Automated Gleason grading is an important preliminary step for quantitative histopathological feature extraction. Different from the traditional task of classifying small pre-selected homogeneous regions, semantic segmentation provides pixel-wise Gleason predictions across an entire slide. Deep learning-based segmentation models can automatically learn visual semantics from data, which alleviates the need for feature engineering. However, performance of deep learning models is limited by the scarcity of large-scale fully annotated datasets, which can be both expensive and time-consuming to create. One way to address this problem is to leverage external weakly labeled datasets to augment models trained on the limited data. In this paper, we developed an expectation maximization-based approach constrained by an approximated prior distribution in order to extract useful representations from a large number of weakly labeled images generated from low-magnification annotations. This method was utilized to improve the performance of a model trained on a limited fully annotated dataset. Our semi-supervised approach trained with 135 fully annotated and 1800 weakly annotated tiles achieved a mean Jaccard Index of 49.5% on an independent test set, which was 14% higher than the initial model trained only on the fully annotated dataset.
Prostate cancer is the most common and second most deadly form of cancer in men in the United States. The classification of prostate cancers based on Gleason grading using histological images is important in risk assessment and treatment planning for patients. Here, we demonstrate a new
Current clinical practice relies on clinical history to determine the time since stroke onset (TSS). Imaging-based determination of acute stroke onset time could provide critical information to clinicians in deciding stroke treatment options such as thrombolysis. Patients with unknown or unwitnessed TSS are usually excluded from thrombolysis, even if their symptoms began within the therapeutic window. In this work, we demonstrate a machine learning approach for TSS classification using routinely acquired imaging sequences. We develop imaging features from the magnetic resonance (MR) images and train machine learning models to classify TSS. We also propose a deep learning model to extract hidden representations for the MR perfusion-weighted images and demonstrate classification improvement by incorporating these additional deep features. The cross-validation results show that our best classifier achieved an area under the curve of 0.765, with a sensitivity of 0.788 and a negative predictive value of 0.609, outperforming
This paper studies adaptive bandwidth selection method for local polynomial regression (LPR) and its application to multi-resolution analysis (MRA) of nonuniformly sampled data. In LPR, the observations are modeled locally by a polynomial using least-squares criterion with a kernel having a certain support or bandwidth so that a better bias-variance tradeoff can be achieved. In this paper, two bandwidth selection methods, namely the Fan and Gijbels's bandwidth selection (FGBS) method (Fan and Gijbels, Local Polynomial Modelling and Its Applications, Chapman and Hall, London, 1996; Fan and Gijbels, Stat Sin 57:371-394, 1995) in the statistical community and the intersection of confidence intervals (ICI) method commonly used in the signal and image processing communities, are reviewed and compared in terms of their performance and implementation complexity using standard testing data sets. Furthermore, using the result of Stankovi (IEEE Trans Signal Proc 52:1228-1234, 2004), a new refined ICI-based adaptive bandwidth selection method for LPR and its associated reliability analysis are proposed. In addition, recursive implementations of LPR with the two classes of bandwidth selection methods are considered for online applications. Simulation results show that the performances of the FGBS method and the refined ICI method are comparable for the data sets tested. Since LPR with adaptive bandwidths can be naturally applied to non-uniformly sampled noisy observations, we propose to use it as a pre-processing step to a conventional MRA so that a MRA of non-uniformly sampled data can be realized. Simulation results show that the proposed LPR-based MRA gives better results than conventional linear interpolation of the data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.