Towards a computer aided diagnosis system dedicated to virtual microscopy based on stereology sampling and diffusion maps

Belhomme, Philippe; Oger, Myriam; Michels, Jean-Jaques; Plancoulaine, Benoît; Herlin, Paulette

doi:10.1186/1746-1596-6-s1-s3

Cited by 12 publications

(13 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We note that this is only an experimental implementation, with processing times too slow for full slide segmentation, but further speed improvements are possible. In addition to this, tissue sampling methods [29,30], and/or supervised extraction of relevant regions of interest [31,32] can be used in order to reduce the number of regions from the full slide that need to be processed, while still providing a relevant result.…”

Section: Discussionmentioning

confidence: 99%

Automatic Nuclei Segmentation in H&E Stained Breast Cancer Histopathology Images

et al. 2013

View full text Add to dashboard Cite

The introduction of fast digital slide scanners that provide whole slide images has led to a revival of interest in image analysis applications in pathology. Segmentation of cells and nuclei is an important first step towards automatic analysis of digitized microscopy images. We therefore developed an automated nuclei segmentation method that works with hematoxylin and eosin (H&E) stained breast cancer histopathology images, which represent regions of whole digital slides. The procedure can be divided into four main steps: 1) pre-processing with color unmixing and morphological operators, 2) marker-controlled watershed segmentation at multiple scales and with different markers, 3) post-processing for rejection of false regions and 4) merging of the results from multiple scales. The procedure was developed on a set of 21 breast cancer cases (subset A) and tested on a separate validation set of 18 cases (subset B). The evaluation was done in terms of both detection accuracy (sensitivity and positive predictive value) and segmentation accuracy (Dice coefficient). The mean estimated sensitivity for subset A was 0.875 (±0.092) and for subset B 0.853 (±0.077). The mean estimated positive predictive value was 0.904 (±0.075) and 0.886 (±0.069) for subsets A and B, respectively. For both subsets, the distribution of the Dice coefficients had a high peak around 0.9, with the vast majority of segmentations having values larger than 0.8.

show abstract

Section: Discussionmentioning

confidence: 99%

Automatic Nuclei Segmentation in H&E Stained Breast Cancer Histopathology Images

et al. 2013

View full text Add to dashboard Cite

show abstract

“…Stratified sampling procedure was performed according to the WHO 2010 classification [17,18]. Ki67 indices were calculated as a percentage of Ki67 positive cells in 500—2000 cells that were counted in areas of strongest nuclear labeling (“hot spots”).…”

Section: Methodsmentioning

confidence: 99%

Clinical validation of the gastrointestinal NET grading system: Ki67 index criteria of the WHO 2010 classification is appropriate to predict metastasis or recurrence

et al. 2013

View full text Add to dashboard Cite

BackgroundIn the WHO 2010 classification, the neuroendocrine tumors (NETs) are subdivided by their mitotic index or Ki67 index into either G1 or G2 NETs. Tumors with a Ki67 index of <2% are classified as G1 and those with 3—20% are classified as G2. However, the assessment of tumors with Ki67 index of greater than 2% and less than or equal to 3% is still unclear. To resolve the problem, we validated the Ki67 index criteria of gastrointestinal NETs of the WHO 2010 classification.MethodsThe medical records of 45 patients who were pathologically diagnosed as having NET G1/G2 of the gastrointestinal tract were analyzed retrospectively. According to the WHO 2010 classification, Ki67 index were calculated. Computer-assisted cytometrical analysis of Ki67 immunoreactivity was performed using the WinRooF image processing software. Receiver operating characteristic (ROC) curves were generated to determine the best discriminating Ki67 index. To clarify the assessment of tumors with Ki67 index between 2—3%, the calculated cutoff of Ki67 index was evaluated using Fisher’s exact test.ResultsROC curve analysis confirmed that 2.8% was the best Ki67 index cutoff value for predicting metastasis or recurrence. The sensitivity of the new Ki67 index cutoff was 42.9%, and the specificity was 86.8%.ConclusionsDivision of NETs into G1/G2 based on Ki67 index of 3% was appropriate to predict metastases or recurrences. The WHO grading system may be the most useful classification to predict metastases or recurrences.Virtual SlidesThe virtual slide(s) for this article can be found here: http://www.diagnosticpathology.diagnomx.eu/vs/1553036118943799

show abstract

“…The first part has been presented in [4]. We have shown that combining stereological sampling and data reduction based on diffusion maps offers an interesting general framework.…”

Section: Resultsmentioning

confidence: 99%

Out-of-sample extension of diffusion maps in a computer aided diagnosis system. Application to breast cancer virtual slide images

et al. 2013

View full text Add to dashboard Cite

Towards a computer aided diagnosis system dedicated to virtual microscopy based on stereology sampling and diffusion maps

Cited by 12 publications

References 8 publications

Automatic Nuclei Segmentation in H&E Stained Breast Cancer Histopathology Images

Automatic Nuclei Segmentation in H&E Stained Breast Cancer Histopathology Images

Clinical validation of the gastrointestinal NET grading system: Ki67 index criteria of the WHO 2010 classification is appropriate to predict metastasis or recurrence

Out-of-sample extension of diffusion maps in a computer aided diagnosis system. Application to breast cancer virtual slide images

Contact Info

Product

Resources

About