Frauke Wilm, Michaela Benz, Volker Bruns, Serop Baghdadlian, Jakob Dexl, David Hartmann, Petr Kuritcyn, Martin Weidenfeller, Thomas Wittenberg, Susanne Merkel, Arndt Hartmann, Markus Eckstein, Carol Immanuel Geppert
Purpose: Automatic outlining of different tissue types in digitized histological specimen provides a basis for follow-up analyses and can potentially guide subsequent medical decisions. The immense size of whole-slide-images (WSIs), however, poses a challenge in terms of computation time. In this regard, the analysis of nonoverlapping patches outperforms pixelwise segmentation approaches but still leaves room for optimization. Furthermore, the division into patches, regardless of the biological structures they contain, is a drawback due to the loss of local dependencies.
Approach: We propose to subdivide the WSI into coherent regions prior to classification by grouping visually similar adjacent pixels into superpixels. Afterward, only a random subset of patches per superpixel is classified and patch labels are combined into a superpixel label. We propose a metric for identifying superpixels with an uncertain classification and evaluate two medical applications, namely tumor area and invasive margin estimation and tumor composition analysis.
Results: The algorithm has been developed on 159 hand-annotated WSIs of colon resections and its performance is compared with an analysis without prior segmentation. The algorithm shows an average speed-up of 41% and an increase in accuracy from 93.8% to 95.7%. By assigning a rejection label to uncertain superpixels, we further increase the accuracy by 0.4%. While tumor area estimation shows high concordance to the annotated area, the analysis of tumor composition highlights limitations of our approach.
Conclusion: By combining superpixel segmentation and patch classification, we designed a fast and accurate framework for whole-slide cartography that is AI-model agnostic and provides the basis for various medical endpoints.
Since the introduction of JPEG 2000, several rate control methods have been proposed. Among them, post-compression rate-distortion optimization (PCRD-Opt) is the most widely used, and the one recommended by the standard. The approach followed by this method is to first compress the entire image split in code blocks, and subsequently, optimally truncate the set of generated bit streams according to the maximum target bit rate constraint. The literature proposes various strategies on how to estimate ahead of time where a block will get truncated in order to stop the execution prematurely and save time. However, none of them have been defined bearing in mind a parallel implementation. Today, multi-core and many-core architectures are becoming popular for JPEG 2000 codecs implementations. Therefore, in this paper, we analyze how some techniques for efficient rate control can be deployed in GPUs. In order to do that, the design of our GPU-based codec is extended, allowing stopping the process at a given point. This extension also harnesses a higher level of parallelism on the GPU, leading to up to 40% of speedup with 4K test material on a Titan X. In a second step, three selected rate control methods are adapted and implemented in our parallel encoder. A comparison is then carried out, and used to select the best candidate to be deployed in a GPU encoder, which gave an extra 40% of speedup in those situations where it was really employed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.