Mutual-information based optimal experimental design for hyperpolarized [Formula: see text]C-pyruvate MRI.

A key parameter of interest recovered from hyperpolarized (HP) MRI measurements is the apparent pyruvate-to-lactate exchange rate, [Formula: see text], for measuring tumor metabolism. This manuscript presents an information-theory-based optimal experimental design approach that minimizes the uncertainty in the rate parameter, [Formula: see text], recovered from HP-MRI measurements. Mutual information is employed to measure the information content of the HP measurements with respect to the first-order exchange kinetics of the pyruvate conversion to lactate. Flip angles of the pulse sequence acquisition are optimized with respect to the mutual information. A time-varying flip angle scheme leads to a higher parameter optimization that can further improve the quantitative value of mutual information over a constant flip angle scheme. However, the constant flip angle scheme, 35 and 28 degrees for pyruvate and lactate measurements, leads to an accuracy and precision comparable to the variable flip angle schemes obtained from our method. Combining the comparable performance and practical implementation, optimized pyruvate and lactate flip angles of 35 and 28 degrees, respectively, are recommended.

Quantitative dual-energy computed tomography with cesium as a novel contrast agent for localization of thermochemical ablation in phantoms and ex vivo models.

BACKGROUND: Thermochemical ablation (TCA) is a minimally invasive therapy under development for hepatocellular carcinoma. TCA simultaneously delivers an acid (acetic acid, AcOH) and base (sodium hydroxide, NaOH) directly into the tumor, where the acid/base chemical reaction produces an exotherm that induces local ablation. However, AcOH and NaOH are not radiopaque, making monitoring TCA delivery difficult. PURPOSE: We address the issue of image guidance for TCA by utilizing cesium hydroxide (CsOH) as a novel theranostic component of TCA that is detectable and quantifiable with dual-energy CT (DECT). MATERIALS AND METHODS: To quantify the minimum concentration of CsOH that can be positively identified by DECT, the limit of detection (LOD) was established in an elliptical phantom (Multi-Energy CT Quality Assurance Phantom, Kyoto Kagaku, Kyoto, Japan) with two DECT technologies: a dual-source system (SOMATOM Force, Siemens Healthineers, Forchheim, Germany) and a split-filter, single-source system (SOMATOM Edge, Siemens Healthineers). The dual-energy ratio (DER) and LOD of CsOH were determined for each system. Cesium concentration quantification accuracy was evaluated in a gelatin phantom before quantitative mapping was performed in ex vivo models. RESULTS: On the dual-source system, the DER and LOD were 2.94 and 1.36-mM CsOH, respectively. For the split-filter system, the DER and LOD were 1.41- and 6.11-mM CsOH, respectively. The signal on cesium maps in phantoms tracked linearly with concentration (R2  = 0.99) on both systems with an RMSE of 2.56 and 6.72 on the dual-source and split-filter system, respectively. In ex vivo models, CsOH was detected following delivery of TCA at all concentrations. CONCLUSIONS: DECT can be used to detect and quantify the concentration of cesium in phantom and ex vivo tissue models. When incorporated in TCA, CsOH performs as a theranostic agent for quantitative DECT image-guidance.

A Simulation of the Effects of Diffusion on Hyperpolarized [1-13C]-Pyruvate Signal Evolution.

OBJECTIVE: Hyperpolarized [1-13C]-pyruvate magnetic resonance imaging is an emerging metabolic imaging method that offers unprecedented spatiotemporal resolution for monitoring tumor metabolism in vivo. To establish robust imaging biomarkers of metabolism, we must characterize phenomena that may modulate the apparent pyruvate-to-lactate conversion rate (kPL). Here, we investigate the potential effect of diffusion on pyruvate-to-lactate conversion, as failure to account for diffusion in pharmacokinetic analysis may obscure true intracellular chemical conversion rates. METHODS: Changes in hyperpolarized pyruvate and lactate signal were calculated using a finite-difference time domain simulation of a two-dimensional tissue model. Signal evolution curves with intracellular kPL values from 0.02 to 1.00 s-1 were analyzed using spatially invariant one-compartment and two-compartment pharmacokinetic models. A second spatially variant simulation incorporating compartmental instantaneous mixing was fit with the same one-compartment model. RESULTS: When fitting with the one-compartment model, apparent kPL underestimated intracellular kPL by approximately 50% at an intracellular kPL of 0.02 s-1. This underestimation increased for larger kPL values. However, fitting the instantaneous mixing curves showed that diffusion accounted for only a small part of this underestimation. Fitting with the two-compartment model yielded more accurate intracellular kPL values. SIGNIFICANCE: This work suggests diffusion is not a significant rate-limiting factor in pyruvate-to-lactate conversion given that our model assumptions hold true. In higher order models, diffusion effects may be accounted for by a term characterizing metabolite transport. Pharmacokinetic models used to analyze hyperpolarized pyruvate signal evolution should focus on carefully selecting the analytical model for fitting rather than accounting for diffusion effects.

