Neural network training for cross-protocol radiomic feature standardization in computed tomography.
Ontology highlight
ABSTRACT:
SUBMITTER: Andrearczyk V
PROVIDER: S-EPMC6553566 | biostudies-literature | 2019 Apr
REPOSITORIES: biostudies-literature
Publications
Neural network training for cross-protocol radiomic feature standardization in computed tomography.
Andrearczyk Vincent V Depeursinge Adrien A Müller Henning H
Journal of medical imaging (Bellingham, Wash.) 20190401 2
Radiomics has shown promising results in several medical studies, yet it suffers from a limited discrimination and informative capability as well as a high variation and correlation with the tomographic scanner types, pixel spacing, acquisition protocol, and reconstruction parameters. We propose and compare two methods to transform quantitative image features in order to improve their stability across varying image acquisition parameters while preserving the texture discrimination abilities. In ...[more]
Similar Datasets
Project description:Variability in the x-ray tube current used in computed tomography may affect quantitative features extracted from the images. To investigate these effects, we scanned the Credence Cartridge Radiomics phantom 12 times, varying the tube current from 25 to 300?mA?s while keeping the other acquisition parameters constant. For each of the scans, we extracted 48 radiomic features from the categories of intensity histogram (n?=?10), gray-level run length matrix (n?=?11), gray-level co-occurrence matrix (n?=?22), and neighborhood gray tone difference matrix (n?=?5). To gauge the size of the tube current effects, we scaled the features by the coefficient of variation of the corresponding features extracted from images of non-small cell lung cancer tumors. Variations in the tube current had more effect on features extracted from homogeneous materials (acrylic, sycamore wood) than from materials with more tissue-like textures (cork, rubber particles). Thirty-eight of the 48 features extracted from acrylic were affected by current reductions compared with only 2 of the 48 features extracted from rubber particles. These results indicate that variable x-ray tube current is unlikely to have a large effect on radiomic features extracted from computed tomography images of textured objects such as tumors.
| S-EPMC5799381 | biostudies-literature
Project description:Radiomics, which extract large amount of quantification image features from diagnostic medical images had been widely used for prognostication, treatment response prediction and cancer detection. The treatment options for lung nodules depend on their diagnosis, benign or malignant. Conventionally, lung nodule diagnosis is based on invasive biopsy. Recently, radiomics features, a non-invasive method based on clinical images, have shown high potential in lesion classification, treatment outcome prediction.Lung nodule classification using radiomics based on Computed Tomography (CT) image data was investigated and a 4-feature signature was introduced for lung nodule classification. Retrospectively, 72 patients with 75 pulmonary nodules were collected. Radiomics feature extraction was performed on non-enhanced CT images with contours which were delineated by an experienced radiation oncologist.Among the 750 image features in each case, 76 features were found to have significant differences between benign and malignant lesions. A radiomics signature was composed of the best 4 features which included Laws_LSL_min, Laws_SLL_energy, Laws_SSL_skewness and Laws_EEL_uniformity. The accuracy using the signature in benign or malignant classification was 84% with the sensitivity of 92.85% and the specificity of 72.73%.The classification signature based on radiomics features demonstrated very good accuracy and high potential in clinical application.
