Limits...
Robust automated tumour segmentation on histological and immunohistochemical tissue images.

Wang CW, Fennell D, Paul I, Savage K, Hamilton P - PLoS ONE (2011)

Bottom Line: The presented technique greatly improves the process of TMA construction and plays an important role in automated IHC quantification in biomarker analysis where excluding stroma areas is critical.With the finest pixel-based evaluation (instead of area-based or object-based), the experimental results show that the proposed method is able to achieve 80% accuracy and 78% accuracy in two different types of pathological virtual slides, i.e., routine histochemical H&E and IHC images, respectively.The presented technique greatly reduces labor-intensive workloads for pathologists and highly speeds up the process of TMA construction and provides a possibility for fully automated IHC quantification.

View Article: PubMed Central - PubMed

Affiliation: Graduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology, Taipei City, Taiwan. cweiwang@mail.ntust.edu.tw

ABSTRACT
Tissue microarray (TMA) is a high throughput analysis tool to identify new diagnostic and prognostic markers in human cancers. However, standard automated method in tumour detection on both routine histochemical and immunohistochemistry (IHC) images is under developed. This paper presents a robust automated tumour cell segmentation model which can be applied to both routine histochemical tissue slides and IHC slides and deal with finer pixel-based segmentation in comparison with blob or area based segmentation by existing approaches. The presented technique greatly improves the process of TMA construction and plays an important role in automated IHC quantification in biomarker analysis where excluding stroma areas is critical. With the finest pixel-based evaluation (instead of area-based or object-based), the experimental results show that the proposed method is able to achieve 80% accuracy and 78% accuracy in two different types of pathological virtual slides, i.e., routine histochemical H&E and IHC images, respectively. The presented technique greatly reduces labor-intensive workloads for pathologists and highly speeds up the process of TMA construction and provides a possibility for fully automated IHC quantification.

Show MeSH

Related in: MedlinePlus

Tissue Microarray Construction.a. donor tissue blocks are selected, b. a needle is used to sample multiple cylindric tissue cores from each donor block and the sampling locations are carefully chosen by an experienced pathologist based on the top H&E slide of the block, c. the obtained tissue cores are assembled in a single microarray, d. the finished tissue microarray block is sectioned to create multiple TMAs where periodically a TMA slide is stained with H&E with all tissue cores examined by an experienced pathologist to verify if cancerous cells exist, e. tumour is with irregular shape and size; sections of cylindric tissue cores may not contain cancerous cells.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3046129&req=5

pone-0015818-g001: Tissue Microarray Construction.a. donor tissue blocks are selected, b. a needle is used to sample multiple cylindric tissue cores from each donor block and the sampling locations are carefully chosen by an experienced pathologist based on the top H&E slide of the block, c. the obtained tissue cores are assembled in a single microarray, d. the finished tissue microarray block is sectioned to create multiple TMAs where periodically a TMA slide is stained with H&E with all tissue cores examined by an experienced pathologist to verify if cancerous cells exist, e. tumour is with irregular shape and size; sections of cylindric tissue cores may not contain cancerous cells.

Mentions: TMAs are produced by a method of re-locating tissue from histologic paraffin blocks such that tissue from multiple patients can be studied on the same slide (commonly, three to five tissue cores are extracted from each donor block). This is done by using a needle to biopsy a standard histologic sections and placing the core into an array on a recipient paraffin block (Fig. 1a,b,c), using a tissue microarrayer. The new block is then cut into 4-micron or 5-micron thick sections that contain 40 to hundreds tissue specimens (Fig. 1d), and these sections can then be stained using standard laboratory methods such as immunohistochemistry for various biomarker studies. In constructing TMAs, the location to sample each tissue core from individual donor blocks is carefully selected by an experienced pathologist at a region containing large amounts of cancer cells of the top H&E section. Tumour is a 3D object and has irregular shape, and thus the obtained cylindrical specimens (tissue cores) may not contain cancerous cell for all TMA sections; as illustrated in Fig. 1e, the tissue core 1 in a number of TMA sections derived from the middle of the cylindric specimens does not contain cancerous cell. In addition, it is unpredictable how deep the tumour is. Hence, periodically TMA slides are stained with H&E and pathologists have to visually examine all the tissue cores across TMAs (Fig. 1d), which is an extremely time consuming and labor-intensive process.