Deep Learning-Based Dose Prediction for Automated, Individualized Quality Assurance of Head and Neck Radiation Therapy Plans.

PURPOSE: This study aimed to use deep learning-based dose prediction to assess head and neck (HN) plan quality and identify suboptimal plans. METHODS AND MATERIALS: A total of 245 volumetric modulated arc therapy HN plans were created using RapidPlan knowledge-based planning (KBP). A subset of 112 high-quality plans was selected under the supervision of an HN radiation oncologist. We trained a 3D Dense Dilated U-Net architecture to predict 3-dimensional dose distributions using 3-fold cross-validation on 90 plans. Model inputs included computed tomography images, target prescriptions, and contours for targets and organs at risk (OARs). The model's performance was assessed on the remaining 22 test plans. We then tested the application of the dose prediction model for automated review of plan quality. Dose distributions were predicted on 14 clinical plans. The predicted versus clinical OAR dose metrics were compared to flag OARs with suboptimal normal tissue sparing using a 2 Gy dose difference or 3% dose-volume threshold. OAR flags were compared with manual flags by 3 HN radiation oncologists. RESULTS: The predicted dose distributions were of comparable quality to the KBP plans. The differences between the predicted and KBP-planned D1%,D95%, and D99% across the targets were within -2.53% ± 1.34%, -0.42% ± 1.27%, and -0.12% ± 1.97%, respectively, and the OAR mean and maximum doses were within -0.33 ± 1.40 Gy and -0.96 ± 2.08 Gy, respectively. For the plan quality assessment study, radiation oncologists flagged 47 OARs for possible plan improvement. There was high interphysician variability; 83% of physician-flagged OARs were flagged by only one of 3 physicians. The comparative dose prediction model flagged 63 OARs, including 30 of 47 physician-flagged OARs. CONCLUSIONS: Deep learning can predict high-quality dose distributions, which can be used as comparative dose distributions for automated, individualized assessment of HN plan quality.

Artificial Intelligence in Diagnostic Radiology: Where Do We Stand, Challenges, and Opportunities.

ABSTRACT: Artificial intelligence (AI) is the most revolutionizing development in the health care industry in the current decade, with diagnostic imaging having the greatest share in such development. Machine learning and deep learning (DL) are subclasses of AI that show breakthrough performance in image analysis. They have become the state of the art in the field of image classification and recognition. Machine learning deals with the extraction of the important characteristic features from images, whereas DL uses neural networks to solve such problems with better performance. In this review, we discuss the current applications of machine learning and DL in the field of diagnostic radiology.Deep learning applications can be divided into medical imaging analysis and applications beyond analysis. In the field of medical imaging analysis, deep convolutional neural networks are used for image classification, lesion detection, and segmentation. Also used are recurrent neural networks when extracting information from electronic medical records and to augment the use of convolutional neural networks in the field of image classification. Generative adversarial networks have been explicitly used in generating high-resolution computed tomography and magnetic resonance images and to map computed tomography images from the corresponding magnetic resonance imaging. Beyond image analysis, DL can be used for quality control, workflow organization, and reporting.In this article, we review the most current AI models used in medical imaging research, providing a brief explanation of the various models described in the literature within the past 5 years. Emphasis is placed on the various DL models, as they are the most state-of-art in imaging analysis.

Machine learning-based texture analysis for differentiation of radiologically indeterminate small adrenal tumors on adrenal protocol CT scans.