| S-EPMC5798832 | biostudies-literature
Project description:PURPOSE:Scatter is a major factor degrading the image quality of cone beam computed tomography (CBCT). Conventional scatter correction strategies require handcrafted analytical models with ad hoc assumptions, which often leads to less accurate scatter removal. This study aims to develop an effective scatter correction method using a residual convolutional neural network (CNN). METHODS:A U-net based 25-layer CNN was constructed for CBCT scatter correction. The establishment of the model consists of three steps: model training, validation, and testing. For model training, a total of 1800 pairs of x-ray projection and the corresponding scatter-only distribution in nonanthropomorphic phantoms taken in full-fan scan were generated using Monte Carlo simulation of a CBCT scanner installed with a proton therapy system. An end-to-end CNN training was implemented with two major loss functions for 100 epochs with a mini-batch size of 10. Image rotations and flips were randomly applied to augment the training datasets during training. For validation, 200 projections of a digital head phantom were collected. The proposed CNN-based method was compared to a conventional projection-domain scatter correction method named fast adaptive scatter kernel superposition (fASKS) method using 360 projections of an anthropomorphic head phantom. Two different loss functions were applied for the same CNN to evaluate the impact of loss functions on the final results. Furthermore, the CNN model trained with full-fan projections was fine-tuned for scatter correction in half-fan scan by using transfer learning with additional 360 half-fan projection pairs of nonanthropomorphic phantoms. The tuned-CNN model for half-fan scan was compared with the fASKS method as well as the CNN-based method without the fine-tuning using additional lung phantom projections. RESULTS:The CNN-based method provides projections with significantly reduced scatter and CBCT images with more accurate Hounsfield Units (HUs) than that of the fASKS-based method. Root mean squared error of the CNN-corrected projections was improved to 0.0862 compared to 0.278 for uncorrected projections or 0.117 for the fASKS-corrected projections. The CNN-corrected reconstruction provided better HU quantification, especially in regions near the air or bone interfaces. All four image quality measures, which include mean absolute error (MAE), mean squared error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity (SSIM), indicated that the CNN-corrected images were significantly better than that of the fASKS-corrected images. Moreover, the proposed transfer learning technique made it possible for the CNN model trained with full-fan projections to be applicable to remove scatters in half-fan projections after fine-tuning with only a small number of additional half-fan training datasets. SSIM value of the tuned-CNN-corrected images was 0.9993 compared to 0.9984 for the non-tuned-CNN-corrected images or 0.9990 for the fASKS-corrected images. Finally, the CNN-based method is computationally efficient - the correction time for the 360 projections only took less than 5 s in the reported experiments on a PC (4.20 GHz Intel Core-i7 CPU) with a single NVIDIA GTX 1070 GPU. CONCLUSIONS:The proposed deep learning-based method provides an effective tool for CBCT scatter correction and holds significant value for quantitative imaging and image-guided radiation therapy.
| S-EPMC6684491 | biostudies-literature
Project description:Computed tomography perfusion (CTP) is used to diagnose ischemic strokes through contralateral hemisphere comparisons of various perfusion parameters. Various perfusion parameter thresholds have been utilized to segment infarct tissue due to differences in CTP software and patient baseline hemodynamics. This study utilized a convolutional neural network (CNN) to eliminate the need for non-universal parameter thresholds to segment infarct tissue. CTP data from 63 ischemic stroke patients was retrospectively collected and perfusion parameter maps were generated using Vitrea CTP software. Infarct ground truth labels were segmented from diffusion-weighted imaging (DWI) and CTP and DWI volumes were registered. A U-net based CNN was trained and tested five separate times using each CTP parameter (cerebral blood flow (CBF), cerebral blood volume (CBV), time-to-peak (TTP), mean-transit-time (MTT), delay time). 8,352 infarct slices were utilized with a 60:30:10 training:testing:validation split and Monte Carlo cross-validation was conducted using 20 iterations. Infarct volumes were reconstructed following segmentation from each CTP slice. Infarct spatial and volumetric agreement was compared between each CTP parameter and DWI. Spatial agreement metrics (Dice coefficient, positive predictive value) for each CTP parameter in predicting infarct volumes are: CBF=(0.67, 0.76), CBV=(0.44, 0.62), TTP=(0.60, 0.67), MTT=(0.58, 0.62), delay time=(0.57, 0.60). 95% confidence intervals for volume differences with DWI infarct are: CBF=14.3±11.5 mL, CBV=29.6±21.2 mL, TTP=7.7±15.2 mL, MTT=-10.7±18.6 mL, delay time=-5.7±23.6 mL. CBF is the most accurate CTP parameter in segmenting infarct tissue. Segmentation of infarct using a CNN has the potential to eliminate non-universal CTP contralateral hemisphere comparison thresholds.
| S-EPMC7946163 | biostudies-literature
Project description:ObjectivesRadiomics is a promising avenue in non-invasive characterisation of diffuse glioma. Clinical translation is hampered by lack of reproducibility across centres and difficulty in standardising image intensity in MRI datasets. The study aim was to perform a systematic review of different methods of MRI intensity standardisation prior to radiomic feature extraction.MethodsMEDLINE, EMBASE, and SCOPUS were searched for articles meeting the following eligibility criteria: MRI radiomic studies where one method of intensity normalisation was compared with another or no normalisation, and original research concerning patients diagnosed with diffuse gliomas. Using PRISMA criteria, data were extracted from short-listed studies including number of patients, MRI sequences, validation status, radiomics software, method of segmentation, and intensity standardisation. QUADAS-2 was used for quality appraisal.ResultsAfter duplicate removal, 741 results were returned from database and reference searches and, from these, 12 papers were eligible. Due to a lack of common pre-processing and different analyses, a narrative synthesis was sought. Three different intensity standardisation techniques have been studied: histogram matching (5/12), limiting or rescaling signal intensity (8/12), and deep learning (1/12)-only two papers compared different methods. From these studies, histogram matching produced the more reliable features compared to other methods of altering MRI signal intensity.ConclusionMultiple methods of intensity standardisation have been described in the literature without clear consensus. Further research that directly compares different methods of intensity standardisation on glioma MRI datasets is required.Key points• Intensity standardisation is a key pre-processing step in the development of robust radiomic signatures to evaluate diffuse glioma. • A minority of studies compared the impact of two or more methods. • Further research is required to directly compare multiple methods of MRI intensity standardisation on glioma datasets.
