Histogram Metrics Research Articles

Comparison of MRI Features of Fat Fraction and ADC for Early Treatment Response Assessment in Participants with Multiple Myeloma

Background An imaging-based predictor of response could provide prognostic information early during treatment course in patients with multiple myeloma (MM). Purpose To investigate if very early changes in bone marrow relative fat fraction (rFF) and apparent diffusion coefficient (ADC) histogram metrics, occurring after one cycle of induction therapy in participants with newly diagnosed MM, could help predict overall best response status. Materials and Methods This prospective study included participants with MM who were enrolled between August 2014 and December 2017. Histogram metrics were extracted from ADC and rFF maps from MRI examinations performed before treatment and after the first treatment cycle. Participants were categorized into the very good partial response (VGPR) or better group and the less than VGPR group per the International Myeloma Working Group response criteria. ADC and rFF map metrics for predicting treatment response were compared using the Wilcoxon rank test, and the false discovery rate (FDR) was used to correct for multiple comparisons. Results A total of 23 participants (mean age, 65 years ± 11 [SD]; 13 men) were evaluated. There was no evidence of a difference in ADC metrics between the two responder groups after correcting for multiple comparisons. The rFF histogram changes between pretreatment MRI and MRI after the first treatment cycle (ΔrFF) that provided significant differences between the VGPR or better and less than VGPR groups were as follows: ΔrFF_10th Percentile (median, 0.5 [95% CI: 0, 1] vs -2.5 [95% CI: -5.1, 0.1], respectively), ΔrFF_90th Percentile (median, 2 [95% CI: 1, 6.8] vs -0.5 [95% CI: -1, 0]), ΔrFF_Mean (median, 3.4 [95% CI: 0.3, 7.6] vs -1.1 [95% CI: -1.8, -0.7]), and ΔrFF_Root Mean Squared (median, 3.2 [95% CI: 0.3, 6.1] vs -0.7 [95% CI: -1.3, -0.4]) (FDR-adjusted P = .03 for all), and the latter two also presented mean group increases in the VGPR or better group that were above the upper 95% CI limit for repeatability. Conclusion Very early changes in bone marrow relative fat fraction histogram metrics, calculated from MRI examination at baseline and after only one cycle of induction therapy, may help to predict very good partial response or better in participants with newly diagnosed multiple myeloma. © RSNA, 2022 Online supplemental material is available for this article.

Autosegmentation based on different-sized training datasets of consistently-curated volumes and impact on rectal contours in prostate cancer radiation therapy.

Background and purposeAutosegmentation techniques are emerging as time-saving means for radiation therapy (RT) contouring, but the understanding of their performance on different datasets is limited. The aim of this study was to determine agreement between rectal volumes by an existing autosegmentation algorithm and manually-delineated rectal volumes in prostate cancer RT. We also investigated contour quality by different-sized training datasets and consistently-curated volumes for retrained versions of this same algorithm.Materials and methodsSingle-institutional data from 624 prostate cancer patients treated to 50–70 Gy were used. Manually-delineated clinical rectal volumes (clinical) and consistently-curated volumes recontoured to one anatomical guideline (reference) were compared to autocontoured volumes by a commercial autosegmentation tool based on deep-learning (v1; n = 891, multiple-institutional data) and retrained versions using subsets of the curated volumes (v32/64/128/256; n = 32/64/128/256). Evaluations included dose-volume histogram metrics, Dice similarity coefficients, and Hausdorff distances; differences between groups were quantified using parametric or non-parametric hypothesis testing.ResultsVolumes by v1-256 (76–78 cm3) were larger than reference (75 cm3) and clinical (76 cm3). Mean doses by v1-256 (24.2–25.2 Gy) were closer to reference (24.2 Gy) than to clinical (23.8 Gy). Maximum doses were similar for all volumes (65.7–66.0 Gy). Dice for v1-256 and reference (0.87–0.89) were higher than for v1-256 and clinical (0.86–0.87) with corresponding Hausdorff comparisons including reference smaller than comparisons including clinical (5–6 mm vs. 7–8 mm).ConclusionUsing small single-institutional RT datasets with consistently-defined rectal volumes when training autosegmentation algorithms created contours of similar quality as the same algorithm trained on large multi-institutional datasets.

