Search

Choice experiments and experimental auctions have become popular mechanisms for estimating willingness to pay (WTP). However, these methods have primarily been used for estimating WTP for single units of goods. We analyze the results from experimental auctions and choice experiments in the context of multiple quantities of a quasi-public good (animal welfare product). We show that the use of WTP values for a single unit of a product, a common practice in experimental valuation literature, can result in underestimation of aggregate demand. We use and compare open ended choice experiments (OECE), second price Vickrey auctions, and random Nth price auctions as mechanisms for valuing WTP. Our results also suggest that individual level demand estimates from OECE are less elastic than demand estimates from uniform price auctions. [ABSTRACT FROM AUTHOR]

We explore the application of genetic algorithms (GA) to deformable models through the proposition of a novel method for medical image segmentation that combines GA with nonconvex, localized, medial-based shape statistics. We replace the more typical gradient descent optimizer used in deformable models with GA, and the convex, implicit, global shape statistics with nonconvex, explicit, localized ones. Specifically, we propose GA to reduce typical deformable model weaknesses pertaining to model initialization, pose estimation and local minima, through the simultaneous evolution of a large number of models. Furthermore, we constrain the evolution, and thus reduce the size of the search-space, by using statistically-based deformable models whose deformations are intuitive (stretch, bulge, bend) and are driven in terms of localized principal modes of variation, instead of modes of variation across the entire shape that often fail to capture localized shape changes. Although GA are not guaranteed to achieve the global optima, our method compares favorably to the prevalent optimization techniques, convex/nonconvex gradient-based optimizers and to globally optimal graph-theoretic combinatorial optimization techniques, when applied to the task of corpus callosum segmentation in 50 mid-sagittal brain magnetic resonance images. [ABSTRACT FROM PUBLISHER]

A method for visualizing manifold-valued medical image data is proposed. The method operates on images in which each pixel is assumed to be sampled from an underlying manifold. For example, each pixel may contain a high dimensional vector, such as the time activity curve (TAC) in a dynamic positron emission tomography (dPET) or a dynamic single photon emission computed tomography (dSPECT) image, or the positive semi-definite tensor in a diffusion tensor magnetic resonance image (DTMRI). A nonlinear mapping reduces the dimensionality of the pixel data to achieve two goals: distance preservation and embedding into a perceptual color space. We use multidimensional scaling distance-preserving mapping to render similar pixels (e.g., DT or TAC pixels) with perceptually similar colors. The 3D CIELAB perceptual color space is adopted as the range of the distance preserving mapping, with a final similarity transform mapping colors to a maximum gamut size. Similarity between pixels is either determined analytically as geodesics on the manifold of pixels or is approximated using manifold learning techniques. In particular, dissimilarity between DTMRI pixels is evaluated via a Log-Euclidean Riemannian metric respecting the manifold of the rank 3, second-order positive semi-definite DTs, whereas the dissimilarity between TACs is approximated via ISOMAP. We demonstrate our approach via artificial high-dimensional, manifold-valued data, as well as case studies of normal and pathological clinical brain and heart DTMRI, dPET, and dSPECT images. Our results demonstrate the effectiveness of our approach in capturing, in a perceptually meaningful way, important features in the data. [ABSTRACT FROM AUTHOR]

The capacity for machine learning (ML) to facilitate radiation therapy (RT) planning for primary brain tumors has not been described. We evaluated ML-assisted RT planning with regard to clinical acceptability, dosimetric outcomes, and planning efficiency for adults and children with primary brain tumors. In this prospective study, children and adults receiving 54 Gy fractionated RT for a primary brain tumor were enrolled. For each patient, one ML-assisted RT plan was created and compared with 1 or 2 plans created using standard ("manual") planning procedures. Plans were evaluated by the treating oncologist, who was blinded to the method of plan creation. The primary endpoint was the proportion of ML plans that were clinically acceptable for treatment. Secondary endpoints included the frequency with which ML plans were selected as preferable for treatment, and dosimetric differences between ML and manual plans. A total of 116 manual plans and 61 ML plans were evaluated across 61 patients. Ninety-four percent of ML plans and 93% of manual plans were judged to be clinically acceptable (P = 1.0). Overall, the quality of ML plans was similar to manual plans. ML plans comprised 34.5% of all plans evaluated and were selected for treatment in 36.1% of cases (P =.82). Similar tumor target coverage was achieved between both planning methods. Normal brain (brain minus planning target volume) received an average of 1 Gy less mean dose with ML plans (compared with manual plans, P <.001). ML plans required an average of 45.8 minutes less time to create, compared with manual plans (P <.001). ML-assisted automated planning creates high-quality plans for patients with brain tumors, including children. Plans created with ML assistance delivered slightly less dose to normal brain tissues and can be designed in less time. [ABSTRACT FROM AUTHOR]