GOAL: To evaluate the ability of radiomic feature extraction and a machine learning algorithm to differentiate between benign and malignant indeterminate adrenal lesions on contrast-enhanced computed tomography (CT) studies. BACKGROUND: Adrenal "incidentalomas" are adrenal lesions that are accidentally discovered during workup not related to the adrenal glands; they have an incidence as high as 5%. Small adrenal incidentalomas (< 4 cm) with high attenuation values on pre-contrast CT(> 10 HU) need further evaluation to calculate the absolute percentage of washout (APW). If the APW is < 60%, these lesions are considered non-adenomas and commonly classified as indeterminate adrenal lesions. Further workup for indeterminate lesions includes more complicated and expensive radiological studies or invasive procedures like biopsy or surgical resection. METHODS: We searched our institutional database for indeterminate adrenal lesions with the following characteristics: < 4 cm, pre-attenuation value > 10 HU, and APW < 60%. Exclusion criteria included pheochromocytoma and no histopathological examination. CT images were converted to Nifti format, and adrenal tumors were segmented using Amira software. Radiomic features from the adrenal mask were extracted using PyRadiomics software after removing redundant features (highly pairwise correlated features and low-variance features) using recursive feature extraction to select the final discriminative set of features. Lastly, the final features were used to build a binary classification model using a random forest algorithm, which was validated and tested using leave-one-out cross-validation, confusion matrix, and receiver operating characteristic curve. RESULTS: We found 40 indeterminate adrenal lesions (21 benign and 19 malignant). Feature extraction resulted in 3947 features, which reduced down to 62 features after removing redundancies. Recursive feature elimination resulted in the following top 4 discriminative features: gray-level size zone matrix-derived size zone non-uniformity from pre-contrast and delayed phases, gray-level dependency matrix-derived large dependence high gray-level emphasis from venous-phase, and gray-level co-occurrence matrix-derived cluster shade from delayed-phase. A binary classification model with leave-one-out cross-validation showed AUC = 0.85, sensitivity = 84.2%, and specificity = 71.4%. CONCLUSION: Machine learning and radiomic features extraction can differentiate between benign and malignant indeterminate adrenal tumors and can be used to direct further workup with high sensitivity and specificity.

The use of laser interstitial thermal therapy in the treatment of brain metastases: a literature review.

PURPOSE: The aim of this paper is to discuss the current evidence for Laser Interstitial Thermal Therapy (LITT) in the treatment of brain metastases, our current recommendations for patient selection and the future perspectives for this therapy. We have also touched upon the possible complications and role of systemic therapy coupled with LITT for the treatment of brain metastases. MATERIAL AND METHODS: Two authors carried out the literature search using two databases independently, including PubMed, and Web of Science. The review included prospective and retrospective studies using LITT to treat brain metastases. RESULTS: Twenty-two original articles were analyzed in this review, particularly clinical outcomes and complications. We have also provided our institutional experience in the use of LITT to treat brain metastases and addressed future perspectives for the use of this technology. CONCLUSIONS: The current literature supports LITT as a safe and effective therapy for patients with brain metastases that have failed SRS. Larger studies are still required to better evaluate the use of systemic therapy in concomitance with LITT. New images modalities may enable optimized treatment and outcomes.

Laser Interstitial Thermal Therapy in the treatment of brain metastases and radiation necrosis.

Stereotactic Radiosurgery has become the main treatment for patients with limited number of brain metastases (BM). Recently, with the increasing use of this modality, there is a growth in recurrence cases. Recurrence after radiation therapy can be divided in changes favoring either tumor recurrence or radiation necrosis (RN). Laser Interstitial Thermal Therapy (LITT) is minimally invasive treatment modality that has been used to treat primary and metastatic brain tumors. When associated with real-time thermometry using Magnetic Resonance Imaging, the extent of ablation can be controlled to provide maximum coverage and avoid eloquent areas. The objective of this study was to investigate the use of LITT in the treatment of BM. An extensive review of the relevant literature was conducted and the outcome results are discussed. There is an emphasis on safety and local control rate of patients treated with this modality. The findings of our study suggest that LITT is a viable safe technique to treat recurrent BM, especially in patients with deep-seated lesions where surgical resection is not an option.

Identification of Kernels in a Convolutional Neural Network: Connections Between Level Set Equation and Deep Learning for Image Segmentation.

Medical image segmentation remains a difficult, time-consuming task; currently, liver segmentation from abdominal CT scans is often done by hand, requiring too much time to construct patient-specific treatment models for hepatocellular carcinoma. Image segmentation techniques, such as level set methods and convolutional neural networks (CNN), rely on a series of convolutions and nonlinearities to construct image features: neural networks that use strictly mean-zero finite difference stencils as convolution kernels can be treated as upwind discretizations of differential equations. If this relationship can be made explicit, one gains the ability to analyze CNN using the language of numerical analysis, thereby providing a well-established framework for proving properties such as stability and approximation accuracy. We test this relationship by constructing a level set network, a type of CNN whose architecture describes the expansion of level sets; forward-propagation through a level set network is equivalent to solving the level set equation; the level set network achieves comparable segmentation accuracy to solving the level set equation, while not obtaining the accuracy of a common CNN architecture. We therefore analyze which convolution filters are present in a standard CNN, to see whether finite difference stencils are learned during training; we observe certain patterns that form at certain layers in the network, where the learned CNN kernels depart from known convolution kernels used to solve the level set equation.