195P Radiomics and dosiomics signature from whole lung predicts radiation pneumonitis: A model development study with prospective external validation and decision-curve analysis

Radiation pneumonitis (RP) is one of the common side effects of radiotherapy in the thoracic region. Radiomics and dosiomics quantifies information implicit within medical images and radiotherapy dose distributions. In this study we demonstrated the prognostic potential of radiomics, dosiomics, and clinical features for RP prediction. Radiomics, dosiomics, dose-volume histogram (DVH) metrics, and clinical parameters were obtained on 314 retrospectively-collected and 35 prospectively-enrolled patients diagnosed with lung cancer between 2013 to 2019. A radiomics risk score (R-score) and dosiomics risk score (D-score) and DVH-score were calculated based on logistic regression after feature selection. Seven models were built using different combinations of R-score, D-score, and clinical parameters to evaluate their added prognostic power. The DVH model was built as a classical model for comparison with the dosiomics model. Over-optimism was evaluated by bootstrap resampling from the training set, and the prospectively-collected cohort was used as the external test set. Model calibration and decision-curve characteristics of the best-performing models were evaluated. For ease of further evaluation, nomograms were constructed for selected models. A model built by integrating all of R-score, D-score, and clinical parameters had the best discriminative ability with area under the curves (AUCs) of 0.793 (95%CI 0.735-0.851),0.774 (95%CI 0.762-0.786), and 0.855 (95%CI 0.719-0.990) in the training set, bootstrapping set, and external test set, respectively. The results of the calibration curve and decision curve analysis showed that the final model of the nomogram has potential for future clinical application.Table: 195PModelTrain (95%CI)Validation by bootstrapping (95%CI)Testing (95%CI)R-score0.676 (0.606-0.745)0.619 (0.592-0.646)0.671 (0.558-0.899)D-score0.728 (0.665-0.790)0.687 (0.667-0.706)0.684 (0.573-0.883)DVH-score0.637 (0.570-0.705)0.628 (0.613-0.642)0.661 (0.551-0.856)Clinical parameters0.664 (0.594-0.735)0.654 (0.628-0.680)0.709 (0.509-0.91)R-score + D-score0.735 (0.673-0.796)0.729 (0.720-0.736)0.739 (0.553-0.926)R-score + C0.717 (0.652-0.782)0.701 (0.683-0.719)0.771 (0.585-0.962)D-score + C0.770 (0.710-0.830)0.755 (0.744-0.765)0.756 (0.559-0.954)R-score + D-score + C0.793 (0.735-0.851)0.774 (0.762-0.786)0.855 (0.719-0.990) Open table in a new tab Radiomics and dosiomics features have potential to assist with the prediction of RP.

Comparative study of conventional ROI-based and volumetric histogram analysis derived from CT enhancement in differentiating malignant and benign renal tumors.

To quantitatively compare the diagnostic values of conventional region of interest (ROI)-based and volumetric histogram analysis derived from CT enhancement in differentiating malignant and benign renal tumors. A total of 230 patients with pathologically confirmed renal tumors who had undergone CT enhancement were classified into clear cell renal cell carcinoma (ccRCC) (n = 133), non-ccRCC (n = 56), and benign renal tumor(n = 41) group. Parametric CT enhancement of each tumor from volumetric histogram were obtained using in-house software, including 10th percentile, 25th percentile, median, 75th percentile, 90th percentile, mean, standard deviation, as well as skewness, kurtosis and entropy, and histogram metrics among these groups were analyzed. ROI-based enhancement density was also analyzed. The entropy and SD values of ccRCCs were higher than those of non-ccRCCs and benign renal tumors (p < 0.05). The 10th percentile, 25th percentile, median, 75th percentile, 90th percentile and mean values of ccRCCs were lower than those of benign renal tumors, however, higher than those of non-ccRCCs (p < 0.05). The ROI-based enhancement density of non-ccRCCs were lower than those of ccRCCs and benign renal tumors(p < 0.05). Receiver operating characteristic (ROC) curve analyses showed that entropy and mean values had the highest diagnostic efficacy in differentiating ccRCCs/non-ccRCCs and benign renal tumors. ROC curve analyses showed that mean values had the highest diagnostic efficacy in differentiating ccRCCs and non-ccRCCs. In terms of pairwise comparisons of ROC curves and diagnostic efficacy, ROI-based CT enhancement density was worse than volumetric histogram analysis (p < 0.05). Volumetric histogram analysis parameters can effectively distinguish malignant and benign renal tumors. 1. Entropy and mean values had the highest diagnostic efficacy in differentiating ccRCCs/ non-ccRCCs and benign renal tumors.2. Mean values had the highest diagnostic efficacy in differentiating ccRCCs and non-ccRCCs.3.Volumetric histogram analysis had better performance than ROI-based enhancement density.

ADC Histogram Analysis of Pediatric Low-Grade Glioma Treated with Selumetinib: A Report from the Pediatric Brain Tumor Consortium.