Purpose: High-quality radiotherapy (RT) planning for children and young adults with primary brain tumours is essential to minimize the risk of late treatment effects. The feasibility of using automated machine-learning (ML) to aid RT planning in this population has not previously been studied.Methods and Materials: We developed a ML model that identifies learned relationships between image features and expected dose in a training set of 95 patients with a primary brain tumour treated with focal radiotherapy to a dose of 54 Gy in 30 fractions. This ML method was then used to create predicted dose distributions for 15 previously-treated brain tumour patients across two institutions, as a testing set. Dosimetry to target volumes and organs-at-risk (OARs) were compared between the clinically-delivered (human-generated) plans versus the ML plans.Results: The ML method was able to create deliverable plans in all 15 patients in the testing set. All ML plans were generated within 30 min of initiating planning. Planning target volume coverage with 95% of the prescription dose was attained in all plans. OAR doses were similar across most structures evaluated; mean doses to brain and left temporal lobe were lower in ML plans than manual plans (mean difference to left temporal, - 2.3 Gy, p = 0.006; mean differences to brain, - 1.3 Gy, p = 0.017), whereas mean doses to right cochlea and lenses were higher in ML plans (+ 1.6-2.2 Gy, p < 0.05 for each).Conclusions: Use of an automated ML method to aid RT planning for children and young adults with primary brain tumours is dosimetrically feasible and can be successfully used to create high-quality 54 Gy RT plans. Further evaluation after clinical implementation is planned. [ABSTRACT FROM AUTHOR]

Extranodal extension (ENE) is strongly associated with worse outcomes in patients with head and neck squamous cell carcinoma (HNSCC) and is included in staging systems for HNSCC2,3. While histopathological determination remains the gold standard, pathologic confirmation is not always available for every patient or may not be available at the time when evidence of ENE may alter the treatment plan. This study aims to develop an automated method of pre-treatment radiographic imaging ENE (iENE) detection without human intervention to increase the chances of clinical adoption and improve risk stratification of patients with ENE. In a previous study by Kann et al, a deep learning model was used to predict the presence of iENE in pre-treatment computed tomography (CT) scans of lymph nodes (LNs) cropped along their borders1. The model trained on 2673 LNs from 270 patients achieved a notable AUC score of 0.91 on the held-out test set, and 0.84 when tested on multi-institutional data. This performance gap may be attributed to the strong correlation between ENE status and LN diameter, as revealed by their regression analysis of clinical variables. Additionally, deploying their model in clinical settings necessitates substantial human involvement, as it requires contouring single or multiple suspicious LNs for inference. To address these issues, we introduce a deep learning model designed to predict patient-level ENE without LN segmentations. Not relying on manual segmentation or cropping along the LNs borders improves potential adoption and serves as a regularization technique to reduce feature correlation with LN volume. We utilized a retrospective dataset comprising 922 oropharyngeal cancer patients and their radiotherapy (RT) treatment planning CT scans, each annotated by a radiologist for the presence of iENE. Our model analyzes 256x256x128 voxels centered around the larynx, encompassing LNs across all neck levels. Due to the input's considerable size, conventional 3D convolutional neural networks proved inefficient due to GPU memory limitations. To circumvent this, we leveraged ACS convolutions (ACSConv), which adapt 2D convolutions for 3D volumetric images, allowing us to capitalize on ImageNet pre-trained weights for more robust feature learning and quicker convergence. Our dataset was divided into 759 cases for training and 163 for held-out testing, with the date of diagnosis as the separation point, simulating a pseudo-prospective in-silico trial. The model was trained across four cross-validation folds on the training set, and the results from these four models were ensembled during inference. Our ACSConv models, without pre-training, achieved an area under the receiver operating characteristic curve (AUROC) of 0.80-0.83 and an area under the precision-recall curve (AUPRC) of 0.47-0.61 on the held-out test set, whereas our ImageNet pre-trained ACSConv model improved with an AUROC of 0.87-0.91 and an AUPRC of 0.52-0.69. This highlights the benefit of using ImageNet pre-trained weights, which encapsulate robust features learned from millions of standard images. However, it's crucial to recognize that these performance metrics solely gauge the model's alignment with iENE, not histopathology. To demonstrate the model's risk stratification capabilities, we examined overall survival using Kaplan-Meier curves. The radiologist's iENE determination did not yield a statistically significant difference (p=0.01), whereas the model's predictions resulted in significant differences (p<0.005). The increased prognostic power may be attributed to inherent limitations of radiological determination compared to histopathology. Alternatively, the model may be less sensitive and highlight severe cases, increasing its prognostic power. External validation on unseen cohorts will further elucidate the model's prognostic capabilities and limitations. [Display omitted] Additionally, we assessed the model's predictions against the volume of the largest LN gross tumor volume (GTVn) and found a moderate Spearman's correlation coefficient of 0.531. This suggests that while the model does consider GTVn volume, it certainly incorporates information beyond LN borders, potentially shedding light on issues with GTVn determination. [Display omitted] To account for potential variability in larynx segmentation models, we explored the impact of test-time augmentations on model inference. This simulated variations in larynx segmentation and its effect on iENE model performance. Notably, we observed no significant differences in prediction standard deviations between patients with or without iENE. An interesting trend emerged, showing that prediction standard deviations increased as the model's confidence decreased (closer to 0.5). This underscores the necessity for a deeper investigation into deep learning model behaviors to enhance safety and increase the likelihood of clinical adoption. [Display omitted] Our study presents a novel deep learning model for the automated detection of pre-treatment iENE. With improved performance and prognostic capabilities, this model holds potential for enhancing risk stratification in HNSCC patients and merits further validation on external datasets and in clinical practice. [ABSTRACT FROM AUTHOR]

