Dataset Information

Non-Gaussian models of diffusion weighted imaging for detection and characterization of prostate cancer: a systematic review and meta-analysis.

ABSTRACT: The importance of Diffusion Weighted Imaging (DWI) in prostate cancer (PCa) diagnosis have been widely handled in literature. In the last decade, due to the mono-exponential model limitations, several studies investigated non-Gaussian DWI models and their utility in PCa diagnosis. Since their results were often inconsistent and conflicting, we performed a systematic review of studies from 2012 examining the most commonly used Non-Gaussian DWI models for PCa detection and characterization. A meta-analysis was conducted to assess the ability of each Non-Gaussian model to detect PCa lesions and distinguish between low and intermediate/high grade lesions. Weighted mean differences and 95% confidence intervals were calculated and the heterogeneity was estimated using the I2 statistic. 29 studies were selected for the systematic review, whose results showed inconsistence and an unclear idea about the actual usefulness and the added value of the Non-Gaussian model parameters. 12 studies were considered in the meta-analyses, which showed statistical significance for several non-Gaussian parameters for PCa detection, and to a lesser extent for PCa characterization. Our findings showed that Non-Gaussian model parameters may potentially play a role in the detection and characterization of PCa but further studies are required to identify a standardized DWI acquisition protocol for PCa diagnosis.

SUBMITTER: Brancato V

PROVIDER: S-EPMC6856159 | biostudies-literature | 2019 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Non-Gaussian models of diffusion weighted imaging for detection and characterization of prostate cancer: a systematic review and meta-analysis.

Brancato V V Cavaliere C C Salvatore M M Monti S S

Scientific reports 20191114 1

The importance of Diffusion Weighted Imaging (DWI) in prostate cancer (PCa) diagnosis have been widely handled in literature. In the last decade, due to the mono-exponential model limitations, several studies investigated non-Gaussian DWI models and their utility in PCa diagnosis. Since their results were often inconsistent and conflicting, we performed a systematic review of studies from 2012 examining the most commonly used Non-Gaussian DWI models for PCa detection and characterization. A meta ...[more]

PMID: 31728007

Similar Datasets

Project description:PurposeThe purpose of this study was to compare the performance of four diffusion models, including mono and bi-exponential both Gaussian and non-Gaussian models, in diffusion weighted imaging of rectal cancer.Material and methodsNineteen patients with rectal adenocarcinoma underwent MRI examination of the rectum before chemoradiation therapy including a 7 b-value diffusion sequence (0, 25, 50, 100, 500, 1000 and 2000 s/mm2) at a 1.5T scanner. Four different diffusion models including mono- and bi-exponential Gaussian (MG and BG) and non-Gaussian (MNG and BNG) were applied on whole tumor volumes of interest. Two different statistical criteria were recruited to assess their fitting performance, including the adjusted-R2 and Root Mean Square Error (RMSE). To decide which model better characterizes rectal cancer, model selection was relied on Akaike Information Criteria (AIC) and F-ratio.ResultsAll candidate models achieved a good fitting performance with the two most complex models, the BG and the BNG, exhibiting the best fitting performance. However, both criteria for model selection indicated that the MG model performed better than any other model. In particular, using AIC Weights and F-ratio, the pixel-based analysis demonstrated that tumor areas better described by the simplest MG model in an average area of 53% and 33%, respectively. Non-Gaussian behavior was illustrated in an average area of 37% according to the F-ratio, and 7% using AIC Weights. However, the distributions of the pixels best fitted by each of the four models suggest that MG failed to perform better than any other model in all patients, and the overall tumor area.ConclusionNo single diffusion model evaluated herein could accurately describe rectal tumours. These findings probably can be explained on the basis of increased tumour heterogeneity, where areas with high vascularity could be fitted better with bi-exponential models, and areas with necrosis would mostly follow mono-exponential behavior.