Selumetinib is a promising MAP (mitogen-activated protein) kinase (MEK) 1/2 inhibitor treatment for pediatric low-grade gliomas. We hypothesized that MR imaging-derived ADC histogram metrics would be associated with survival and response to treatment with selumetinib. Children with recurrent, refractory, or progressive pediatric low-grade gliomas who had World Health Organization grade I pilocytic astrocytoma with KIAA1549-BRAF fusion or the BRAF V600E mutation (stratum 1), neurofibromatosis type 1-associated pediatric low-grade gliomas (stratum 3), or sporadic non-neurofibromatosis type 1 optic pathway and hypothalamic glioma (OPHG) (stratum 4) were treated with selumetinib for up to 2 years. Quantitative ADC histogram metrics were analyzed for total and enhancing tumor volumes at baseline and during treatment. Each stratum comprised 25 patients. Stratum 1 responders showed lower values of SD of baseline ADC_total as well as a larger decrease with time on treatment in ADC_total mean, mode, and median compared with nonresponders. Stratum 3 responders showed a greater longitudinal decrease in ADC_total. In stratum 4, higher baseline ADC_total skewness and kurtosis were associated with shorter progression-free survival. When all 3 strata were combined, responders showed a greater decrease with time in ADC_total mode and median. Compared with sporadic OPHG, neurofibromatosis type 1-associated OPHG had lower values of ADC_total mean, mode, and median as well as ADC_enhancement mean and median and higher values of ADC_total skewness and kurtosis at baseline. The longitudinal decrease in ADC_total median during treatment was significantly greater in sporadic OPHG compared with neurofibromatosis type 1-associated OPHG. ADC histogram metrics are associated with progression-free survival and response to treatment with selumetinib in pediatric low-grade gliomas.

The Value of Whole-Tumor Histogram and Texture Analysis Using Intravoxel Incoherent Motion in Differentiating Pathologic Subtypes of Locally Advanced Gastric Cancer.

PurposeTo determine if whole-tumor histogram and texture analyses using intravoxel incoherent motion (IVIM) parameters values could differentiate the pathologic characteristics of locally advanced gastric cancer.MethodsEighty patients with histologically confirmed locally advanced gastric cancer who received surgery in our institution were retrospectively enrolled into our study between April 2017 and December 2018. Patients were excluded if they had lesions with the smallest diameter < 5 mm and severe image artifacts. MR scanning included IVIM sequences (9 b values, 0, 20, 40, 60, 100, 150,200, 500, and 800 s/mm2) used in all patients before treatment. Whole tumors were segmented by manually drawing the lesion contours on each slice of the diffusion-weighted imaging (DWI) images (with b=800). Histogram and texture metrics for IVIM parameters values and apparent diffusion coefficient (ADC) values were measured based on whole-tumor volume analyses. Then, all 24 extracted metrics were compared between well, moderately, and poorly differentiated tumors, and between different Lauren classifications, signet-ring cell carcinomas, and other poorly cohesive carcinomas using univariate analyses. Multivariate logistic analyses and multicollinear tests were used to identify independent influencing factors from the significant variables of the univariate analyses to distinguish tumor differentiation and Lauren classifications. ROC curve analyses were performed to evaluate the diagnostic performance of these independent influencing factors for determining tumor differentiation and Lauren classifications and identifying signet-ring cell carcinomas. The interobserver agreement was also conducted between the two observers for image quality evaluations and parameter metric measurements.ResultsFor diagnosing tumor differentiation, the ADCmedian, pure diffusion coefficient median (Dslowmedian), and pure diffusion coefficient entropy (Dslowentropy) showed the greatest AUCs: 0.937, 0.948, and 0.850, respectively, and no differences were found between the three metrics, P>0.05). The 95th percentile perfusion factor (FP P95th) was the best metric to distinguish diffuse-type GCs vs. intestinal/mixed (AUC=0.896). The ROC curve to distinguish signet-ring cell carcinomas from other poorly cohesive carcinomas showed that the Dslowmedian had AUC of 0.738. For interobserver reliability, image quality evaluations showed excellent agreement (interclass correlation coefficient [ICC]=0.85); metrics measurements of all parameters indicated good to excellent agreement (ICC=0.65-0.89), except for the Dfast metric, which showed moderate agreement (ICC=0.41-0.60).ConclusionsThe whole-tumor histogram and texture analyses of the IVIM parameters based on the biexponential model provided a non-invasive method to discriminate pathologic tumor subtypes preoperatively in patients with locally advanced gastric cancer. The metric FP P95th derived from IVIM performed better in determining Lauren classifications than the mono-exponential model.

Personalized Automation of Treatment Planning for Linac-Based Stereotactic Body Radiotherapy of Spine Cancer.