With the purported future of oncological care being precision medicine, the hunt for predictive biomarkers has become a focal point. A potential source lies in radiological imaging, which has motivated the field of radiomics for the last decade [1–4]. Radiomics research, however, has been hampered by inconsistent methodology, despite efforts to establish standard features [5]. The release of the open-source PyRadiomics toolkit [6] was a significant and necessary step to standardize radiomics analysis, but the collation and distribution of publically available radiomics datasets remains poorly organized within the community. As a result, significant overhead remains when dealing with multiple training, testing, and validation datasets from both internal and external sources. Further, a recent study has raised the question of whether radiomic features with high predictive value are surrogates for tumour volume measurements [7]. There is a need for standard methodology for radiomic feature extraction, as well as large, publicly available radiomic datasets that have undergone rigorous processing to benchmark analyses. In this study, we have developed a reproducible, automated, open-source processing pipeline to generate analysis-ready radiomics data. We showcase the pipeline's capabilities by processing and analyzing the largest publicly available head and neck cancer (HNC) dataset, RADCURE [8], and compare three previously published radiomics models [1,7,9] using the resulting data. Data outputs have been made available via https://www.orcestra.ca/, a web-app that hosts processed 'omics data. Our proposed pipeline leverages three main tools: Med-ImageTools [10], PyRadiomics [6], and ORCESTRA [11]. While the former two are imaging-specific, we have modified ORCESTRA to work with clinical radiological data. The proposed pipeline was developed using the RADCURE [12] dataset. It consists of 3,346 HNC CT image volumes, corresponding radiotherapy structure sets (RTSTRUCT) containing primary gross tumour volume (GTVp) contours, and clinical data. The Med-ImageTools library was used to generate complete file lists for each CT acquisition, associate these with the correct RTSTRUCT, and load both as Simple ITK [13] images. For each GTVp, preprocessing, quality checking, and radiomic feature extraction was performed using PyRadiomics. Extraction settings from the RADCURE prognostic modelling challenge [8] were applied. Feature extraction was repeated with two negative control samples for each CT, either by shuffling voxel index values or randomly generating voxel values within the range of values in the original CT [7] (Figure 1). [Display omitted] The standard for data organization on ORCESTRA [11] is the MultiAssayExperiment R object [14], designed to harmonize multiple experimental assays from an overlapping patient set. To leverage this for radiomics, each set of extracted features becomes an experiment, with clinical data included as the primary metadata describing each patient. To demonstrate the pipeline's utility, we replicated previously published survival analysis models with the training and test cohorts from the RADCURE challenge subset [8]. Coefficients from the MW2018 [7] and Kwan [9] models were used to calculate prognostic index values for the test cohort. For comparison, we fit a Cox model to the RADCURE training cohort using the same radiomic signature and applied it to the test cohort. A univariate model for GTVp Mesh Volume was also tested. All models were compared using the concordance index. We processed 2,949 patients with GTVp contours, for a total of 2,988 GTVps from patients with varying primary tumour sites. We extracted 1,317 radiomic features from the CT and the negative control volume for each GTVp. For the 37 patients with multiple GTVps, features were extracted independently for each contour. The final data object containing all of these features, along with the clinical data and PyRadiomics configuration file, are available at https://www.orcestra.ca/radiomicset/10.5281/zenodo.8332910. The pipeline implementation is published at https://github.com/BHKLAB-DataProcessing/RADCUREradiomics. Results from our radiomics analysis are available in Table 1. The subset of 2400 GTVs was split into training and test cohorts based on the 'RADCURE-challenge' label in the clinical data. The Kwan model was tested with the oropharynx patients only. Model performance is similar whether the features were extracted from the CT or negative control samples, signaling that the radiomic signature is likely highly correlated with tumour volume, a known confounder of radiomics analysis. This is confirmed by the comparable performance of the univariable volume model. [Display omitted] This standardized architecture framework and the publicly available processed RADCURE dataset can be used to benchmark new datasets or radiomics models semi-automatically. Future work will include organ at risk and nodal targets in the RADCURE dataset and the production of ORCESTRA objects for other publicly available HNC datasets. We anticipate that this pipeline and the RADCURE objects generated could be a standard testing benchmark for future radiomics analyses and publications. [ABSTRACT FROM AUTHOR]