Feasibility of prior‐informed domain‐based electrical impedance tomography for deep tissue thermometry: Phantom experiments and computed tomography‐derived numerical simulations

BACKGROUND: Accurate monitoring of deep tissue temperature is important during hyperthermia treatment. Because the electrical conductivity of biological tissue generally increases with temperature, Electrical Impedance Tomography (EIT) may provide a non-invasive approach for estimating temperature changes in deep tissue. PURPOSE: This study aimed to evaluate a domain-based EIT thermometry framework for quantitative temperature estimation in electrically heterogeneous settings composed of piecewise homogeneous domains, where conventional voxel-based EIT reconstruction is limited by the ill-posed inverse problem and inaccurate separation of domain-level conductivity changes. METHODS: Simulation and phantom studies were performed. In the simulation study, reconstruction performance was evaluated using reconstruction error and stability. In the phantom experiments, performance was evaluated using the mean and maximum absolute temperature errors over reconstructed frames by comparing sensor-measured temperatures with EIT-based estimates. An anatomically realistic model was additionally constructed from a single respiratory phase of one patient computed tomography (CT) dataset, and nine anatomical subdomains were reconstructed under target-centered and external-entry heating scenarios. RESULTS: In the simulation study, the proposed domain-based reconstruction yielded the lowest reconstruction error across the evaluated noise conditions. In phantom experiments, it showed close agreement with direct measurements under all tested conditions. The average conductivity reconstruction error in the homogeneous phantom was 0.29%, while the mean absolute temperature errors in the multi-domain phantoms ranged from 0.22°C to 0.71°C. In pooled phantom comparisons, the proposed method showed mean and maximum absolute temperature errors of 0.44 ± 0.31°C and 0.91 ± 0.77°C, respectively. In the CT-derived model, the mean absolute temperature errors were 0.50°C and 1.43°C for the target-centered and external-entry scenarios, compared with 1.21-1.25°C and 2.60-2.67°C for the benchmark methods, respectively, and the location of the hottest region was preserved in both scenarios. CONCLUSION: The proposed domain-based EIT thermometry demonstrated feasibility in phantom experiments and a CT-derived anatomical model, establishing a foundation for its potential application as a non-invasive temperature monitoring approach.

Authors

Institutions

Publication Details

Journal
Medical Physics
Published
2026-11-18
DOI
https://doi.org/10.1002/mp.70774
Primary Topic
Electrical and Bioimpedance Tomography
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Feasibility of prior‐informed domain‐based electrical impedance tomography for deep tissue thermometry: Phantom experiments and computed tomography‐derived numerical simulations

Yongha Gi, Bumsoo Lee, Sung Hwan Ahn, Myonggeun Yoon et al.
Medical Physics
Electrical and Bioimpedance Tomography
article

Feasibility of prior‐informed domain‐based electrical impedance tomography for deep tissue thermometry: Phantom experiments and computed tomography‐derived numerical simulations

Yongha Gi, Bumsoo Lee, Sung Hwan Ahn, Myonggeun Yoon, JongHyun Kim, Jinyoung Hong, Byung‐Jou Lee, Yunhui Jo
article en

Abstract

BACKGROUND: Accurate monitoring of deep tissue temperature is important during hyperthermia treatment. Because the electrical conductivity of biological tissue generally increases with temperature, Electrical Impedance Tomography (EIT) may provide a non-invasive approach for estimating temperature changes in deep tissue. PURPOSE: This study aimed to evaluate a domain-based EIT thermometry framework for quantitative temperature estimation in electrically heterogeneous settings composed of piecewise homogeneous domains, where conventional voxel-based EIT reconstruction is limited by the ill-posed inverse problem and inaccurate separation of domain-level conductivity changes. METHODS: Simulation and phantom studies were performed. In the simulation study, reconstruction performance was evaluated using reconstruction error and stability. In the phantom experiments, performance was evaluated using the mean and maximum absolute temperature errors over reconstructed frames by comparing sensor-measured temperatures with EIT-based estimates. An anatomically realistic model was additionally constructed from a single respiratory phase of one patient computed tomography (CT) dataset, and nine anatomical subdomains were reconstructed under target-centered and external-entry heating scenarios. RESULTS: In the simulation study, the proposed domain-based reconstruction yielded the lowest reconstruction error across the evaluated noise conditions. In phantom experiments, it showed close agreement with direct measurements under all tested conditions. The average conductivity reconstruction error in the homogeneous phantom was 0.29%, while the mean absolute temperature errors in the multi-domain phantoms ranged from 0.22°C to 0.71°C. In pooled phantom comparisons, the proposed method showed mean and maximum absolute temperature errors of 0.44 ± 0.31°C and 0.91 ± 0.77°C, respectively. In the CT-derived model, the mean absolute temperature errors were 0.50°C and 1.43°C for the target-centered and external-entry scenarios, compared with 1.21-1.25°C and 2.60-2.67°C for the benchmark methods, respectively, and the location of the hottest region was preserved in both scenarios. CONCLUSION: The proposed domain-based EIT thermometry demonstrated feasibility in phantom experiments and a CT-derived anatomical model, establishing a foundation for its potential application as a non-invasive temperature monitoring approach.

Medical PhysicsVol. 53(12)
Korea University (KR), Samsung Medical Center (KR), Inje University Ilsan Paik Hospital (KR)
Electrical and Bioimpedance Tomography
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon