Automatic signal classification in fluorescence in situ hybridization images

Lerner, Boaz; Clocksin, William F.; Dhanjal, Seema; Hultén, Maj; Bishop, Christopher M.

doi:10.1002/1097-0320(20010201)43:2<87::aid-cyto1022>3.0.co;2-#

Cited by 42 publications

(54 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After background correction (SBMinGray algorithm) and intensity thresholding (counterstain object threshould: 10%), the DAPI objects were used as a counterstain mask; FISH signals that appeared outside of DAPI objects were disregarded. FISH signals were segmented after a TopHat transformation as background correction using parameters attained through an optimization process, as described previously (18,19). Grid units without red or green signals were discarded.…”

Section: Automated Microscopymentioning

confidence: 99%

Automated FISH analysis using dual‐fusion and break‐apart probes on paraffin‐embedded tissue sections

et al. 2008

View full text Add to dashboard Cite

Detecting balanced translocations using tissue sections plays an important diagnostic role in cases of hematological malignancies. Manual scoring is often problematic due to truncation and overlapping of nuclei. Reports have described automated analysis using primarily tile sampling. The aim of this study was to investigate an automated fluorescent in situ hybridization analysis method using grid sampling on tissue sections, and compare the performance of dual-fusion (DF) and break-apart (BA) probes in this setting. Ten follicular, 10 mantle cell lymphoma, and 10 translocation-negative samples were used to set the threshold of false positivity using IGH/CCND1, IGH/BCL-2 DF, and IGH BA probes. The cut-off distances of red and green signals to define fusion signals were 0.5, 1.0, and 1.2 lm for the IGH/CCND1, IGH/BCL-2 DF, and IGH BA probes, respectively. The mean false positivity of grid units was 5.3, 11.4, and 28.1%, respectively. Ten to 14 additional samples analyzed blindly and were correctly classified using each probe. Discriminating positive and negative samples using automated analysis and grid sampling was possible with each probe, although different definitions of fusion signals were required due to the different physical distances between the DNA probes. Using the DF probes resulted in lower false positivity, which was less affected by signal numbers per grid units. ' International Society for Advancement of Cytometry Key termsFISH; automated analysis; cytogenetics; paraffin; lymphoma; FFPE; tissue sections; dual-fusion; break-apart; grid sampling RECURRENT, structural cytogenetic abnormalities, most frequently balanced translocations play an important diagnostic and prognostic role in cases of hematological malignancies (1). Frequently, only formalin-fixed, paraffin-embedded (FFPE) material is available for demonstrating such abnormalities. Several recent reports suggest that out of the available molecular techniques, fluorescent in situ hybridization (FISH) is the most reliable and appropriate for detecting translocations in FFPE samples (2-6). Several groups reported on FISH analysis of FFPE material, which may be performed on isolated nuclei (7-10) or on tissue sections (11-15). The major advantage of the former is that whole nuclei can be analyzed free from truncation artefacts. However, isolation of nuclei is more time-and resource consuming, than analyzing tissue sections and valuable information regarding histological topography is lost. If the sample contains only a small focus of neoplastic cells, their frequency in suspensions of isolated nuclei may drop below the false positivity rate of the FISH analysis.Because of recent advancement of image analysis and computer science, automated FISH analysis is becoming a desired possibility (16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26). A recent report

show abstract

Section: Automated Microscopymentioning

confidence: 99%

Automated FISH analysis using dual‐fusion and break‐apart probes on paraffin‐embedded tissue sections

et al. 2008

View full text Add to dashboard Cite

show abstract

“…For example, Netten et al, reported the first automated FISH image analysis system that enables to count FISH signals that are neither split nor stringy in a single spectrum [16]. Since then, several other automated FISH image analysis systems and schemes have been developed and tested [3,[17][18][19]. In these schemes, different image processing methods, including the user-defined thresholds [3], artificial neural networks [17], the watershed algorithm [19], and the Isodata algorithm [20], were used to segment interphase cell nuclei.…”

Section: Introductionmentioning

confidence: 99%

“…Since then, several other automated FISH image analysis systems and schemes have been developed and tested [3,[17][18][19]. In these schemes, different image processing methods, including the user-defined thresholds [3], artificial neural networks [17], the watershed algorithm [19], and the Isodata algorithm [20], were used to segment interphase cell nuclei. Despite the reported encouraging results and progress, none of these automated FISH image analysis methods have been routinely used in cytogenetic laboratories due to both the hardware and software limitations (i.e., the existing commercialized FISH image scanning systems are unable to acquire high-resolution FISH images in a fully-automated image scanning mode and the computerized schemes are unable to accurately distinguish and merge splitting and stringy FISH signals).…”

Section: Introductionmentioning

confidence: 99%

FluorescenceIn SituHybridization (FISH) Signal Analysis Using Automated Generated Projection Images

