 68 (1) , 21-29
Orthopedic Implants Induced Unintended Nerve Stimulation During Transcutaneous Electrical Nerve Stimulation Treatment
Transcutaneous electrical nerve stimulation (TENS) is a widely used modality for pain relief and neuromuscular rehabilitation. However, the presence of implantable orthopedic devices can significantly distort the surrounding electric field, potentially leading to unintended neuronal activation. This study examines the influence of such implants on nerve stimulation during TENS therapy through a combination of numerical modeling and experimental validation. Electromagnetic simulations and neurophysiological modeling using NEURON, based on an anatomically accurate human body model, demonstrate pronounced electric field distortions at the edges of implantable devices, resulting in a lowered activation threshold for adjacent neurons. These findings are corroborated by phantom experiments, which show substantial alterations in electric field distribution in the presence of orthopedic plates and screws. The results suggest that TENS application should maintain a safe distance from implantable devices to minimize the risk of inadvertent nerve activation.
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electromagnetic Compatibility
- Vol. 67 (5) , 1502-1509
A Combined RF and Gradient Network for MRI Safety Testing of Implantable Devices
- 1 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electromagnetic Compatibility
- Vol. 67 (5) , 1502-1509
A Combined RF and Gradient Network for MRI Safety Testing of Implantable Devices
- 18 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 204-209
RF-Induced Heating in Orthopedic Implants: In Silico Assessment of Surgical and Healed Models Under 1.5 T and 3 T MRI
This study investigates RF-induced heating in tibial nail implants using computational models (Duke, Ella, Fats) at 1.5T and 3T MRI. Post-surgery with bone adhesives and healed models were analyzed to assess the impact of implant depth, bone adhesive properties, and nail length on RF-induced heating. Results show higher heating may occur in surgical models, especially with synovial adhesives due to their conductivity. Heating decreased with implant depth, reflecting tissue conductivity variations from surface to core. Nail length influenced worst-case heating, varying across models and MRI field strengths. These findings highlight the importance of in silico simulations to label the implants, ensuring patient safety under MRI.
- 18 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 204-209
RF-Induced Heating in Orthopedic Implants: In Silico Assessment of Surgical and Healed Models Under 1.5 T and 3 T MRI
This study investigates RF-induced heating in tibial nail implants using computational models (Duke, Ella, Fats) at 1.5T and 3T MRI. Post-surgery with bone adhesives and healed models were analyzed to assess the impact of implant depth, bone adhesive properties, and nail length on RF-induced heating. Results show higher heating may occur in surgical models, especially with synovial adhesives due to their conductivity. Heating decreased with implant depth, reflecting tissue conductivity variations from surface to core. Nail length influenced worst-case heating, varying across models and MRI field strengths. These findings highlight the importance of in silico simulations to label the implants, ensuring patient safety under MRI.
- 11 August 2025
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 94 (6) , 2632-2642
Numerical and experimental study of RF‐induced heating of passive implantable medical devices at 5T MRI
Purpose: To assess the RF‐induced heating of orthopedic implants in a 5T whole‐body MRI system through electromagnetic simulations and experimental validation, with the goal of ensuring patient safety in ultra‐high field (UHF) MRI.Methods: Numerical and experimental studies were conducted to evaluate RF‐induced heating in five titanium screws (4–12 cm) inside a 60‐cm wide 5T whole‐body MRI scanner using the standard ASTM phantom. The temperature rise over 15 min was determined through full‐wave electromagnetic simulations and direct measurements. The Finite Difference Time Domain (FDTD) method was used to quantify the 1 g mass‐averaged specific absorption rate (pSAR1g) in 10 clinically relevant plate‐and‐screw configurations implanted in the Duke and Ella human body models at three anatomical locations: humerus, femur, and tibia.Results: In the phantom study, the 6 cm screw exhibited the highest SAR and temperature rise, demonstrating a resonance effect at 5T. However, in human body models, the worst‐case implant lengths shifted to 7–11 cm, highlighting the influence of tissue heterogeneity on resonance conditions. SAR values were also affected by the implant's position within the RF coil. The strong agreement between simulations and measurements validates the computational approach.Conclusion: This study systematically evaluates RF‐induced heating in orthopedic implants within the newly approved 5T whole‐body MRI, demonstrating that implant length, positioning, and surrounding media significantly impact heating risks. The findings highlight the necessity for updated MRI safety guidelines at UHF strengths, as implant safety conditions at 5T systems may differ from those at 1.5 and 3T MRI systems.
