FEM Based 3D Tumor Growth Prediction for Kidney Tumor

Abstract

It is important to predict the tumor growth so that appropriate treatment can be planned in the early stage. In this letter, we propose a finite-element method (FEM)-based 3-D tumor growth prediction system using longitudinal kidney tumor images. To the best of our knowledge, this is the first kidney tumor growth prediction system. The kidney tissues are classified into three types: renal cortex, renal medulla, and renal pelvis. The reaction-diffusion model is applied as the tumor growth model. Different diffusion properties are considered in the model: the diffusion for renal medulla is considered as anisotropic, while those of renal cortex and renal pelvis are considered as isotropic. The FEM is employed to solve the diffusion model. Themodel parameters are estimated by the optimization of an objective function of overlap accuracy using a hybrid optimization parallel search package. The proposed method was tested on two longitudinal studies with seven time points on five tumors. The average true positive volume fraction and false positive volume fraction on all tumors is 91.4% and 4.0%, respectively. The experimental results showed the feasibility and efficacy of the proposed method.

Tumour growth prediction of follow‐up lung cancer via conditional recurrent variational autoencoder

“…The qualitative comparison results are shown as follows: Fig. 5 a shows the FEM results [1], Fig. 5 b shows the CEN results [7], Fig.…”

“…Besides using the generative model to predict tumour growth, the main distinction against [1, 6] is using the information from patients as constraints, such as smoking history and family disease status. In this way, while considering the common characteristics of tumour growth, it can also take into account the characteristics of a specific patient.…”

“…For longitudinal medical image research, physicians [3][4][5] usually set the baseline as the beginning of the study and obtain the screened image from the recruited lung cancer patients as experiment data. Among them, representative methods include finite element [1], mathematical modelling [5,6], and convolution neural network [7,8]. Chen et al [1] proposed a feasible finiteelement method (FEM) via a three-dimensional tumour growth prediction system using longitudinal kidney tumour images.…”

“…With the popularity of medical examinations and people's awareness of tumours has deepened, the longitudinal images are available and grow rapidly. Recently, deep learning and ConvNets [7, 8] have shown more fascinating and better performance than the mathematical modelling approach [1] when using a large‐scale dataset. Deep learning is a deep neural network that can learn the inherent representation from the sample data [12] and possesses extensibility to a certain extent.…”

“…Among them, representative methods include finite element [1], mathematical modelling [5, 6], and convolution neural network [7, 8]. Chen et al [1] proposed a feasible finite‐element method (FEM) via a three‐dimensional tumour growth prediction system using longitudinal kidney tumour images. David et al [5] developed the methods needed to accurately estimate tumour specific reaction–diffusion model parameters, and predicted the future gliomas tumour growth.…”