Wang

Chen

et al. 2012

Analytical Cellular Pathology

View full text Add to dashboard Cite

Fluorescencein situhybridization (FISH) tests provide promising molecular imaging biomarkers to more accurately and reliably detect and diagnose cancers and genetic disorders. Since current manual FISH signal analysis is low-efficient and inconsistent, which limits its clinical utility, developing automated FISH image scanning systems and computer-aided detection (CAD) schemes has been attracting research interests. To acquire high-resolution FISH images in a multi-spectral scanning mode, a huge amount of image data with the stack of the multiple three-dimensional (3-D) image slices is generated from a single specimen. Automated preprocessing these scanned images to eliminate the non-useful and redundant data is important to make the automated FISH tests acceptable in clinical applications. In this study, a dual-detector fluorescence image scanning system was applied to scan four specimen slides with FISH-probed chromosome X. A CAD scheme was developed to detect analyzable interphase cells and map the multiple imaging slices recorded FISH-probed signals into the 2-D projection images. CAD scheme was then applied to each projection image to detect analyzable interphase cells using an adaptive multiple-threshold algorithm, identify FISH-probed signals using a top-hat transform, and compute the ratios between the normal and abnormal cells. To assess CAD performance, the FISH-probed signals were also independently visually detected by an observer. The Kappa coefficients for agreement between CAD and observer ranged from 0.69 to 1.0 in detecting/counting FISH signal spots in four testing samples. The study demonstrated the feasibility of automated FISH signal analysis that applying a CAD scheme to the automated generated 2-D projection images.

show abstract

“…Reliable spot detection is also essential in the Photoactivation Localization Microscopy (PALM) and (14,15). The importance of using a 3D approach in spatial arrangement studies, in comparison to a 2D analysis, has been noted (16,17).Existing automated methods for spot detection can be subdivided into two main groups: (1) unsupervised methods, which do not include a learning step, and (2) supervised (machine learning) methods (18)(19)(20). In a previous comparative study (21) on 2D detection methods, it was shown that for images whose signal-to-noise ratio (SNR) was very low (SNR % 2), the supervised methods performed best overall.…”

mentioning

confidence: 99%

Performance and sensitivity evaluation of 3D spot detection methods in confocal microscopy

Štěpka

Matula

et al. 2015

Cytometry Pt A

View full text Add to dashboard Cite

Reliable 3D detection of diffraction-limited spots in fluorescence microscopy images is an important task in subcellular observation. Generally, fluorescence microscopy images are heavily degraded by noise and non-specifically stained background, making reliable detection a challenging task. In this work, we have studied the performance and parameter sensitivity of eight recent methods for 3D spot detection. The study is based on both 3D synthetic image data and 3D real confocal microscopy images. The synthetic images were generated using a simulator modeling the complete imaging setup, including the optical path as well as the image acquisition process. We studied the detection performance and parameter sensitivity under different noise levels and under the influence of uneven background signal. To evaluate the parameter sensitivity, we propose a novel measure based on the gradient magnitude of the F 1 score. We measured the success rate of the individual methods for different types of the image data and found that the type of image degradation is an important factor. Using the F 1 score and the newly proposed sensitivity measure, we found that the parameter sensitivity is not necessarily proportional to the success rate of a method. This also provided an explanation why the best performing method for synthetic data was outperformed by other methods when applied to the real microscopy images. On the basis of the results obtained, we conclude with the recommendation of the HDome method for data with relatively low variations in quality, or the Sorokin method for image sets in which the quality varies more. We also provide alternative recommendations for high-quality images, and for situations in which detailed parameter tuning might be deemed expensive. V C 2015 International Society for Advancement of Cytometry Key terms fluorescence microscopy; 3D imaging; diffraction-limited spot detection; parameter sensitivity RELIABLE spot detection in 3D confocal microscopy is an important task in cell observation. The objects of interest such as proteins or other biomolecules (e.g., sequences of nucleic acids) are typically fluorescently tagged and they appear as bright, diffraction-limited particles in 3D images. For example, the proteins of interest can be genetically modified to become fluorescent while keeping their function (1), or fluorescent antibodies can be attached to proteins (2). The sequences of nucleic acids can be labeled by Fluorescence In-Situ Hybridization (FISH), which allows us to visualize parts of chromosomes such as telomeres or centromeres or even individual genes (3). It is known that the spatial distribution of subcellular objects is related to their function, and thus reliable 3D spot detection is of paramount importance in many biological applications, for example, in particle tracking (4), studies on the spatial arrangement of genes (5-7), in telomere studies (8-11), kinetochore studies (12), and in co-localization studies (13). Reliable spot detection is also essential in the Photoac...

show abstract

Automatic signal classification in fluorescence in situ hybridization images

Cited by 42 publications

References 8 publications

Automated FISH analysis using dual‐fusion and break‐apart probes on paraffin‐embedded tissue sections

Automated FISH analysis using dual‐fusion and break‐apart probes on paraffin‐embedded tissue sections

FluorescenceIn SituHybridization (FISH) Signal Analysis Using Automated Generated Projection Images

Performance and sensitivity evaluation of 3D spot detection methods in confocal microscopy

Contact Info

Product

Resources

About