Mirada Medical (United Kingdom)

PurposeAtlas-based and deep-learning contouring (DLC) are methods for automatic segmentation of organs-at-risk (OARs). The European Particle Therapy Network (EPTN) published a consensus-based atlas for delineation of OARs in neuro-oncology. In this study, geometric and dosimetric evaluation of automatically-segmented neuro-oncological OARs was performed using CT- and MR-models following the EPTN-contouring atlas. MethodsImage and contouring data from 76 neuro-oncological patients were included. Two atlas-based models (CT-atlas and MR-atlas) and one DLC-model (MR-DLC) were created. Manual contours on registered CT-MR-images were used as ground-truth. Results were analyzed in terms of geometrical (volumetric Dice similarity coefficient (vDSC), surface DSC (sDSC), added path length (APL), and mean slice-wise Hausdorff distance (MSHD)) and dosimetrical accuracy. Distance-to-tumor analysis was performed to analyze to which extent the location of the OAR relative to planning target volume (PTV) has dosimetric impact, using Wilcoxon rank-sum tests. ResultsCT-atlas outperformed MR-atlas for 22/26 OARs. MR-DLC outperformed MR-atlas for all OARs. Highest median (95 %CI) vDSC and sDSC were found for the brainstem in MR-DLC: 0.92 (0.88–0.95) and 0.84 (0.77–0.89) respectively, as well as lowest MSHD: 0.27 (0.22–0.39)cm. Median dose differences (ΔD) were within ± 1 Gy for 24/26(92 %) OARs for all three models. Distance-to-tumor showed a significant correlation for ΔDmax,0.03cc-parameters when splitting the data in ≤ 4 cm and > 4 cm OAR-distance (p < 0.001). ConclusionMR-based DLC and CT-based atlas-contouring enable high-quality segmentation. It was shown that a combination of both CT- and MR-autocontouring models results in the best quality.

To investigate a deep learning approach that enables three-dimensional (3D) segmentation of an arbitrary structure of interest given a user provided two-dimensional (2D) contour for context. Such an approach could decrease delineation times and improve contouring consistency, particularly for anatomical structures for which no automatic segmentation tools exist. A series of deep learning segmentation models using a Recurrent Residual U-Net with attention gates was trained with a successively expanding training set. Contextual information was provided to the models, using a previously contoured slice as an input, in addition to the slice to be contoured. In total, 6 models were developed, and 19 different anatomical structures were used for training and testing. Each of the models was evaluated for all 19 structures, even if they were excluded from the training set, in order to assess the model's ability to segment unseen structures of interest. Each model's performance was evaluated using the Dice similarity coefficient (DSC), Hausdorff distance, and relative added path length (APL). The segmentation performance for seen and unseen structures improved when the training set was expanded by addition of structures previously excluded from the training set. A model trained exclusively on heart structures achieved a DSC of 0.33, HD of 44mm, and relative APL of 0.85 when segmenting the spleen, whereas a model trained on a diverse set of structures, but still excluding the spleen, achieved a DSC of 0.80, HD of 13mm, and relative APL of 0.35. Iterative prediction performed better compared to direct prediction when considering unseen structures. Training a contextual deep learning model on a diverse set of structures increases the segmentation performance for the structures in the training set, but importantly enables the model to generalize and make predictions even for unseen structures that were not represented in the training set. This shows that user-provided context can be incorporated into deep learning contouring to facilitate semi-automatic segmentation of CT images for any given structure. Such an approach can enable faster de-novo contouring in clinical practice.

Este es un libro introductorio al electromagnetismo con los siguientes contenidos: electrostática y magnetostática en el vacío, campos dependientes del tiempo en el vacío (formulación maxwelliana básica), aspectos básicos sobre circuitos eléctricos en regímenes estacionarios y transitorios, propiedades electromagnéticas de los medios materiales, ondas electromagnéticas, óptica geométrica y óptica física.

This paper describes a novel automatic system for Modic changes classification of vertebral endplates. Modic changes are classes of vertebral degenerations visible as intensity variations in magnetic resonance images (MRI). The system operates on T1 and T2 MRI. We introduce three main novelties: 1. a vertebrae alignment scheme via precise bounding boxes obtained through corner localisation, 2. vertebral endplate classification in three dimensions, and 3. Modic changes classification. The system was trained and validated using a large dataset of 785 patients, containing MRIs sourced from a wide range of acquisition protocols. The proposed system achieved 87.8 % classification accuracy on our dataset.

