Multi-Parametric Deep Learning Model for Prediction of Overall Survival after Postoperative Concurrent Chemoradiotherapy in Glioblastoma Patients

Abstract

This study aimed to investigate the performance of a deep learning-based survival-prediction model, which predicts the overall survival (OS) time of glioblastoma patients who have received surgery followed by concurrent chemoradiotherapy (CCRT). The medical records of glioblastoma patients who had received surgery and CCRT between January 2011 and December 2017 were retrospectively reviewed. Based on our inclusion criteria, 118 patients were selected and semi-randomly allocated to training and test datasets (3:1 ratio, respectively). A convolutional neural network–based deep learning model was trained with magnetic resonance imaging (MRI) data and clinical profiles to predict OS. The MRI was reconstructed by using four pulse sequences (22 slices) and nine images were selected based on the longest slice of glioblastoma by a physician for each pulse sequence. The clinical profiles consist of personal, genetic, and treatment factors. The concordance index (C-index) and integrated area under the curve (iAUC) of the time-dependent area-under-the-curve curves of each model were calculated to evaluate the performance of the survival-prediction models. The model that incorporated clinical and radiomic features showed a higher C-index (0.768 (95% confidence interval (CI): 0.759, 0.776)) and iAUC (0.790 (95% CI: 0.783, 0.797)) than the model using clinical features alone (C-index = 0.693 (95% CI: 0.685, 0.701); iAUC = 0.723 (95% CI: 0.716, 0.731)) and the model using radiomic features alone (C-index = 0.590 (95% CI: 0.579, 0.600); iAUC = 0.614 (95% CI: 0.607, 0.621)). These improvements to the C-indexes and iAUCs were validated using the 1000-times bootstrapping method; all were statistically significant (p < 0.001). This study suggests the synergistic benefits of using both clinical and radiomic parameters. Furthermore, it indicates the potential of multi-parametric deep learning models for the survival prediction of glioblastoma patients.

Deep learning features from diffusion tensor imaging improve glioma stratification and identify risk groups with distinct molecular pathway activities

“…Yoon HG et al. trained a deep CNN from 88 GBM patients for survival prediction and tested its performance in 30 patients [17] . Previous studies built their prediction model using conventional MR sequences (T1, T1c, T2, Flair) with limited sample size, while the use of CNN for survival prediction on DTI has not been investigated.…”

“…Lao et al developed a machine learning model combining both radiomics and deep features from standard MRI for survival prediction in GBM on a small cohort of 112 patients [16]. Yoon HG et al trained a deep CNN from 88 GBM patients for survival prediction and tested its performance in 30 patients [17]. Previous studies built their prediction model using conventional MR sequences (T1, T1c, T2, Flair) with limited sample size, while the use of CNN for survival prediction on DTI has not been investigated.…”

“…Recently, machine learning methods have been applied in extracting imaging features for prediction of clinical outcomes in glioma [ 6 , [12] , [13] , [14] , [15] , [16] , [17] ]. Specifically, there are two most popular imaging-based machine learning approaches: handcrafted radiomics analysis and convolutional neural networks (CNN).…”

A Survey on Deep Learning for Precision Oncology

“…Traditional cancer therapies like chemotherapy are cytotoxic to most cells, and thus they could damage healthy cells as well as cancer cells, while chemotherapy could be effective and a mainstay of cancer treatment for many patients, it also comes with the potential for many side effects. Figure 4 illustrates four kinds of deep-learning networks in this field, including Figure 4 a CNN [ 23 ], Figure 4 b DeepSurv [ 52 ], Figure 4 c residual CNN [ 41 ], and Figure 4 d survival recurrent network (SRN) [ 7 ].…”

“…Yoon et al [ 23 ] proposed a CNN to predict the overall survival time from MRI images of glioblastoma patients who had surgery and concurrent chemoradiation. As shown in Figure 4 a, their proposed method consists of an input layer, a hidden layer (composed of six convolution layers and six fully connected layers, some of which were followed by Leaky ReLU as the activation function and max-pooling), and output layer to predict the overall survival time.…”