검색 상세

조직유형에 기반한 인체 전도도 추정 방법

Estimation of Human Conductivity Based on Segmentation of Tissue Type

  • 박상민 (Sang Min Park, 대학원 바이오의공학과 바이오의공학)

초록/요약

Human tissue conductivity can be utilized as a crucial factor in various therapeutic planning systems, including electroceutical. This study aimed to propose a method for estimating human tissue conductivity that can be used in treatment planning systems. In this study, we constructed an electric field model for Tumor Treatment Fields (TTFields) therapy. A 3D electric field model of the human brain was created using patient medical images (DICOM). The model comprised five tissue types (skin, skull, cerebrospinal fluid, gray matter, and white matter), And a 3x3 electrode array was placed on all four sides of the model's surface. Subsequently, electrical stimulation simulations were conducted to obtain voltage calculations at each electrode array. Subsequently, electrical stimulation simulations were conducted to obtain calculated voltage at each electrode array. These results were compared to measured voltages, and tissue-specific conductivities were estimated within a specific range through iterative optimization. Tissue conductivity estimates were calculated for various tissue types through repeated simulations. To minimize the difference between voltage values calculated from the simulations and measured voltage values, a total of eight iterations were performed, resulting in the following estimated conductivities compared to the initially intended values: Skin: estimated conductivity of 0.4001 S/m (initially 0.42 S/m), Skull: estimated conductivity of 0.008 S/m (initially 0.008 S/m), Cerebrospinal fluid: estimated conductivity of 1.9399 S/m (initially 1.92 S/m), Gray matter: estimated conductivity of 0.4999 S/m (initially 0.48 S/m), White matter: estimated conductivity of 0.178 S/m (initially 0.22 S/m). There were differences between the initially intended conductivities and the estimated conductivities for each tissue type, with the magnitudes of these differences ranked in the following order: white matter, gray matter, cerebrospinal fluid, skull, and skin. This study successfully estimated human tissue conductivity within a certain margin of error, which is considered a significant advancement in reducing the uncertainty associated with measuring human tissue conductivity. Based on these research findings, it is anticipated that further studies on the accuracy and detailed analysis of tissue conductivity will be conducted, enhancing the precision of electric stimulation therapy planning systems and optimizing the treatment efficacy of stimulation therapy such as TTFields.

more

초록/요약

인체의 전도도는 전자약을 비롯한 다양한 치료 계획 시스템에 중요한 요소로 활용될 수 있다. 이 연구는 치료 계획 시스템에서 사용할 수 있는 인체 전도도를 추정하는 방법을 제시하는 것을 목표로 한다. 본 연구에서는 전기장 치료(Tumor Treatment Fields, TTFields)에 사용되는 전기장 모델을 제작하였다. 환자의 의료영상(DICOM)을 이용해 인체 두뇌의 3D 전기장 모델을 제작하였으며, 생성된 전기장 모델은 5개의 조직유형(피부, 두개골, 뇌척수액, 회백질, 백질)으로 구성하였으며, 모델 겉면에는 3x3형태의 전극 어레이가 모델의 앞, 뒤, 좌, 우 네 면에 배치하였다. 그 후 전기 자극 시뮬레이션을 통해 각 전극 어레이에서 계산된 전압을 획득하였으며, 이 결과를 측정된 전압과 비교하여, 모델의 각 조직유형별 전도도를 일정 범위 내에서 재구성을 통해 최적화된 조직유형별 전도도를 추정하였다. 본 연구에서는 반복 시뮬레이션을 통해 조직 유형에 따른 전도도 추정치를 계산했다. 시뮬레이션에서 계산된 전극 어레이의 전압값과 측정된 전압값의 차이를 최소화하기 위해 총 8번의 반복을 수행하여 다음과 같이 초기에 의도한 전도도 값과 추정된 전도도 값을 얻었다. 피부의 경우 초기 0.42 S/m에서 추정된 값은 0.4001 S/m, 두개골은 초기 0.008 S/m에서 추정된 값은 0.008 S/m, 뇌척수액은 초기 1.92 S/m에서 추정된 값은 1.9399 S/m, 회백질은 초기 0.48 S/m에서 추정된 값은 0.4999 S/m, 그리고 백질은 초기 0.22 S/m에서 추정된 값은 0.178 S/m. 각 조직 유형에 대한 초기 의도한 전도도와 추정된 전도도 간의 차이가 있었으며, 이러한 차이의 크기는 백질, 회백질, 뇌척수액, 두개골, 피부 순서로 나타났다. 본 연구를 통해 인체의 일정 오차범위 내 전도도를 추정하였으며, 이는 인체 전도도 측정의 불확실성을 감소시킬 수 있는 중요한 진전이라고 생각한다. 이 연구 결과를 기반으로, 인체 전도도 추정의 정확도 및 세분화된 전도도 연구가 진행될 것으로 예상되며, 이는 전기 자극 치료 계획 시스템의 정확성을 향상해 전기장 치료 등 전기 자극 치료의 치료 효과를 최적화하는 데 유용할 것으로 기대한다.

more

목차

초록················································································································i
1 배경 ··········································································································· 1
1.1 인체 전도도의 필요성 ················································································1
1.2 관련연구 ··································································································1
1.3 문제점 ·····································································································2
1.4 연구목표 ·································································································3
2 방법 ··········································································································4
2.1 전기장 모델 설정 생성··············································································4
2.2 전도도 설정·····························································································7
2.3 유한요소법을 통한 전기장 모델 계산 ·······················································8
2.4 시뮬레이션을 통한 전도도 최적화 ··························································12
3 연구 결과 ·································································································13
3.1 전도도 추정 결과·····················································································13
3.2 조직유형별 전도도에 따른 민감도 ···························································16
4 토론·············································· ···························································23
4.1 연구의 한계·····························································································23
4.2 연구의 발전 가능성·················································································25
5 결론··········································································································28
참고문헌·······································································································29

more
반출 Meta View 목록