Project description:BackgroundMetastatic complications are a major cause of cancer-related morbidity, with up to 40% of cancer patients experiencing at least one brain metastasis. Earlier detection may significantly improve patient outcomes and overall survival. We investigated machine learning (ML) models for early detection of brain metastases based on diffusion weighted imaging (DWI) radiomics.MethodsLongitudinal diffusion imaging from 116 patients previously treated with stereotactic radiosurgery (SRS) for brain metastases were retrospectively analyzed. Clinical contours from 600 metastases were extracted from radiosurgery planning computed tomography, and rigidly registered to corresponding contrast enhanced-T1 and apparent diffusion coefficient (ADC) maps. Contralateral contours located in healthy brain tissue were used as control. The dataset consisted of (I) radiomic features using ADC maps, (II) radiomic feature change calculated using timepoints before the metastasis manifested on contrast enhanced-T1, (III) primary cancer, and (IV) anatomical location. The dataset was divided into training and internal validation sets using an 80/20 split with stratification. Four classification algorithms [Linear Support Vector Machine (SVM), Random Forest (RF), AdaBoost, and XGBoost] underwent supervised classification training, with contours labeled either 'control' or 'metastasis'. Hyperparameters were optimized towards balanced accuracy. Various model metrics (receiver operating characteristic curve area scores, accuracy, recall, and precision) were calculated to gauge performance.ResultsThe radiomic and clinical data set, feature engineering, and ML models developed were able to identify metastases with an accuracy of up to 87.7% on the training set, and 85.8% on an unseen test set. XGBoost and RF showed superior accuracy (XGBoost: 0.877±0.021 and 0.833±0.47, RF: 0.823±0.024 and 0.858±0.045) for training and validation sets, respectively. XGBoost and RF also showed strong area under the receiver operating characteristic curve (AUC) performance on the validation set (0.910±0.037 and 0.922±0.034, respectively). AdaBoost performed slightly lower in all metrics. SVM model generalized poorly with the internal validation set. Important features involved changes in radiomics months before manifesting on contrast enhanced-T1.ConclusionsThe proposed models using diffusion-based radiomics showed encouraging results in differentiating healthy brain tissue from metastases using clinical imaging data. These findings suggest that longitudinal diffusion imaging and ML may help improve patient care through earlier diagnosis and increased patient monitoring/follow-up. Future work aims to improve model classification metrics, robustness, user-interface, and clinical applicability.

Project description:PurposeTo compare integrated slice-specific dynamic shimming (iShim) diffusion weighted imaging (DWI) and single-shot echo-planar imaging (SS-EPI) DWI in image quality and pathological characterization of rectal cancer.Materials and methodsA total of 193 consecutive rectal tumor patients were enrolled for retrospective analysis. Among them, 101 patients underwent iShim-DWI (b = 0, 800, and 1600 s/mm2) and 92 patients underwent SS-EPI-DWI (b = 0, and 1000 s/mm2). Qualitative analyses of both DWI techniques was performed by two independent readers; including adequate fat suppression, the presence of artifacts and image quality. Quantitative analysis was performed by calculating standard deviation (SD) of the gluteus maximus, signal intensity (SI) of lesion and residual normal rectal wall, apparent diffusion coefficient (ADC) values (generated by b values of 0, 800 and 1600 s/mm2 for iShim-DWI, and by b values of 0 and 1000 s/mm2 for SS-EPI-DWI) and image quality parameters, such as signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of primary rectal tumor. For the primary rectal cancer, two pathological groups were divided according to pathological results: Group 1 (well-differentiated) and Group 2 (poorly differentiated). Statistical analyses were performed with p < 0.05 as significant difference.ResultsCompared with SS-EPI-DWI, significantly higher scores of image quality were obtained in iShim-DWI cases (P < 0.001). The SDbackground was significantly reduced on b = 1600 s/mm2 images and ADC maps of iShim-DWI. Both SNR and CNR of b = 800 s/mm2 and b = 1600 s/mm2 images in iShim-DWI were higher than those of b = 1000 s/mm2 images in SS-EPI-DWI. In primary rectal cancer of iShim-DWI cohort, SIlesion was significantly higher than SIrectum in both b = 800 and 1600 s/mm2 images. ADC values were significantly lower in Group 2 (0.732 ± 0.08) × 10- 3 mm2/s) than those in Group 1 ((0.912 ± 0.21) × 10- 3 mm2/s). ROC analyses showed significance of ADC values and SIlesion between the two groups.ConclusioniShim-DWI with b values of 0, 800 and 1600 s/mm2 is a promising technique of high image quality in rectal tumor imaging, and has potential ability to differentiate rectal cancer from normal wall and predicting pathological characterization.

Dataset Information

Non-Gaussian models of diffusion weighted imaging for detection and characterization of prostate cancer: a systematic review and meta-analysis.

Publications

Non-Gaussian models of diffusion weighted imaging for detection and characterization of prostate cancer: a systematic review and meta-analysis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets