Biomarker estimation methods from medical images have traditionally followed a segment-and-measure strategy. Deep-learning regression networks have changed such a paradigm, enabling the direct estimation of biomarkers in databases where segmentation masks are not present. While such methods achieve high performance, they operate as a black-box. In this work, we present a novel deep learning network structure that, when trained with only the value of the biomarker, can perform biomarker regression and the generation of an accurate localization mask simultaneously, thus enabling a qualitative assessment of the image locus that relates to the quantitative result. We showcase the proposed method with three different network structures and compare their performance against direct regression networks in four different problems: pectoralis muscle area (PMA), subcutaneous fat area (SFA), liver mass area in single slice computed tomography (CT), and Agatston score estimated from non-contrast thoracic CT images (CAC). Our results show that the proposed method improves the performance with respect to direct biomarker regression methods (correlation coefficient of 0.978, 0.998, and 0.950 for the proposed method in comparison to 0.971, 0.982, and 0.936 for the reference regression methods on PMA, SFA and CAC respectively) while achieving good localization (DICE coefficients of 0.875, 0.914 for PMA and SFA respectively, p < 0.05 for all pairs). We observe the same improvement in regression results comparing the proposed method with those obtained by quantify the outputs using an U-Net segmentation network (0.989 and 0.951 respectively). We, therefore, conclude that it is possible to obtain simultaneously good biomarker regression and localization when training biomarker regression networks using only the biomarker value.

Motivation: The PAM50 classifier is used to assign patients to the highest correlated breast cancer subtype irrespectively of the obtained value. Nonetheless, all subtype correlations are required to build the risk of recurrence (ROR) score, currently used in therapeutic decisions. Present subtype uncertainty estimations are not accurate, seldom considered or require a population-based approach for this context. Results: Here we present a novel single-subject non-parametric uncertainty estimation based on PAM50's gene label permutations. Simulations results ( n = 5228) showed that only 61% subjects can be reliably 'Assigned' to the PAM50 subtype, whereas 33% should be 'Not Assigned' (NA), leaving the rest to tight 'Ambiguous' correlations between subtypes. The NA subjects exclusion from the analysis improved survival subtype curves discrimination yielding a higher proportion of low and high ROR values. Conversely, all NA subjects showed similar survival behaviour regardless of the original PAM50 assignment. We propose to incorporate our PAM50 uncertainty estimation to support therapeutic decisions. Availability and Implementation: Source code can be found in 'pbcmc' R package at Bioconductor. Contacts: [email protected] or [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.

We present a novel approach to fully automated delineation of tree structures in noisy 2D images and 3D image stacks. Unlike earlier methods that rely mostly on local evidence, our method builds a set of candidate trees over many different subsets of points likely to belong to the final one and then chooses the best one according to a global objective function. Since we are not systematically trying to span all nodes, our algorithm is able to eliminate noise while retaining the right tree structure. Manually annotated dendrite micrographs and retinal scans are used to evaluate the performance of our method, which is shown to be able to reject noise while retaining the tree structure.

State-of-the-art approaches for detecting filament-like structures in noisy images rely on filters optimized for signals of a particular shape, such as an ideal edge or ridge. While these approaches are optimal when the image conforms to these ideal shapes, their performance quickly degrades on many types of real data where the image deviates from the ideal model, and when noise processes violate a Gaussian assumption. In this paper, we show that by learning rotational features, we can outperform state-of-the-art filament detection techniques on many different kinds of imagery. More specifically, we demonstrate superior performance for the detection of blood vessel in retinal scans, neurons in brightfield microscopy imagery, and streets in satellite imagery.