Purpose/Objective(s)Stereotactic ablative body radiotherapy (SBRT) for vertebral metastases is a challenging treatment process. Planning automation has recently reported the potential to improve plan quality and increase planning efficiency. We performed a dosimetric evaluation of the new Personalized engine implemented in Pinnacle3 for full planning automation of SBRT spine treatments in terms of plan quality, treatment efficiency, and delivery accuracy.Materials/MethodsThe Pinnacle3 treatment planning system was used to reoptimize six patients with spinal metastases, employing two separate automated engines. These two automated engines, the existing Autoplanning and the new Personalized, are both template-based algorithms that employ a wishlist to construct planning goals and an iterative technique to replicate the planning procedure performed by skilled planners. The boost tumor volume (BTV) was defined as the macroscopically visible lesion on RM examination, and the planning target volume (PTV) corresponds with the entire vertebra. Dose was prescribed according to simultaneous integrated boost strategy with BTV and PTV irradiated simultaneously over 3 fractions with a dose of 30 and 21 Gy, respectively. Dose-volume histogram (DVH) metrics and conformance indices were used to compare clinically accepted manual plans (MP) with automated plans developed using both Autoplanning (AP) and Personalized engines (Pers). All plans were evaluated for planning efficiency and dose delivery accuracy.ResultsFor similar spinal cord sparing, automated plans reported a significant improvement of target coverage and dose conformity. On average, Pers plans increased near-minimal dose D98% by 10.4% and 8.9% and target coverage D95% by 8.0% and by 4.6% for BTV and PTV, respectively. Automated plans provided significantly superior dose conformity and dose contrast by 37%–47% and by 4.6%–5.7% compared with manual plans. Overall planning times were dramatically reduced to about 15 and 23 min for Pers and AP plans, respectively. The average beam-on times were found to be within 3 min for all plans. Despite the increased complexity, all plans passed the 2%/2 mm γ-analysis for dose verification.ConclusionAutomated planning for spine SBRT through the new Pinnacle3 Personalized engine provided an overall increase of plan quality in terms of dose conformity and a major increase in efficiency. In this complex anatomical site, Personalized strongly reduce the tradeoff between optimal accurate dosimetry and planning time.

Site-agnostic 3D dose distribution prediction with deep learning neural networks.

Typically, the current dose prediction models are limited to small amounts of data and require retraining for a specific site, often leading to suboptimal performance. We propose a site-agnostic, three-dimensional dose distribution prediction model using deep learning that can leverage data from any treatment site, thus increasing the total data available to train the model. Applying our proposed model to a new target treatment site requires only a brief fine-tuning of the model to the new data and involves no modifications to the model input channels or its parameters. Thus, it can be efficiently adapted to a different treatment site, even with a small training dataset. This study uses two separate datasets/treatment sites: data from patients with prostate cancer treated with intensity-modulated radiation therapy (source data), and data from patients with head-and-neck cancer treated with volumetric-modulated arc therapy (target data). We first developed a source model with 3D UNet architecture, trained from random initial weights on the source data. We evaluated the performance of this model on the source data. We then studied the generalizability of the model to the new target dataset via transfer learning. To do this, we built three more models, all with the same 3D UNet architecture: target model, adapted model, and combined model. The source and target models were trained on the source and target data from random initial weights, respectively. The adapted model fine-tuned the source model to the target domain by using the target data. Finally, the combined model was trained from random initial weights on a combined data pool consisting of both target and source datasets. We tested all four models on the target dataset and evaluated quantitative dose-volume histogram metrics for the planning target volume (PTV) and organs at risk (OARs). When tested on the source treatment site, the source model accurately predicted the dose distributions with average (mean, max) absolute dose errors of (0.32%±0.14, 2.37%±0.93) (PTV) relative to the prescription dose, and highest mean dose error of 1.68%±0.76, and highest max dose error of 5.47%± 3.31 for femoral head right. The error in PTV dose coverage prediction is 3.21%±1.51 for D98 , 3.04%±1.69 for D95 , and 1.83%±1.01 for D02 . Averaging across all OARs, the source model predicted the OAR mean dose within 1.38% and the OAR max dose within 3.64%. For the target treatment site, the target model average (mean, max) absolute dose errors relative to the prescription dose for the PTV were (1.08%±0.95, 2.90%±1.35). Left cochlea had the highest mean and max dose errors of 5.37%±5.82 and 8.33%±8.88, respectively. The errors in PTV dose coverage prediction for D98 and D95 were 2.88%±1.59 and 2.55%±1.28, respectively. The target model can predict the OAR mean dose within 2.43% and the OAR max dose within 4.33% on average across all OARs. We developed a site-agnostic model for three-dimensional dose prediction and tested its adaptability to a new target treatment site via transfer learning. Our proposed model can make accurate predictions with limited training data.

Impact of white-matter mask selection on DTI histogram-based metrics as potential biomarkers in cerebral small vessel disease.

Histogram-based metrics extracted from diffusion-tensor imaging (DTI) have been suggested as potential biomarkers for cerebral small vessel disease (SVD), but methods and results have varied across studies. This work aims to assess the impact of mask selection for extracting histogram-based metrics of fractional anisotropy (FA) and mean diffusivity (MD) on their sensitivity as SVD biomarkers. DTI data were collected from 17 SVD patients and 12 healthy controls. FA and MD maps were estimated; from these, histograms were computed on two whole-brain white-matter masks: normal-appearing white-matter (NAWM) and mean FA tract skeleton (TBSS). Histogram-based metrics (median, peak height, peak width, peak value) were extracted from the FA and MD maps. These were compared between groups and correlated with the patients' cognitive scores (executive function and processing speed). White-matter mask selection significantly impacted FA and MD histogram metrics. In particular, significant interactions were found between Mask and Group for FA peak height (p = 0.027), MD Median (p = 0.035) and MD peak width (p = 0.047); indicating that the mask used affected their ability to discriminate between groups. In fact, MD peak width showed a significant 8.8% increase in patients when using TBSS (p = 0.037), but not when using NAWM (p = 0.69). Moreover, the mask may have an effect on the correlations with cognitive measures. Nevertheless, MD peak width (TBSS: r = - 0.75, NAWM: r = - 0.71) and MD peak height (TBSS: r = 0.65, NAWM: r = 0.62) remained significantly correlated with executive function, regardless of the mask. The impact of the processing methodology, in particular the choice of white-matter mask, highlights the need for standardized MRI data-processing pipelines.

