Search

ObjectiveThe purpose of this study was to evaluate the reliability and quality of radiomic features in glioblastoma multiforme (GBM) derived from tumor volumes obtained with semi-automated tumor segmentation software.Materials and methodsMR images of 45 GBM patients (29 males, 16 females) were downloaded from The Cancer Imaging Archive, in which post-contrast T1-weighted imaging and fluid-attenuated inversion recovery MR sequences were used. Two raters independently segmented the tumors using two semi-automated segmentation tools (TumorPrism3D and 3D Slicer). Regions of interest corresponding to contrast-enhancing lesion, necrotic portions, and non-enhancing T2 high signal intensity component were segmented for each tumor. A total of 180 imaging features were extracted, and their quality was evaluated in terms of stability, normalized dynamic range (NDR), and redundancy, using intra-class correlation coefficients, cluster consensus, and Rand Statistic.ResultsOur study results showed that most of the radiomic features in GBM were highly stable. Over 90% of 180 features showed good stability (intra-class correlation coefficient [ICC] ≥ 0.8), whereas only 7 features were of poor stability (ICC < 0.5). Most first order statistics and morphometric features showed moderate-to-high NDR (4 > NDR ≥1), while above 35% of the texture features showed poor NDR (< 1). Features were shown to cluster into only 5 groups, indicating that they were highly redundant.ConclusionThe use of semi-automated software tools provided sufficiently reliable tumor segmentation and feature stability; thus helping to overcome the inherent inter-rater and intra-rater variability of user intervention. However, certain aspects of feature quality, including NDR and redundancy, need to be assessed for determination of representative signature features before further development of radiomics.

Gaining the ability to audit the behavior of deep learning (DL) denoising models is of crucial importance to prevent potential hallucinations and adversarial clinical consequences. We present a preliminary version of AntiHalluciNet, which is designed to predict spurious structural components embedded in the residual noise from DL denoising models in low-dose CT and assess its feasibility for auditing the behavior of DL denoising models. We created a paired set of structure-embedded and pure noise images and trained AntiHalluciNet to predict spurious structures in the structure-embedded noise images. The performance of AntiHalluciNet was evaluated by using a newly devised residual structure index (RSI), which represents the prediction confidence based on the presence of structural components in the residual noise image. We also evaluated whether AntiHalluciNet could assess the image fidelity of a denoised image by using only a noise component instead of measuring the SSIM, which requires both reference and test images. Then, we explored the potential of AntiHalluciNet for auditing the behavior of DL denoising models. AntiHalluciNet was applied to three DL denoising models (two pre-trained models, RED-CNN and CTformer, and a commercial software, ClariCT.AI [version 1.2.3]), and whether AntiHalluciNet could discriminate between the noise purity performances of DL denoising models was assessed. AntiHalluciNet demonstrated an excellent performance in predicting the presence of structural components. The RSI values for the structural-embedded and pure noise images measured using the 50% low-dose dataset were 0.57 ± 31 and 0.02 ± 0.02, respectively, showing a substantial difference with a p-value < 0.0001. The AntiHalluciNet-derived RSI could differentiate between the quality of the degraded denoised images, with measurement values of 0.27, 0.41, 0.48, and 0.52 for the 25%, 50%, 75%, and 100% mixing rates of the degradation component, which showed a higher differentiation potential compared with the SSIM values of 0.9603, 0.9579, 0.9490, and 0.9333. The RSI measurements from the residual images of the three DL denoising models showed a distinct distribution, being 0.28 ± 0.06, 0.21 ± 0.06, and 0.15 ± 0.03 for RED-CNN, CTformer, and ClariCT.AI, respectively. AntiHalluciNet has the potential to predict the structural components embedded in the residual noise from DL denoising models in low-dose CT. With AntiHalluciNet, it is feasible to audit the performance and behavior of DL denoising models in clinical environments where only residual noise images are available. [ABSTRACT FROM AUTHOR]