| S-EPMC9474349 | biostudies-literature
Project description:IntroductionAlthough positron emission tomography/computed tomography (PET/CT) is a common tool for measuring breast cancer (BC), subtypes are not automatically classified by it. Therefore, the purpose of this research is to use an artificial neural network (ANN) to evaluate the clinical subtypes of BC based on the value of the tumor marker.Materials and methodsIn our nuclear medical facility, 122 BC patients (training and testing) had 18F-fluoro-D-glucose (18F-FDG) PET/CT to identify the various subtypes of the disease. 18F-FDG-18 injections were administered to the patients before the scanning process. We carried out the scan according to protocol. Based on the tumor marker value, the ANN's output layer uses the Softmax function with cross-entropy loss to detect different subtypes of BC.ResultsWith an accuracy of 95.77%, the result illustrates the ANN model for K-fold cross-validation. The mean values of specificity and sensitivity were 0.955 and 0.958, respectively. The area under the curve on average was 0.985.ConclusionSubtypes of BC may be categorized using the suggested approach. The PET/CT may be updated to diagnose BC subtypes using the appropriate tumor maker value when the suggested model is clinically implemented.
| S-EPMC11309150 | biostudies-literature
Project description:INTRODUCTION:Pulmonary sarcoidosis is a rare heterogeneous lung disease of unknown aetiology, with limited treatment options. Phenotyping relies on clinical testing including visual scoring of chest radiographs. Objective radiomic measures from high-resolution computed tomography (HRCT) may provide additional information to assess disease status. As the first radiomics analysis in sarcoidosis, we investigate the potential of radiomic measures as biomarkers for sarcoidosis, by assessing 1) differences in HRCT between sarcoidosis subjects and healthy controls, 2) associations between radiomic measures and spirometry, and 3) trends between Scadding stages. METHODS:Radiomic features were computed on HRCT in three anatomical planes. Linear regression compared global radiomic features between sarcoidosis subjects (n=73) and healthy controls (n=78), and identified associations with spirometry. Spatial differences in associations across the lung were investigated using functional data analysis. A subanalysis compared radiomic features between Scadding stages. RESULTS:Global radiomic measures differed significantly between sarcoidosis subjects and controls (p<0.001 for skewness, kurtosis, fractal dimension and Geary's C), with differences in spatial radiomics most apparent in superior and lateral regions. In sarcoidosis subjects, there were significant associations between radiomic measures and spirometry, with a large association found between Geary's C and forced vital capacity (FVC) (p=0.008). Global radiomic measures differed significantly between Scadding stages (p<0.032), albeit nonlinearly, with stage IV having more extreme radiomic values. Radiomics explained 71.1% of the variability in FVC compared with 51.4% by Scadding staging alone. CONCLUSIONS:Radiomic HRCT measures objectively differentiate disease abnormalities, associate with lung function and identify trends in Scadding stage, showing promise as quantitative biomarkers for pulmonary sarcoidosis.
