Efficient autofocus method for sequential automatic capturing of high-magnification microscopic images

Tello-Mijares, Santiago; Flores, Franklin César; Bescós, Jesús; Valdez, Edgar

doi:10.3788/col201311.121102

Cited by 3 publications

(8 citation statements)

References 13 publications

(30 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used normalized variance as focus metric and fast hill-climbing search as autofocus algorithm [22,23]. The algorithm searches a depth range with progressively smaller steps (here 20µm, 5µm and 1µm).…”

Section: Specimen Image Acquisitionmentioning

confidence: 99%

A method for quantitative analysis of clump thickness in cervical cytology slides

Fan

Bradley

2016

Micron

View full text Add to dashboard Cite

A c c e p t e d M a n u s c r i p tWe developed an algorithm to estimate the thickness of cervical cytology specimens.The proposed algorithm reached an accuracy of 1 micron at 90% of times.The algorithm was used to quantitatively analysis of ten normal Thin-prep cervical cytology slides.It was found that the distribution of cells is skewed towards the cover-slip (top of the slide).It was also proved that considering the thickness of focal points produced focus maps in superior qualities compared to conventional ones. Highlights (for review)Page 2 of 28 A c c e p t e d M a n u s c r i p t AbstractKnowledge of the spatial distribution and thickness of cytology specimens is critical to the development of digital slide acquisition techniques that minimise both scan times and image file size. In this paper, we evaluate a novel method to achieve this goal utilising an exhaustive high-resolution scan, an over-complete wavelet transform across multi-focal planes and a clump segmentation of all cellular material on the slide. The method is demonstrated with a quantitative analysis of ten normal, but difficult to scan Pap stained, Thin-prep, cervical cytology slides. We show that with this method the top and bottom of the specimen can be estimated to an accuracy of 1 micron in 88% and 97% of the fields of view respectively. Overall, cellar material can be over 30 microns thick and the distribution of cells is skewed towards the cover-slip (top of the slide). However, the median clump thickness is 10 microns and only 31% of clumps contain more than three nuclei. Therefore, by finding a focal map of the specimen the number of 1 micron spaced focal planes that are required to be scanned to acquire 95% of the in-focus material can be reduced from 25.4 to 21.4 on average. In addition, we show that by considering the thickness of the specimen, an improved focal map can be produced which further reduces the required number of 1 micron spaced focal planes to 18.6. This has the potential to reduce scan times and raw image data by over 25%.

show abstract

Section: Specimen Image Acquisitionmentioning

confidence: 99%

A method for quantitative analysis of clump thickness in cervical cytology slides

Fan

Bradley

2016

Micron

View full text Add to dashboard Cite

show abstract

“…This requires capturing hundredths of focused images per tissue sample (see our autofocus contributions in Ref. 6) and analyzing of these images to identify and segment cervical nuclei (see our contributions on this area in Ref. 7).…”

Section: Introductionmentioning

confidence: 99%

Region-based multifocus image fusion for the precise acquisition of Pap smear images

Tello-Mijares

Bescós

2018

J. Biomed. Opt.

View full text Add to dashboard Cite

A multifocus image fusion method to obtain a single focused image from a sequence of microscopic high-magnification Papanicolau source (Pap smear) images is presented. These images, captured each in a different position of the microscope lens, frequently show partially focused cells or parts of cells, which makes them unpractical for the direct application of image analysis techniques. The proposed method obtains a focused image with a high preservation of original pixels information while achieving a negligible visibility of the fusion artifacts. The method starts by identifying the best-focused image of the sequence; then, it performs a mean-shift segmentation over this image; the focus level of the segmented regions is evaluated in all the images of the sequence, and best-focused regions are merged in a single combined image; finally, this image is processed with an adaptive artifact removal process. The combination of a region-oriented approach, instead of block-based approaches, and a minimum modification of the value of focused pixels in the original images achieve a highly contrasted image with no visible artifacts, which makes this method especially convenient for the medical imaging domain. The proposed method is compared with several state-of-the-art alternatives over a representative dataset. The experimental results show that our proposal obtains the best and more stable quality indicators.