- 11 August 2025
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 94 (6) , 2632-2642
Numerical and experimental study of RF‐induced heating of passive implantable medical devices at 5T MRI
Purpose: To assess the RF‐induced heating of orthopedic implants in a 5T whole‐body MRI system through electromagnetic simulations and experimental validation, with the goal of ensuring patient safety in ultra‐high field (UHF) MRI.Methods: Numerical and experimental studies were conducted to evaluate RF‐induced heating in five titanium screws (4–12 cm) inside a 60‐cm wide 5T whole‐body MRI scanner using the standard ASTM phantom. The temperature rise over 15 min was determined through full‐wave electromagnetic simulations and direct measurements. The Finite Difference Time Domain (FDTD) method was used to quantify the 1 g mass‐averaged specific absorption rate (pSAR1g) in 10 clinically relevant plate‐and‐screw configurations implanted in the Duke and Ella human body models at three anatomical locations: humerus, femur, and tibia.Results: In the phantom study, the 6 cm screw exhibited the highest SAR and temperature rise, demonstrating a resonance effect at 5T. However, in human body models, the worst‐case implant lengths shifted to 7–11 cm, highlighting the influence of tissue heterogeneity on resonance conditions. SAR values were also affected by the implant's position within the RF coil. The strong agreement between simulations and measurements validates the computational approach.Conclusion: This study systematically evaluates RF‐induced heating in orthopedic implants within the newly approved 5T whole‐body MRI, demonstrating that implant length, positioning, and surrounding media significantly impact heating risks. The findings highlight the necessity for updated MRI safety guidelines at UHF strengths, as implant safety conditions at 5T systems may differ from those at 1.5 and 3T MRI systems.
- 23 May 2025
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 94 (4) , 1714-1730
Impact of lateral movement of patients on radiofrequency ‐induced heating of active and passive implantable medical devices at 3T MRI
Purpose: To investigate the implications of lateral patient position on radiofrequency (RF)–induced heating of active and passive implantable medical devices (AIMDs and PIMDs) in a wide‐bore 3T MRI system. Methods: In vitro simulations for two rod positions and three lateral phantom shifts were experimentally validated inside a wide‐bore 3T MRI scanner. Three commercially available AIMDs (40‐cm peripheral nerve stimulator [PNS], 45‐cm restorative neurostimulator, and 50‐cm cardiac rhythm management system) were analyzed. RF‐induced heating for AIMDs was assessed using the transfer function method, which combines transfer functions with tangential electric fields along clinically relevant device pathways for the Duke, Ella, and Fats human body models. Thirty‐minute temperature rises were numerically simulated near four orthopedic PIMDs (shoulder, hand‐wrist plate, femoral plate, hip replacement) inside human body models. Results: RF‐induced heating of the titanium rod varied with lateral phantom shifts due to changes in electric‐field distribution. Centrally positioned cardiac rhythm management and restorative neurostimulator devices exhibited less than 4°C and 1°C 95th percentile temperature rise variation, respectively. Meanwhile, some PNS devices, due to their peripheral placement in regions with varying electric‐field exposure, showed up to 8.8°C variation in 95th percentile temperature rise. Similarly, 1‐g mass‐averaged specific absorption rate near PIMDs increased by up to 50 W/Kg, and a 19.7% increase in temperature rise is noted. Conclusion: The increased flexibility in lateral patient position within wide‐bore MRI can increase RF‐induced heating with implications for MR conditional labeling, especially for PNS devices and PIMDs.
- 23 May 2025
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 94 (4) , 1714-1730
Impact of lateral movement of patients on radiofrequency ‐induced heating of active and passive implantable medical devices at 3T MRI
Purpose: To investigate the implications of lateral patient position on radiofrequency (RF)–induced heating of active and passive implantable medical devices (AIMDs and PIMDs) in a wide‐bore 3T MRI system. Methods: In vitro simulations for two rod positions and three lateral phantom shifts were experimentally validated inside a wide‐bore 3T MRI scanner. Three commercially available AIMDs (40‐cm peripheral nerve stimulator [PNS], 45‐cm restorative neurostimulator, and 50‐cm cardiac rhythm management system) were analyzed. RF‐induced heating for AIMDs was assessed using the transfer function method, which combines transfer functions with tangential electric fields along clinically relevant device pathways for the Duke, Ella, and Fats human body models. Thirty‐minute temperature rises were numerically simulated near four orthopedic PIMDs (shoulder, hand‐wrist plate, femoral plate, hip replacement) inside human body models. Results: RF‐induced heating of the titanium rod varied with lateral phantom shifts due to changes in electric‐field distribution. Centrally positioned cardiac rhythm management and restorative neurostimulator devices exhibited less than 4°C and 1°C 95th percentile temperature rise variation, respectively. Meanwhile, some PNS devices, due to their peripheral placement in regions with varying electric‐field exposure, showed up to 8.8°C variation in 95th percentile temperature rise. Similarly, 1‐g mass‐averaged specific absorption rate near PIMDs increased by up to 50 W/Kg, and a 19.7% increase in temperature rise is noted. Conclusion: The increased flexibility in lateral patient position within wide‐bore MRI can increase RF‐induced heating with implications for MR conditional labeling, especially for PNS devices and PIMDs.