| S-EPMC7261239 | biostudies-literature
Project description:ObjectivesThis study aims to evaluate the diagnostic performance of machine-learning-based contrast-enhanced CT radiomic analysis for categorizing benign and malignant ovarian tumors.MethodsA total of 1,329 patients with ovarian tumors were randomly divided into a training cohort (N=930) and a validation cohort (N=399). All tumors were resected, and pathological findings were confirmed. Radiomic features were extracted from the portal venous phase images of contrast-enhanced CT. The clinical predictors included age, CA-125, HE-4, ascites, and margin of tumor. Both radiomics model (including selected radiomic features) and mixed model (incorporating selected radiomic features and clinical predictors) were constructed respectively. Six classifiers [k-nearest neighbor (KNN), support vector machines (SVM), random forest (RF), logistic regression (LR), multi-layer perceptron (MLP), and eXtreme Gradient Boosting (XGBoost)] were used for each model. The mean relative standard deviation (RSD) and area under the receiver operating characteristic curve (AUC) were applied to evaluate and select the best classifiers. Then, the performances of the two models with selected classifiers were assessed in the validation cohort.ResultsThe MLP classifier with the least RSD (1.21 and 0.53, respectively) was selected as the best classifier in both radiomics and mixed models. The two models with MLP classifier performed well in the validation cohort, with the AUCs of 0.91 and 0.96 and with accuracies (ACCs) of 0.83 and 0.87, respectively. The Delong test showed that the AUC of mixed model was statistically different from that of radiomics model (p<0.001).ConclusionsMachine-learning-based CT radiomic analysis could categorize ovarian tumors with good performance preoperatively. The mixed model with MLP classifier may be a potential tool in clinical applications.
| S-EPMC9395674 | biostudies-literature
Project description:Objectives: Determining the presence of extrathyroidal extension (ETE) is important for patients with papillary thyroid carcinoma (PTC) in selecting the proper surgical approaches. This study aimed to explore a radiomic model for preoperative prediction of ETE in patients with PTC. Methods: The study included 624 PTC patients (without ETE, n = 448; with minimal ETE, n = 52; with gross ETE, n = 124) whom were divided randomly into training (n = 437) and validation (n = 187) cohorts; all data were gathered between January 2016 and November 2017. Radiomic features were extracted from computed tomography (CT) images of PTCs. Key radiomic features were identified and incorporated into a radiomic signature. Combining the radiomic signature with clinical risk factors, a radiomic nomogram was constructed using multivariable logistic regression. Delong test was used to compare different receiver operating characteristic curves. Results: Five key radiomic features were incorporated into the radiomic signature, which were significantly associated with ETE (p < 0.001 for both cohorts) and slightly better than clinical model integrating significant clinical risk factors in the training cohort (area under the receiver operating characteristic curve (AUC), 0.791 vs. 0.778; F1 score, 0.729 vs. 0.714) and validation cohort (AUC, 0.772 vs. 0.756; F1 score, 0.710 vs. 0.692). The radiomic nomogram significantly improved predictive value in the training cohort (AUC, 0.837, p < 0.001; F1 score, 0.766) and validation cohort (AUC, 0.812, p = 0.024; F1 score, 0.732). Conclusions: The radiomic nomogram significantly improved the preoperative prediction of ETE in PTC patients. It indicated that radiomics could be a valuable method in PTC research.
| S-EPMC6736997 | biostudies-literature
Project description:Quantitative analysis of computed tomography (CT) radiomic features is an indirect measure of tumor heterogeneity, which has been associated with prognosis in human lung carcinoma. Canine lung tumors share similar features to human lung tumors and serve as a model in which to investigate the utility of radiomic features in differentiating tumor type and prognostication. The purpose of this study was to correlate first-order radiomic features from canine pulmonary tumors to histopathologic characteristics and outcome. Disease-free survival, overall survival time and tumor-specific survival were calculated as days from the date of CT scan. Sixty-seven tumors from 65 dogs were evaluated. Fifty-six tumors were classified as primary pulmonary adenocarcinomas and 11 were non-adenocarcinomas. All dogs were treated with surgical resection; 14 dogs received adjuvant chemotherapy. Second opinion histopathology in 63 tumors confirmed the histologic diagnosis in all dogs and further characterized 53 adenocarcinomas. The median overall survival time was longer (p = 0.004) for adenocarcinomas (339d) compared to non-adenocarcinomas (55d). There was wide variation in first-order radiomic statistics across tumors. Mean Hounsfield units (HU) ratio (p = 0.042) and median mean HU ratio (p = 0.042) were higher in adenocarcinomas than in non-adenocarcinomas. For dogs with adenocarcinoma, completeness of excision was associated with overall survival (p<0.001) while higher mitotic index (p = 0.007) and histologic score (p = 0.037) were associated with shorter disease-free survival. CT-derived tumor variables prognostic for outcome included volume, maximum axial diameter, and four radiomic features: integral total, integral total mean ratio, total HU, and max mean HU ratio. Tumor volume was also significantly associated with tumor invasion (p = 0.044). Further study of radiomic features in canine lung tumors is warranted as a method to non-invasively interrogate CT images for potential predictive and prognostic utility.
| S-EPMC8370631 | biostudies-literature