Finite element 3D models of melanoma growth and time-dependent backscattered data for dielectric properties of melanoma at 6 GHz
Eric Lindström, Larisa Beilina · 2025
Read Finite element 3D models of melanoma growth and time-dependent backscattered data for dielectric properties of melanoma at 6 GHz by Eric Lindström, Larisa Beilina (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-07794.txt.
Introduction
Eric Lindström and Larisa Beilina build 3D computer models of skin melanoma growth in 2025. The models use the finite element method, a standard way to solve physics on a mesh. They include true electric traits of skin tissue. The aim is to make fake sensor data that mimics what a scan would record. Such data let other teams test methods that rebuild a tumor's electric traits from scattered waves. The work focuses on one wave rate used in imaging research.
Catalog record
| CREATOR | Eric Lindström, Larisa Beilina |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 14 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-07794
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Finite element 3D models of melanoma growth and time-dependent backscattered data for dielectric properties of melanoma at 6 GHz Eric Lindström, Larisa Beilina Abstract Finite element meshes for 3D models simulating realistic malignant melanoma (MM) growth, incorporating accurate dielectric properties of the skin, have been developed. Numerical simulations illustrate how 3D finite element meshes can be utilized to generate backscattered data, enabling the evaluation of reconstruction algorithms designed to determine the dielectric properties of the proposed 3D model. Keywords— Finite element mesh, finite element method, Maxwell’s equations, dielectric properties of skin, malignant melanoma, microwave imaging MSC codes : 65J22; 65K10; 65M32; 65M55; 65M60; 65M70 1 Introduction This study presents finite element meshes for 3D models that simulate the growth of realistic malignant melanoma (MM). These models are integrated with accurate representations of skin properties corresponding to the regions where MM develops. To generate finite element meshes we combine geometrical models of melanoma growth presented recently in [ 8 ] with skin models studied in [ 7 ] . One of the key applications of the developed models is microwave medical imaging [ 1 ] for detecting malignant melanoma, the most lethal form of skin cancer, despite accounting for only 1% of all skin cancer cases [ 8 , 1 ] . The prognosis for MM is closely linked to the depth of primary tumor invasion in the skin [ 16 , 7 , 8 ] . Microwave medical imaging stands out as a non-invasive imaging technique, offering a significant advantage over methods such as X-ray imaging [ 15 , 32 ] . This makes it a compelling addition to existing medical imaging modalities like ultrasound [ 19 , 20 ] , X-ray, and MRI [ 21 ] . Notably, X-ray and MRI are not utilized for diagnosing primary skin cancers. Consequently, microwave imaging emerges as an attractive non-invasive technology that can be applied directly to the skin [ 8 ] . Developing this technology holds great promise as a valuable complement to current diagnostic approaches for MM. Previous studies [ 8 , 26 , 34 ] have highlighted variations in permittivity contrasts between malignant and normal tissues, demonstrating that malignant tumors exhibit higher relative permittivity values compared to normal tissues at frequencies below 10 GHz. However, accurately estimating the relative permittivity of the skin’s internal structure with malignant melanoma (MM) remains a significant challenge. This estimation relies on analyzing backscattered electromagnetic waves collected by detectors positioned near or on the skin affected by MM. In this study, we have modeled MM using realistic dielectric properties measured at 6 GHz, based on data from [ 8 ] . Potential application of the finite element meshes presented in this work is qualitative and quantitative malign melanoma detection using microwaves. One possibility to do this is to use an adaptive domain decomposition finite element/finite difference method (ADDFE/FDM) developed recently in [ 5 ] . This method was applied in [ 5 , 6 , 31 ] for reconstruction of dielectric permittivity and conductivity functions at 6 GHz for anatomically realistic breast phantom from database of [ 34 ] , as well as in [ 29 ] for reconstructing the dielectric properties of MM for another halph-spheric 3D melanoma model. Additional challenge is apply real measured backscattered microwave data at different frequencies in order to use generated finite element meshes in the ADDFE/FDM method for qualitative and quantitative MM detection. This challenge can be considered as a future work which should be performed in collaboration with electrical engineers. The problem of determination of the spatially distributed relative dielectric permittivity and conductivity functions using backscattered data of the electric field collected at the boundary of the investigated domain in scientific community is called Coefficient Inverse Problems (CIPs). We refer to [ 2 ] , [ 3 ] , [ 17 ] , [ 18 ] , [ 24 ] , [ 25 ] , [ 27 ] and references therein for mathematical methods related to the solution of CIPs using minimization of the functional, and to [ 3 , 14 , 22 , 23 , 28 ] for globally convergent methods for solution of CIPs. In the numerical studies of the current work we demonstrate how the developed 3D finite element mesh with MM can be applied in the ADDFE/FDM method of [ 5 ] to generate backscattered data. This data serves as a testing platform for reconstruction algorithms aimed at determining the dielectric properties of the proposed 3D models. Figure 1: Melanoma model shapes for different depths. 2 Melanoma growth models In this study we are taking geometrical models of MM growth developed in [ 8 ] and combine them with accurate representation of skin properties as was proposed in [ 7 ] . The prognosis of MM is related to the diameter and depth of MM invasion [ 9 , 10 ] and we modelled geometrical properties and shapes of MM accordingly to Figure 1 and Tables 1, 2. Next, we combined models shown on Figure 1 with skin properties described in the Table 2 . We note that this table highlights the dielectric properties of skin with MM at 6 GHz. However, the depth of all tissue types are consistent across all frequencies. Let the computational domain \Omega\subset\mathbb{R}^{3} be a convex domain with a smooth boundary \Gamma . To generate finite element meshes we choose \Omega of the size 10\times 10\times 10 mm such that the dimensionless domain is \Omega=\left\{x=(x_{1},x_{2},x_{3})\in(0,10)\times(0,10)\times(0,10)\right\}. Next, we decompose \Omega into tetrahedral elements with the mesh size h=0.5 such that every tetrahedral element has type of material corresponding to the tissue type of the Tables 1 , 2 , and assign then values for the relative dielectric permittivity \varepsilon_{r} and effective conductivity \sigma accordingly to the Tables 1 , 2 . Figure 2 presents finite element models of MM corresponding to the geometrical models presented in Figure 1 and Table 1 combined with realistic values of skin shown in Table 2 . Using Figure 2 and Tables 1 , 2 we observe that MM models for months 0,2,4,6,8,10 correspond to MM "in-situ" [ 9 , 10 ] and these MM are not invasive since the transition from non-invasive to invasive MM starts from the diameter of MM d=6 mm [ 8 ] . Next, MM models for months 12,14,16,18 are located in epidermis and dermis and they are invasive [ 11 ] . Finally, MM models for month 20,22 are propagated from dermis to fat and can reach the lymphatic system where MM begins to metastasize [ 13 ] . Month 0 Month 2 Month 4 Month 6 Month 8 Month 10 Month 12 Month 14 Month 16 Month 18 Month 20 Month 22 Figure 2: Models used for calculations visualized based on geometric properties shown in Table 1 . Month 0 Month 2 Month 4 Month 6 Month 8 Month 10 Month 12 Month 14 Month 16 Month 18 Month 20 Month 22 Figure 3: Finite element meshes for models presented on Figure 2 . Table 1: Tumor model geometrical properties Month Shape Diameter (mm) Depth (mm) 0 Cylinder 6 0.15 2 Cone 6.09 0.34 4 Cone 6.19 0.71 6 Cone 6.28 1.08 8 Cone 6.36 1.46