Robust automated tumour segmentation on histological and immunohistochemical tissue images.

Wang CW, Fennell D, Paul I, Savage K, Hamilton P - PLoS ONE (2011)

Tissue Microarray Construction.a. donor tissue blocks are selected, b. a needle is used to sample multiple cylindric tissue cores from each donor block and the sampling locations are carefully chosen by an experienced pathologist based on the top H&E slide of the block, c. the obtained tissue cores are assembled in a single microarray, d. the finished tissue microarray block is sectioned to create multiple TMAs where periodically a TMA slide is stained with H&E with all tissue cores examined by an experienced pathologist to verify if cancerous cells exist, e. tumour is with irregular shape and size; sections of cylindric tissue cores may not contain cancerous cells.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3046129&req=5

pone-0015818-g001: Tissue Microarray Construction.a. donor tissue blocks are selected, b. a needle is used to sample multiple cylindric tissue cores from each donor block and the sampling locations are carefully chosen by an experienced pathologist based on the top H&E slide of the block, c. the obtained tissue cores are assembled in a single microarray, d. the finished tissue microarray block is sectioned to create multiple TMAs where periodically a TMA slide is stained with H&E with all tissue cores examined by an experienced pathologist to verify if cancerous cells exist, e. tumour is with irregular shape and size; sections of cylindric tissue cores may not contain cancerous cells.
Mentions: TMAs are produced by a method of re-locating tissue from histologic paraffin blocks such that tissue from multiple patients can be studied on the same slide (commonly, three to five tissue cores are extracted from each donor block). This is done by using a needle to biopsy a standard histologic sections and placing the core into an array on a recipient paraffin block (Fig. 1a,b,c), using a tissue microarrayer. The new block is then cut into 4-micron or 5-micron thick sections that contain 40 to hundreds tissue specimens (Fig. 1d), and these sections can then be stained using standard laboratory methods such as immunohistochemistry for various biomarker studies. In constructing TMAs, the location to sample each tissue core from individual donor blocks is carefully selected by an experienced pathologist at a region containing large amounts of cancer cells of the top H&E section. Tumour is a 3D object and has irregular shape, and thus the obtained cylindrical specimens (tissue cores) may not contain cancerous cell for all TMA sections; as illustrated in Fig. 1e, the tissue core 1 in a number of TMA sections derived from the middle of the cylindric specimens does not contain cancerous cell. In addition, it is unpredictable how deep the tumour is. Hence, periodically TMA slides are stained with H&E and pathologists have to visually examine all the tissue cores across TMAs (Fig. 1d), which is an extremely time consuming and labor-intensive process.

Bottom Line: The presented technique greatly improves the process of TMA construction and plays an important role in automated IHC quantification in biomarker analysis where excluding stroma areas is critical.With the finest pixel-based evaluation (instead of area-based or object-based), the experimental results show that the proposed method is able to achieve 80% accuracy and 78% accuracy in two different types of pathological virtual slides, i.e., routine histochemical H&E and IHC images, respectively.The presented technique greatly reduces labor-intensive workloads for pathologists and highly speeds up the process of TMA construction and provides a possibility for fully automated IHC quantification.

View Article: PubMed Central - PubMed

Affiliation: Graduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology, Taipei City, Taiwan. cweiwang@mail.ntust.edu.tw

ABSTRACT
Tissue microarray (TMA) is a high throughput analysis tool to identify new diagnostic and prognostic markers in human cancers. However, standard automated method in tumour detection on both routine histochemical and immunohistochemistry (IHC) images is under developed. This paper presents a robust automated tumour cell segmentation model which can be applied to both routine histochemical tissue slides and IHC slides and deal with finer pixel-based segmentation in comparison with blob or area based segmentation by existing approaches. The presented technique greatly improves the process of TMA construction and plays an important role in automated IHC quantification in biomarker analysis where excluding stroma areas is critical. With the finest pixel-based evaluation (instead of area-based or object-based), the experimental results show that the proposed method is able to achieve 80% accuracy and 78% accuracy in two different types of pathological virtual slides, i.e., routine histochemical H&E and IHC images, respectively. The presented technique greatly reduces labor-intensive workloads for pathologists and highly speeds up the process of TMA construction and provides a possibility for fully automated IHC quantification.

Show MeSH
Related in: MedlinePlus