Evaluation of microvascular invasion of hepatocellular carcinoma using whole-lesion histogram analysis with the stretched-exponential diffusion model.

To evaluate the potential role of histogram analysis of stretched exponential model (SEM) through whole-tumor volume for preoperative prediction of microvascular invasion (MVI) in single hepatocellular carcinoma (HCC). This study included 43 patients with pathologically proven HCCs by surgery who underwent multiple b-values diffusion-weighted imaging (DWI) and contrast-enhanced MRI. The histogram metrics of distributed diffusion coefficient (DDC) and heterogeneity index (α) from SEM were compared between HCCs with and without MVI, by using the independent t-test. Morphologic features of conventional MRI and clinical data were evaluated with chi-squared or Fisher's exact tests. Receiver operating characteristic (ROC) and multivariable logistic regression analyses were performed to evaluate the diagnostic performance of different parameters for predicting MVI. The tumor size and non-smooth tumor margin were significantly associated with MVI (all p < 0.05). The mean, fifth, 25th, 50th percentiles of DDC, and the fifth percentile of ADC between HCCs with and without MVI were statistically significant differences (all p < 0.05). The histogram parameters of α showed no statistically significant differences (all p > 0.05). At multivariate analysis,the fifth percentile of DDC was independent risk factor for MVI of HCC(p = 0.006). Histogram parameters DDC and ADC, but not the α value, are useful predictors of MVI. The fifth percentile of DDC was the most useful value to predict MVI of HCC. There is limited literature addressing the role of SEM for evaluating MVI of HCC. Our findings suggest that histogram analysis of SEM based on whole-tumor volume can be useful for MVI prediction.

Exploring Contrast Multi-Attribute Representation With Deep Network for No-Reference Perceptual Quality Assessment

Aiming at the effectiveness of contrast feature design, we proposed a promising novel non-reference quality assessment approach in exploring Attribute-Based representation. The method generates three perceptual attribute categories tailored to contrast. The first is semantic attribute derived from deep convolutional neural network, which implements adaptive contrast prediction relevant to scenario content. Second, for perceiving Spatial channel attribute, the global and local features generated by dark channel map through the designed dual convolution structures. Third, for statistical attribute, we assume the enhanced image as &#x201C;reference&#x201D; and calculate the structural similarity with pristine image, and the entropy and histogram metrics are also employed to assist learning. After that, for maximizing utilization, the features are embedded and integrated hierarchically to translate into objective score. In addition, a medium-scale contrast distortion database is established to support further research, which is more challenging than existing datasets because of the sufficient content and sophisticated changes. We demonstrate the availability of structures quantitatively and verify the rationality of hypothesis. Extensive experiments reveal that the proposed method outperforms advanced methods and achieves the state-of-the-art on the created database and CSIQ, TID2013, CCID2014.

3D Textured Model Encryption Using 2D Logistic and 3D Lorenz Chaotic Map

The widespread of recent multimedia, including various 3D model applications in different domains of areas, may lead to 3D models being stolen and attacked by hackers. Moreover, 3D models must be protected from unauthorized users and when transmitting over the internet. Nowadays the 3D model protection is a very important issue. This paper proposed a scheme that provides high protection for the textured 3D model by implementing multiple levels of security. The first level of security is achieved by encrypting the texture map based on a key generated by a 2D Logistic chaotic map. The second level of security is implemented by modifying the vertices values of the 3D mesh based on keys generated by the 3D Lorenz chaotic map. The proposed scheme was implemented on various 3D textured models varying in the number of vertices and faces. The experimental results show that the proposed scheme has a good encryption and provides high security by completely deforms the whole texture and 3D mesh of the textured 3D model into the two levels. The encryption scheme has a large key space 10135 making the scheme resists violent attacks. The Hausdorff Distance (HD) and histogram metrics are adopted to calculate the matching degree between the original and extracted model. The results show that the original and extracted model are identical through the values of HD, which are approximate to zero, and the histogram visually is similar. Index Terms— 3D textured model, encryption, 2D Logistic map, 3D Lorenz map, chaotic map

Comparison of different ROI analysis methods for liver lesion characterization with simplified intravoxel incoherent motion (IVIM)