Predictors of Local Control of Brain Metastasis Treated With Laser Interstitial Thermal Therapy.

BACKGROUND: Laser Interstitial Thermal Therapy (LITT) has been used to treat recurrent brain metastasis after stereotactic radiosurgery (SRS). Little is known about how best to assess the efficacy of treatment, specifically the ability of LITT to control local tumor progression post-SRS. OBJECTIVE: To evaluate the predictive factors associated with local recurrence after LITT. METHODS: Retrospective study with consecutive patients with brain metastases treated with LITT. Based on radiological aspects, lesions were divided into progressive disease after SRS (recurrence or radiation necrosis) and new lesions. Primary endpoint was time to local recurrence. RESULTS: A total of 61 consecutive patients with 82 lesions (5 newly diagnosed, 46 recurrence, and 31 radiation necrosis). Freedom from local recurrence at 6 mo was 69.6%, 59.4% at 12, and 54.7% at 18 and 24 mo. Incompletely ablated lesions had a shorter median time for local recurrence (P < .001). Larger lesions (>6 cc) had shorter time for local recurrence (P = .03). Dural-based lesions showed a shorter time to local recurrence (P = .01). Tumor recurrence/newly diagnosed had shorter time to local recurrence when compared to RN lesions (P = .01). Patients receiving systemic therapy after LITT had longer time to local recurrence (P = .01). In multivariate Cox-regression model, the HR for incomplete ablated lesions was 4.88 (P < .001), 3.12 (P = .03) for recurrent tumors, and 2.56 (P = .02) for patients not receiving systemic therapy after LITT. Complication rate was 26.2%. CONCLUSION: Incompletely ablated and recurrent tumoral lesions were associated with higher risk of treatment failure and were the major predicting factors for local recurrence. Systemic therapy after LITT was a protective factor regarding local recurrence.

Mechanism-Based Modeling of Tumor Growth and Treatment Response Constrained by Multiparametric Imaging Data.

Multiparametric imaging is a critical tool in the noninvasive study and assessment of cancer. Imaging methods have evolved over the past several decades to provide quantitative measures of tumor and healthy tissue characteristics related to, for example, cell number, blood volume fraction, blood flow, hypoxia, and metabolism. Mechanistic models of tumor growth also have matured to a point where the incorporation of patient-specific measures could provide clinically relevant predictions of tumor growth and response. In this review, we identify and discuss approaches that use multiparametric imaging data, including diffusion-weighted magnetic resonance imaging, dynamic contrast-enhanced magnetic resonance imaging, diffusion tensor imaging, contrast-enhanced computed tomography, [18F]fluorodeoxyglucose positron emission tomography, and [18F]fluoromisonidazole positron emission tomography to initialize and calibrate mechanistic models of tumor growth and response. We focus the discussion on brain and breast cancers; however, we also identify three emerging areas of application in kidney, pancreatic, and lung cancers. We conclude with a discussion of the future directions for incorporating multiparametric imaging data and mechanistic modeling into clinical decision making for patients with cancer.

Lung tumor segmentation methods: Impact on the uncertainty of radiomics features for non-small cell lung cancer.

PURPOSE: To evaluate the uncertainty of radiomics features from contrast-enhanced breath-hold helical CT scans of non-small cell lung cancer for both manual and semi-automatic segmentation due to intra-observer, inter-observer, and inter-software reliability. METHODS: Three radiation oncologists manually delineated lung tumors twice from 10 CT scans using two software tools (3D-Slicer and MIM Maestro). Additionally, three observers without formal clinical training were instructed to use two semi-automatic segmentation tools, Lesion Sizing Toolkit (LSTK) and GrowCut, to delineate the same tumor volumes. The accuracy of the semi-automatic contours was assessed by comparison with physician manual contours using Dice similarity coefficients and Hausdorff distances. Eighty-three radiomics features were calculated for each delineated tumor contour. Informative features were identified based on their dynamic range and correlation to other features. Feature reliability was then evaluated using intra-class correlation coefficients (ICC). Feature range was used to evaluate the uncertainty of the segmentation methods. RESULTS: From the initial set of 83 features, 40 radiomics features were found to be informative, and these 40 features were used in the subsequent analyses. For both intra-observer and inter-observer reliability, LSTK had higher reliability than GrowCut and the two manual segmentation tools. All observers achieved consistently high ICC values when using LSTK, but the ICC value varied greatly for each observer when using GrowCut and the manual segmentation tools. For inter-software reliability, features were not reproducible across the software tools for either manual or semi-automatic segmentation methods. Additionally, no feature category was found to be more reproducible than another feature category. Feature ranges of LSTK contours were smaller than those of manual contours for all features. CONCLUSION: Radiomics features extracted from LSTK contours were highly reliable across and among observers. With semi-automatic segmentation tools, observers without formal clinical training were comparable to physicians in evaluating tumor segmentation.

