Fast Segmentation for Texture-based Cartography of whole Slide Images

Apou, Grégory; Naegel, Benoît; Forestier, Germain; Feuerhake, Friedrich; Wemmert, Cédric

doi:10.5220/0004687403090319

Cited by 2 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On prostate specimens, Gorelick et al 18 achieved a class-averaged recall of 83.88% on eight tissue classes with an inference time of 2 min per 300 × 300 pixel sized patch. Apou et al 19 segmented breast cancer WSI into six classes and achieved a class-averaged sensitivity of 55.83% and a class-averaged specificity of 91.4%. The authors stated inference times of under 2 h per WSI.…”

Section: Applications In Digital Pathologymentioning

confidence: 99%

“…26 The authors, however, merely performed a segmentation and did not infer labels for the computed superpixels. Other existing works manually extracted handcrafted superpixel feature vectors which were then classified using machine learning-based classifiers and thereby enabled a binary 22,27 or multiclass 12,18,19,28 semantic segmentation of medical images. On histological image data, this approach has facilitated the binary segmentation of WSIs in 20 to 45 min by Bejnordi et al 27 and up to 60 min by Balazsi et al 22 with good performance results indicated by Dice scores of 92.43% 27 and 69%, 22 respectively.…”

Section: Superpixel Classificationmentioning

confidence: 99%

See 1 more Smart Citation

Fast whole-slide cartography in colon cancer histology using superpixels and CNN classification

et al. 2022

View full text Add to dashboard Cite

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.

show abstract

Section: Applications In Digital Pathologymentioning

confidence: 99%

Section: Superpixel Classificationmentioning

confidence: 99%

Fast whole-slide cartography in colon cancer histology using superpixels and CNN classification

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In the field of texture-based segmenta-tion approaches Signolle et al 17 propose a method that incorporates several binary hidden wavelet domain Markov tree classifiers whose outputs are combined using majority voting. 19 . Due to the high variations in hardware resources and annotation quality, it is oftentimes difficult to compare image analysis algorithms in terms of classification accuracy and computational costs.…”

Section: Applications In Digital Pathologymentioning

confidence: 99%

“…superpixels, has proven advantageous. The potential of superpixel segmentation prior to classification has been shown for various applications including salient object detection 25 , road segmentation 26,27 and binary 22,28 or multi-class 18,19 semantic segmentation of medical images. The authors manually extract superpixel feature vectors which are provided as input to machine learning-based classifiers.…”

Section: Superpixel Classificationmentioning

confidence: 99%

Fast whole-slide cartography in colon cancer histology using superpixels and CNN classification

Wilm,

Benz,

Bruns

et al. 2021

Preprint

View full text Add to dashboard Cite

Whole-slide-image cartography is the process of automatically detecting and outlining different tissue types in digitized histological specimen. This semantic segmentation provides a basis for many follow-up analyses and can potentially guide subsequent medical decisions. Due to their large size, whole-slide-images typically have to be divided into smaller patches which are then analyzed individually using machine learning-based approaches. Thereby, local dependencies of image regions get lost and since a whole-slide-image comprises many thousands of such patches this process is inherently slow. We propose to subdivide the image into coherent regions prior to classification by grouping visually similar adjacent image pixels into larger segments, i.e. superpixels. Afterwards, only a random subset of patches per superpixel is classified and patch labels are combined into a single superpixel label. The algorithm has been developed and validated on a dataset of 159 hand-annotated whole-slideimages of colon resections and its performance has been compared to a standard patch-based approach. The algorithm shows an average speed-up of 41 % on the test data and the overall accuracy is increased from 93.8 % to 95.7 %. We additionally propose a metric for identifying superpixels with an uncertain classification so they can be excluded from further analysis. Finally, we evaluate two potential medical applications, namely tumor area estimation including tumor invasive margin generation and tumor composition analysis.

show abstract

Fast Segmentation for Texture-based Cartography of whole Slide Images

Cited by 2 publications

References 17 publications

Fast whole-slide cartography in colon cancer histology using superpixels and CNN classification

Fast whole-slide cartography in colon cancer histology using superpixels and CNN classification

Fast whole-slide cartography in colon cancer histology using superpixels and CNN classification

Contact Info

Product

Resources

About