Relaxation time (T 10) estimation using variable flip angle sequences is a key step for pharmacokinetic (PK) analysis of tumours in DCE-MRI exams. In this study, the effects of motion within flip angle sequences on the T 10 and subsequent K trans and k ep estimation were examined. It was found that errors in T 10 estimation caused by motion had a significant impact on subsequent PK analysis. A new similarity metric, based on the T 10 regression error, for groupwise motion correction of variable flip angle sequences is proposed and compared against Groupwise Normalized Mutual Information (GNMI). In rigid registration experiments on simulated data, the new metric outperformed GNMI, showing an improvement alignment of over 14% in terms of average target registration error, which is also reflected by a lower T 10 estimation error. Finally, registration was applied to 46 clinical sequences to identify the average amount of motion found in this type of acquisition; this showed an estimated displacement of 0.98mm, which could lead to over 25% K trans estimation error if motion were not corrected.

A significant arising complication in the care of patients with diabetes is increased susceptibility to chronic wounds, including diabetic foot ulcers and pressure sores. This is driven by, e.g., neuropathy and peripheral arterial disease. It is well recognized that best practice in wound care requires wound assessment, including measurement, on presentation and regularly throughout the treatment program. Proper assessment is necessary to ensure that the most appropriate and cost-effective therapy is used at all times, with quantitative measurement necessary to track the efficacy of the chosen approach. A documented assessment can also assist patient-clinician dialog and discussion within the multidisciplinary team. Remote evaluation and assessment of the wound is also of increasing importance and practicality through the use of a telemedicine approach. There has been considerable progress in the space of imaging for wounds, including systems that include three-dimensional measurement and telemedicine features. This literature review examines the available options and reviews the clinical evidence for measurement accuracy, scope for remote assessment, and published user feedback on the systems.

Purpose: To systematically study the effects of computation details on DVH accuracy for the first time in a teletherapy planning context, measure the effect sizes of different parameters and make recommendations for quality assurance in multicentric studies. Methods: We defined two IMRT plans (maximum doses: 84/79 Gy) on a physical phantom (using a commercial TPS) and exported five dose volumes with different resolutions (1 * 1 * 0.83 mm^3, 2 * 2 * 2.5 mm^3, 3 * 3 * 2.5 mm^3, 4 * 4 * 5 mm^3, 5 * 5 * 5 mm^3) and contours for fifteen structures (volumes between 1 and 349 cm^3). We computed 7200 DVHs by independently varying dose bin number (2^7, 2^8, 2^9, 2^10), grid geometry (CT, dose volume), voxel counting scheme (completely inside, center inside, partial volume‐weighted), interpolation method (nearest neighbor, linear, cubic). For each structure, the DVH with optimal simulation / computation parameters was interpreted as ground truth, and compared against the suboptimal DVHs using the percentage of failing gammas (PFG) distance (doi:10.1088/0031‐9155/55/11/N04) [1 Gy dose tolerance, 2% volume tolerance]. The effect sizes of the various parameters were estimated via ANOVA. Results: DVH accuracy (as measured by PFG) depended significantly on dose resolution (p&lt;1e‐15), grid geometry (p&lt;1e‐15) and voxel counting scheme (p=9e‐4), with partial ŋ2 values of 0.408 / 0.104 / 0.002. Bin number and interpolation method were negligible. DVHs computed from a 4 mm grid had an almost three times higher average PFG than DVHs from a 2 mm grid (0.42 vs. 0.16). Conclusion: For applications where DVH comparability between clinical sites with different TPS is imperative (e.g. multicentric studies), the dose resolution should be strictly identical, and researchers should ensure that DVHs are consistently computed either in the dose or CT grid geometry. Differences in the other computation parameters can be tolerated. Dose volumes with a resolution of &gt;3 mm, whilst occasionally encountered in clinical routine, are unsuitable for accurate DVH computation. FK and TK are employees of Mirada Medical. However, the contents of this abstract constitute strictly basic research, so that there is no conflict of interest.

We propose a method for accurately localizing anatomical landmarks in 3D medical volumes based on dense matching of parts-based graphical models. Our novel approach replaces population mean models by jointly leveraging weighted combinations of labeled exemplars (both spatial and appearance) to obtain personalized models for the localization of arbitrary landmarks in upper body images. We compare the method to a baseline population-mean graphical model and atlas-based deformable registration optimized for CT-CT registration, by measuring the localization accuracy of 22 anatomical landmarks in clinical 3D CT volumes, using a database of 83 lung cancer patients. The average mean localization error across all landmarks is 2.35 voxels. Our proposed method outperforms deformable registration by 73%, 93% for the most improved landmark. Compared to the baseline population-mean graphical model, the average improvement of localization accuracy is 32%; 67% for the most improved landmark.