This study investigated the impact of different ROI placement and analysis methods on the diagnostic performance of simplified IVIM-DWI for differentiating liver lesions. 1.5/3.0-T DWI data from a respiratory-gated MRI sequence (b = 0, 50, 250, 800 s/mm2) were analyzed in patients with malignant (n = 74/54) and benign (n = 35/19) lesions. Apparent diffusion coefficient ADC = ADC(0,800) and IVIM parameters D1′ = ADC(50,800), D2′ = ADC(250,800), f1′ = f(0,50,800), f2′ = f(0,250,800), and D*' = D*(0,50,250,800) were calculated voxel-wise. For each lesion, a representative 2D-ROI, a 3D-ROI whole lesion, and a 3D-ROI from “good” slices were placed, including and excluding centrally deviating areas (CDA) if present, and analyzed with various histogram metrics. The diagnostic performance of 2D- and 3D-ROIs was not significantly different; e.g. AUC (ADC/D1′/f1′) were 0.958/0.902/0.622 for 2D- and 0.942/0.892/0.712 for whole lesion 3D-ROIs excluding CDA at 1.5 T (p > 0.05). For 2D- and 3D-ROIs, AUC (ADC/D1′/D2′) were significantly higher, when CDA were excluded. With CDA included, AUC (ADC/D1′/D2′/f1′/D*') improved when low percentiles were used instead of averages, and was then comparable to the results of average ROI analysis excluding CDA. For lesion differentiation the use of a representative 2D-ROI is sufficient. CDA should be excluded from ROIs by hand or automatically using low percentiles of diffusion coefficients.

Learning-based synthetic dual energy CT imaging from single energy CT for stopping power ratio calculation in proton radiation therapy.

Dual energy CT (DECT) has been shown to estimate stopping power ratio (SPR) map with a higher accuracy than conventional single energy CT (SECT) by obtaining the energy dependence of photon interactions. This work presents a learning-based method to synthesize DECT images from SECT image for proton radiotherapy. The proposed method uses a residual attention generative adversarial network. Residual blocks with attention gates were used to force the model to focus on the difference between DECT images and SECT images. To evaluate the accuracy of the method, we retrospectively investigated 70 head-and-neck cancer patients whose DECT and SECT scans were acquired simultaneously. The model was trained to generate both a high and low energy DECT image based on a SECT image. The generated synthetic low and high DECT images were evaluated against the true DECT images using leave-one-out cross-validation. To evaluate our method in the context of a practical application, we generated SPR maps from synthetic DECT (sDECT) using a dual-energy based stoichiometric method and compared the SPR maps to those generated from DECT. A dosimetric comparison for dose obtained from DECT was performed against that derived from sDECT. The mean of mean absolute error, peak signal-to-noise ratio and normalized cross-correlation for the synthetic high and low energy CT images was 36.9 HU, 29.3 dB, 0.96 and 35.8 HU, 29.2 dB, and 0.96, respectively. The corresponding SPR maps generated from synthetic DECT showed an average normalized mean square deviation of about 1% with reduced noise level and artifacts than those from original DECT. Dose-volume histogram (DVH) metrics for the clinical target volume agree within 1% between the DECT and sDECT calculated dose. Our method synthesized accurate DECT images and showed a potential feasibility for proton SPR map generation. This study investigated a learning-based method to synthesize DECT images from SECT image for proton radiotherapy.

Pre-Treatment Apparent Diffusion Coefficient Histogram Metrics as a Predictor of Local Tumor Control After Proton Beam Therapy in Patients With Hepatocellular Carcinomas

Apparent diffusion coefficient (ADC) values extractable by diffusion-weighted imaging (DWI) are reported as useful in predicting the outcome of surgery and radiofrequency ablation (RFA) in hepatocellular carcinoma (HCC). However, their relationship with local tumor control (LC) after proton beam therapy (PBT) has not been fully evaluated. This study aimed to investigate interobserver variability in ADC histogram metrics and their usefulness in predicting LC in HCC patients who underwent PBT.The ADC maps of the liver of HCC patients who underwent PBT between 2015 and 2020 were reviewed. The inclusion criteria were imaging DWI at 1.5T within 6 weeks before PBT and the largest tumor dimension between 2 and 10 cm. Twenty-nine patients and 32 tumors were eligible. Axial DWI was acquired using a single-shot spin-echo echo-planar sequence and the b values of 0 and 1000 s mm-2. The ADC maps generated were transferred to a workstation. Five board-certified radiation oncologists (ROs) contoured the whole tumor regions-of-interest (ROIs) on these maps, using anatomical images obtained together as reference. For each ROI, the minimum (ADCmin), mean (ADCmean), maximum (ADCmax), and the 5th (ADC5), 25th (ADC25), 50th (ADC50), 75th (ADC75), and 95th (ADC95) percentiles were extracted. The intraclass correlation coefficient (2,1) (ICC) was calculated to determine interobserver reproducibility: ICC of 0.75 - 0.90 was considered good and > 0.90 excellent. The ADC metrics were averaged among 5 ROs. The optimal cut-off values that determined LC, i.e., non-progressive disease (PD) based on the modified RECIST, were identified using receiver-operating characteristics (ROC) analysis. The LC rates were calculated using the Kaplan-Meier method, and statistical significance was assessed by log-rank test.The median calculated biologically effective dose (α / β = 10) was 97 GyE. With a median follow-up of 39 months (5 - 62), 5 tumors had local PD. The ADCmin, ADCmean, ADCmax, ADC5, ADC25, ADC50, ADC75, and ADC95 of all tumors ranged between 0 - 1131, 748 - 1726, 1566 - 3149, 396 - 1430, 600 - 1566, 723 - 1719, 858 - 1874, and 1166 - 2426 (×10-6 mm 2 s-1), respectively. The ICC were 0.761, 0.969, 0.513, 0.864, 0.978, 0.983, 0.974, and 0.859, respectively. For ADCmean, ADC25, ADC50, and ADC75 which showed excellent reproducibility, the optimal cut-off values to determine LC were 1025, 1005, 1156, and 1132 (×10-6 mm 2 s-1), respectively. The tumors with higher ADC metrics had higher 2-year LC rates than their counterparts (100 % vs 40 %, P = 0.002; 100% vs 74%, P = 0.011; 100% vs 76%, P = 0.024; 100% vs 25%, P < 0.001, respectively).ADCmean, ADC25, ADC50, and ADC75 showed excellent interobserver reproducibility. Pre-treatment ADC histogram metrics may be useful in predicting LC after PBT in HCC.