show abstract

“…The easiest strategy for mataining the high quality focus is to find the accurate focal depth for every point or FOV of the specimen before taking the high resolution image (Bradley et al 2005). However, this method is extremely time-consuming with average hardware, because the focusing methods need the microscopic stage to move to multiple focal depths in order to find the in-focus focal depth, hence it is subject to the mechanical limitations of the stage (Tello-Mijares et al 2013). Additionally, given the main imaging sensors for high resolution and large size image acquisition often have low frame rates, focusing in on a large number of different points can lead to a significant accumulation of scan time.…”

Section: Specimen Localisationmentioning

confidence: 99%

“…To ensure these FOVs match exactly the tiles found from processing low resolution slide images, the mechnical stage for the microscope needs to have high positional accuracy, i.e., close to the pixel resolution of low resolution images. The focus metric used was normalised variance and for each tile, and the fast hill-climbing search method for focusing, which samples a range of focal depths with progressively smaller focal steps till a maximum focus value is found (Sun et al 2004;Tello-Mijares et al 2013). Next, we manually classify inlier FOVs that had cervical cells and outlier FOVs that contained glue, ink marker artefacts or were focused on the top of the cover-slip.…”

Section: Focus Map Ground-truth Generationmentioning

confidence: 99%

See 1 more Smart Citation

Methods for rapid and high quality acquisition of whole slide images

Fan¹

View full text Add to dashboard Cite

In the past two decades, whole slide imaging (WSI) has been increasingly used in clinical pathology, research and education thanks to advances in technology. However, existing digital slide scanners for WSI are incapable of efficiently scanning cytology specimens in "glass-faithful quality". In particular, cervical cytology specimens used for cervical cancer screening need to be imaged at multiple focal planes and at high spatial resolution. So, the aim of this thesis is to develop novel methods for improving the scan efficiency of cytology specimens. Preliminary investigations narrowed the research into two directions: to optimise the scan parameters (e.g. the scan map, the focus profile map and the number of focal planes to be scanned) for scanning cytology specimens, and to investigate the fast slanted (specimen) scan.Creating an accurate scan map followed by estimating the focus profile map of the specimen can significantly reduce the scan time, but previous methods are not robust to slide artefacts such as ink marker and dusts. The first research aspect of this thesis has focused on the development of a robust method that not only creates an accurate scan map but also provides a ranked list of focus candidates required for estimating the focus map. Specifically, tiles representing field-of-views to be scanned at high resolution are located on the low resolution images and evaluated with regards to whether they contain foreground objects and how good they are as focus candidates. Four metrics were proposed for the tile evaluation: threshold index (TI) that measures the image intensity; normalised auto-correlation index (NACI) that measures the spatial image similarity; auto-phase correlation index (APCI) that measures the image phase diversity; and entropy index (EI) that measures the predictability of image intensities. The experimental results indicated that while NACI, APCI and EI are effective for specimen delineation, only APCI is capable of selecting superior focus candidates and ignoring artefacts.Knowing the three-dimensional distribution of the specimen can help to determine the minimal number of focal planes to be acquired, and hence the minimal scan time. A method has been developed to evaluate the thickness of all cell and cell clumps in cervical cytology specimens incorporating an exhaustive high-resolution scan, an over-complete wavelet transform across multi-focal planes and a cell clump segmentation method. The accuracy and usefulness of the method was evaluated and demonstrated by quantitative analysis of ten Pap stained Thin-prep cervical slides. In particular, it was found that creating a focus profile map was able to reduce the number of scanning focal planes from 25.4 to 21.4 on average. It was also shown that the distribution of cells was skewed towards the cover-slip (top of the slide), so by considering the thickness of the specimen, an improved focal map i can be produced to further reduce the required number of 1 micron spaced focal planes to 18.6.The last research aspect of...

show abstract

Efficient autofocus method for sequential automatic capturing of high-magnification microscopic images

Cited by 3 publications

References 13 publications

A method for quantitative analysis of clump thickness in cervical cytology slides

A method for quantitative analysis of clump thickness in cervical cytology slides

Region-based multifocus image fusion for the precise acquisition of Pap smear images

Methods for rapid and high quality acquisition of whole slide images

Contact Info

Product

Resources

About