- 15 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electromagnetic Compatibility
- Vol. 67 (3) , 761-769
Lead-Specific Optimization of Transfer Function Validation Pathways for Magnetic Resonance RF Safety Evaluation
This article presents a novel lead-specific validation method for magnetic resonance radiofrequency safety evaluations on active implantable medical devices (AIMDs). The proposed method designs a set of optimized validation pathways for the specific AIMD lead which achieve 1) high temperature rise or high induced voltages during the validation measurements with 2) minimally correlated electric fields along the pathways inside a electric field generating phantom box. The novel strategy 1) calculates the heating, or the induced voltage, for 603 preselected pathways and ranks these pathways based on temperature rise or voltage values and 2) performs matrix operations on these pathways to select pathways with minimally correlated tangential electric fields to each other. Two commercially available AIMDs were used to demonstrate the effectiveness of this novel strategy. Consequently, the strategy potentially provides a better transfer function validation for AIMD leads in terms of accuracy and completion.
- 15 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electromagnetic Compatibility
- Vol. 67 (3) , 761-769
Lead-Specific Optimization of Transfer Function Validation Pathways for Magnetic Resonance RF Safety Evaluation
This article presents a novel lead-specific validation method for magnetic resonance radiofrequency safety evaluations on active implantable medical devices (AIMDs). The proposed method designs a set of optimized validation pathways for the specific AIMD lead which achieve 1) high temperature rise or high induced voltages during the validation measurements with 2) minimally correlated electric fields along the pathways inside a electric field generating phantom box. The novel strategy 1) calculates the heating, or the induced voltage, for 603 preselected pathways and ranks these pathways based on temperature rise or voltage values and 2) performs matrix operations on these pathways to select pathways with minimally correlated tangential electric fields to each other. Two commercially available AIMDs were used to demonstrate the effectiveness of this novel strategy. Consequently, the strategy potentially provides a better transfer function validation for AIMD leads in terms of accuracy and completion.
- 18 February 2025
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 93 (6) , 2583-2594
Computational study of the effects of orthopedic plates on gradient‐induced peripheral nerve stimulation under MRI using electromagnetic and neurophysiological modeling
Purpose: This study investigates the impact of orthopedic plates on peripheral nerve stimulation (PNS) during MRI, focusing on how the presence of implants affects PNS thresholds. Methods: A combination of anatomical electromagnetic and neurophysiological modeling was utilized. Electromagnetic fields in human body models were simulated, both with and without orthopedic plates. Simulations were performed under x ‐, y ‐, and z ‐axis gradient coils, with plates implanted at two clinically relevant locations. Nerve responses were modeled using an established neurophysiological model. Results: The presence of orthopedic implants significantly influenced nerve stimulation, leading to reductions in stimulation thresholds of up to 80%. Some of the reduced thresholds were close to the PNS limits outlined in International Electrotechnical Commission (IEC) 60601‐2‐33, suggesting a considerably reduced safety margin compared to cases without implants. Conclusion: Orthopedic implants can substantially lower the activation thresholds of nearby nerves, with some thresholds approaching the PNS limits defined in IEC 60601‐2‐33 for MRI gradient field. This finding indicates a reduced safety margin for patients with implants, highlighting the need for more comprehensive safety assessments.
- 18 February 2025
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 93 (6) , 2583-2594
Computational study of the effects of orthopedic plates on gradient‐induced peripheral nerve stimulation under MRI using electromagnetic and neurophysiological modeling
Purpose: This study investigates the impact of orthopedic plates on peripheral nerve stimulation (PNS) during MRI, focusing on how the presence of implants affects PNS thresholds. Methods: A combination of anatomical electromagnetic and neurophysiological modeling was utilized. Electromagnetic fields in human body models were simulated, both with and without orthopedic plates. Simulations were performed under x ‐, y ‐, and z ‐axis gradient coils, with plates implanted at two clinically relevant locations. Nerve responses were modeled using an established neurophysiological model. Results: The presence of orthopedic implants significantly influenced nerve stimulation, leading to reductions in stimulation thresholds of up to 80%. Some of the reduced thresholds were close to the PNS limits outlined in International Electrotechnical Commission (IEC) 60601‐2‐33, suggesting a considerably reduced safety margin compared to cases without implants. Conclusion: Orthopedic implants can substantially lower the activation thresholds of nearby nerves, with some thresholds approaching the PNS limits defined in IEC 60601‐2‐33 for MRI gradient field. This finding indicates a reduced safety margin for patients with implants, highlighting the need for more comprehensive safety assessments.