A Comparative Study Between AI-Generated, Real-Life Clinical as Well as Reference Rectal Volumes Defined in Accordance With the Swedish National STRONG Guidelines in Prostate Cancer Radiotherapy

Successful delivery of radiotherapy (RT) relies on correct and consistent definition of organs at risk (OAR). Artificial intelligence (AI) auto-segmentation based on deep learning (DL) is emerging as a means to reduce variability of OARs. Knowledge about how auto-segmentation tools perform on standardized datasets is limited. Our aim is to evaluate the quality of rectal volumes defined by an existing DL auto-contouring algorithm on a standardized dataset for rectal volumes in prostate cancer RT as proposed by the Swedish STRONG guidelines for male pelvis [1][2][3]. [1] Olsson et al. PhIRO 2019:11:88 [2] Gay et al. IJROBP 2012:83:e353 [3] Salembier et al. RO 2018:127:49 MATERIALS/METHODS: DICOM-RT datasets for 19 patients extracted from a treatment planning system at one institution in Sweden was used. The patients were treated for prostate cancer in 2018-2019 using 15 MV photon beams to total doses of 50-70Gy@2-3Gy/fraction. Rectal volumes were defined on planning CT images clinically (clinical), according to proposed standard (reference), and auto-contoured using the existing version of the MVision software. The degree of variation between volumes was investigated. Statistics included dose volume histogram metrics, the Jaccard index, and the Dice similarity coefficient with comparisons between groups based on hypothesis testing with a two-sided P-value≤0.05 indicating a statistically significant difference.The clinical and the MVision DVH points were in general closer to each other than to the points of the reference DVH. Clinical and MVision volumes were also somewhat larger than the reference volume (83cc and 84cc versus 80cc; P > > 0.05). Overall, mean doses were 0.8/0.9Gy lower for clinical/MVision compared with the reference (21.9 ± 5.1/21.8 ± 5.1Gy versus 22.7 ± 5.9Gy; P > > 0.05) whilst maximum doses were similar for all (67.8-67.9 ± 4.3). Both overlap metrics were similar for volume comparisons between MVision and reference and for MVision and clinical but somewhat higher for clinical and reference (Jaccard index/Dice coefficient: 0.74/0.85 versus 0.73/0.83 versus 0.90/0.95, respectively). Differences were primarily found in cranio-caudal direction between any two data sets.This is the first systematic comparison of an auto-contouring algorithm's performance on a standardized OAR dataset in RT. Our results need to be confirmed on a larger patient cohort, but indicate that the MVision algorithm did very well since the size of identified differences generally was small both with respect to clinical and to reference volumes. However, to increase the use of AI-algorithms for clinical studies and research, retraining them on standardized reference datasets has the potential to improve performance further.