Development of an Electroporation and Nanoparticle-based Therapeutic Platform for Bone Metastases.

Purpose To assess for nanopore formation in bone marrow cells after irreversible electroporation (IRE) and to evaluate the antitumoral effect of IRE, used alone or in combination with doxorubicin (DOX)-loaded superparamagnetic iron oxide (SPIO) nanoparticles (SPIO-DOX), in a VX2 rabbit tibial tumor model. Materials and Methods All experiments were approved by the institutional animal care and use committee. Five porcine vertebral bodies in one pig underwent intervention (IRE electrode placement without ablation [n = 1], nanoparticle injection only [n = 1], and nanoparticle injection followed by IRE [n = 3]). The animal was euthanized and the vertebrae were harvested and evaluated with scanning electron microscopy. Twelve rabbit VX2 tibial tumors were treated, three with IRE, three with SPIO-DOX, and six with SPIO-DOX plus IRE; five rabbit VX2 tibial tumors were untreated (control group). Dynamic T2*-weighted 4.7-T magnetic resonance (MR) images were obtained 9 days after inoculation and 2 hours and 5 days after treatment. Antitumor effect was expressed as the tumor growth ratio at T2*-weighted MR imaging and percentage necrosis at histologic examination. Mixed-effects linear models were used to analyze the data. Results Scanning electron microscopy demonstrated nanopores in bone marrow cells only after IRE (P , .01). Average volume of total tumor before treatment (503.1 mm3 ± 204.6) was not significantly different from those after treatment (P = .7). SPIO-DOX was identified as a reduction in signal intensity within the tumor on T2*-weighted images for up to 5 days after treatment and was related to the presence of iron. Average tumor growth ratios were 103.0% ± 75.8 with control treatment, 154.3% ± 79.7 with SPIO-DOX, 77% ± 30.8 with IRE, and -38.5% ± 24.8 with a combination of SPIO-DOX and IRE (P = .02). The percentage residual viable tumor in bone was significantly less for combination therapy compared with control (P = .02), SPIO-DOX (P , .001), and IRE (P = .03) treatment. The percentage residual viable tumor in soft tissue was significantly less with IRE (P = .005) and SPIO-DOX plus IRE (P = .005) than with SPIO-DOX. Conclusion IRE can induce nanopore formation in bone marrow cells. Tibial VX2 tumors treated with a combination of SPIO-DOX and IRE demonstrate enhanced antitumor effect as compared with individual treatments alone. © RSNA, 2017 Online supplemental material is available for this article.

Irreversible Electroporation in the Epidural Space of the Porcine Spine: Effects on Adjacent Structures.

Purpose To determine the effects of irreversible electroporation (IRE) on the neural tissues after ablation in the epidural space of the porcine spine. Materials and Methods The institutional animal care and use committee approved this study. With the IRE electrode positioned in the right lateral recess of the spinal epidural space, 20 IRE ablations were performed with computed tomographic (CT) guidance by using different applied voltages in four animals that were euthanized immediately after magnetic resonance (MR) imaging of the spine, performed 6 hours after IRE (terminal group). Histopathologic characteristics of the neural tissues were assessed and used to select a voltage for a survival study. Sixteen CT-guided IRE ablations in the epidural space were performed by using 667 V in four animals that were survived for 7 days (survival group). Clinical characteristics, MR imaging findings (obtained 6 hours after IRE and before euthanasia), histopathologic characteristics, and simulated electric field strengths were assessed. A one-way analysis of variance was used to compare the simulated electric field strength to histologic findings. Results The mean distance between the IRE electrode and the spinal cord and nerve root was 1.71 mm ± 0.90 and 8.47 mm + 3.44, respectively. There was no clinical evidence of paraplegia after IRE ablation. MR imaging and histopathologic examination showed no neural tissue lesions within the spinal cord; however, five of 16 nerve roots (31.2%) demonstrated moderate wallerian degeneration in the survival group. The severity of histopathologic injury in the survival group was not significantly related to either the simulated electric field strength or the distance between the IRE electrode and the neural structure (P > .05). Conclusion Although the spinal cord appears resistant to the toxic effects of IRE, injury to the nerve roots may be a limiting factor for the use of IRE ablation in the epidural space. © RSNA, 2016 Online supplemental material is available for this article.