Learning-Based Prostate Needle Position Prediction for HDR Brachytherapy

During prostate high-dose-rate (HDR) brachytherapy, physicians select needle positions in a peripheral loading technique based on experience but without knowledge of the achievable dose distribution. Sub-optimal needle placement may lead to unfavorable dose distributions even after plan optimization. We propose a new deep learning-based method to predict HDR needle position for prostate HDR brachytherapy.The proposed framework consists of three major steps: (1) deformable registration via registration network (Reg-Net), (2) multi-atlas ranking and (3) needle regression. To model the global spatial relationship among multiple organs, binary masks of the target and organs at risk are transformed into distance maps which describe the distance of each local voxel to the organ surfaces. Reg-Net is utilized to deformably register the distance maps from multi-atlas set to match a patient's distance map and then bring needle maps from multi-atlas to this patient via spatial transformation. Several criteria are used for multi-atlas ranking including prostate volume similarity, multi-organ semantic image similarity and needle position criteria. Finally, needle regression is used to refine the final needle positions. A retrospective study of consecutively treated patients was used to evaluate the feasibility of the proposed method. We performed a five-fold cross validation to evaluate the proposed method. We generated a new dose plan for each patient based on predicted needle location. Target and organ dose volume histogram (DVH) metrics were used to quantify the difference between the clinical and predicted needle plans. All plans were normalized to prostate V100 of 95%.Ninety predicted needle plans were compared to clinical plans. In each case, the needle prediction algorithm completed within 1 minute. See table for comparison of target and organ DVH metrics (mean and standard deviation). Predicted needle plans had slightly greater target dose heterogeneity. The RTOG constraints of bladder V75% < 1 cc, rectum V75% < 1cc, and urethra V125% < 1 cc were met in all predicted plans.It is feasible to use this novel deep-learning-based method to predict needle locations for HDR prostate brachytherapy a priori. This strategy merits further study to improve utilization and quality of prostate brachytherapy.

Predicting T and N Staging of Resectable Gastric Cancer According to Whole Tumor Histogram Analysis About a Non-Cartesian k-Space Acquisition DCE-MRI: A Feasibility Study.

ObjectiveTo explore the feasibility of the whole tumor histogram analysis parameters derived from dynamic contrast-enhanced MRI (DCE-MRI) based on stack-of stars (StarVIBE) to predict T and N staging of resectable gastric cancer (GC).MethodsEighty-seven patients confirmed as GC by histopathology were enrolled in this prospective study. DCE-MRI were performed before surgery, and quantitative DCE parameters (Ktrans, Kep, Ve) and histogram metrics (Skewness, Kurtosis and Entropy) were measured by Omni-Kinetics software. Intraclass correlation coefficient (ICC) testing was used to determine the consistency of Ktrans, Kep and Ve values and histogram metrics values between two radiologists using Bland–Altman analysis. The quantitative DCE parameters or histogram metrics values between T stage or N stage were compared using ANOVA or Kruskal–Wallis testing. Receiver operating characteristic (ROC) analyses was performed to find out the best parameters for identifying T and N staging.ResultsThere was statistical difference in Ktrans, Kep, Ve and entropy to identify T staging (P=0.015, 0.033, <0.001, and 0.007, respectively), and in pairwise comparisons of Ve values showed statistically difference between T1+2 and T3 group (P<0.001), T1+2 and T4 group (P<0.001). There were statistical differences in Ve to identify N staging (P=0.041). In ROC analysis, Ve was the best parameter for identifying T staging (AUC: 0.788, the sensitivity and specificity was 0.929 and 0.578, respectively) and N staging (AUC: 0.590, the sensitivity and specificity was 0.714 and 0.899, respectively).ConclusionThe whole tumor histogram analysis parameters derived from StarVIBE DCE-MRI may be able to quantitatively evaluate T and N staging of GC, so as to help clinical treatment decision optimization.

Diffusion kurtosis imaging–derived histogram metrics for prediction of resistance to neoadjuvant chemoradiotherapy in rectal adenocarcinoma: Preliminary findings

PurposeThis study aimed to evaluate the potential role of diffusion kurtosis imaging (DKI)-derived parameters for assessing resistance to CRT in patients with Locally advanced rectal cancer (LARC) by using histogram analysis derived from whole-tumor volumes. Method136 consecutive patients with histologically confirmed rectal adenocarcinoma who underwent MRI examination before and after chemoradiotherapy were enrolled in our retrospective study. The parameters D, K, and conventional apparent diffusion coefficient (ADC) were measured using whole-tumor volume histogram analysis. The AJCC tumor regression grading (TRG) system was the standard reference (resistance: TRG 3; non-resistance: TRG 0–2). Receiver operating characteristic (ROC) curves were used for evaluating the diagnostic performance. ResultsAside from the skew and kurtosis values, we found all the histogram metrics of D and ADC values significantly increased after CRT (all p < 0.001). In contrast, the histogram metrics of K values significantly decreased after CRT. The majority of percentiles metrics of D, K, and ADC values were correlated with tumor resistance before and after CRT (P < 0.05), except for the skew and kurtosis values. Regarding the comparison of the diagnostic performance of all the histogram metrics, the percentage Dmean change (ΔDmean) showed the highest AUC value of 0.939, and the corresponding sensitivity, specificity, PPV, and NPV were 84.1% and 94.6%, 88.1% and 92.6%, respectively. ConclusionsThese preliminary results demonstrated that DKI-derived histogram metrics, especially the pre-treatment metrics and ΔDmean, were useful to assess tumoral resistance to CRT and individual clinical management for patients with LARC.

IMAGING: EP.338 Fat fraction histogram metrics assessed by quantitative MRI demonstrate differences between neuromuscular diseases

IMAGING: EP.338 Fat fraction histogram metrics assessed by quantitative MRI demonstrate differences between neuromuscular diseases