Focus classification in digital holographic microscopy using deep convolutional neural networks

Pitkäaho, Tomi; Manninen, Aki; Naughton, Thomas J.

doi:10.1117/12.2286161

Cited by 7 publications

(10 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The greatest benefit of the deep learning method outlined in this paper is that after training, the in-focus depth can be obtained from the hologram plane intensity directly, in constant time, without any numerical propagation. It should be noted that this paper extends results reported in [26,27] by rigorously verifying the performance of the approach in a regression context. Ren et al [30] reported a deep-learning-based approach for both amplitude and phase objects.…”

Section: Digital Holographic Microscopysupporting

confidence: 75%

“…In a rigorous treatment of our preliminary results [26,27], one of the first applications of deep learning to digital holographic microscopy, we show that a deep artificial neural network can be designed to learn the appropriate in-focus depth of an arbitrary MDCK cell cluster encoded in a digital hologram. Its greatest benefit is that the in-focus depth can be obtained from the hologram plane intensity only, and in constant time, without any numerical propagation.…”

Section: Discussionmentioning

confidence: 99%

“…Conventional artificial neural networks have been applied before in the fields of digital holographic microscopy [21,23] and, more generally, digital holography [24,25]. Recently, deep learning has been applied in different holographic applications [26][27][28][29][30][31][32][33][34][35][36][37][38].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Focus prediction in digital holographic microscopy using deep convolutional neural networks

2019

Self Cite

View full text Add to dashboard Cite

Deep artificial neural network learning is an emerging tool in image analysis. We demonstrate its potential in the field of digital holographic microscopy by addressing the challenging problem of determining the in-focus reconstruction depth of Madin-Darby canine kidney cell clusters encoded in digital holograms. A deep convolutional neural network learns the in-focus depths from half a million hologram amplitude images. The trained network correctly determines the in-focus depth of new holograms with high probability, without performing numerical propagation. This paper reports on extensions to preliminary work published earlier as one of the first applications of deep learning in the field of digital holographic microscopy.

show abstract

Section: Digital Holographic Microscopysupporting

confidence: 75%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Focus prediction in digital holographic microscopy using deep convolutional neural networks

2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…In Fig. 8(g), the Region-recognition Filter (F RR ) is built with shape recognition of the spectrum and iterative calculation of threshold [21] and the shape of filtered spectrum is consistent with the energy distribution of + 1 term. In Fig.…”

Section: Experimental Testing and Resultsmentioning

confidence: 99%

“…Recently, deep learning has been introduced to reconstruction processes of digital holography, such as autofocusing [21], phase aberration compensations [22], etc. The technique can realize adaptation and full automation because it takes pre-labeled results as a guide to learn the intrinsic features of the data set.…”

Section: Introductionmentioning

confidence: 99%

Adaptive frequency filtering based on convolutional neural networks in off-axis digital holographic microscopy

Xiao¹,

Wang²,

Pan³

et al. 2019

Biomed. Opt. Express

View full text Add to dashboard Cite

Digital holographic microscopy (DHM) as a label-free quantitative imaging tool has been widely used to investigate the morphology of living cells dynamically. In the offaxis DHM, the spatial filtering in the frequency spectrum of the hologram is vital to the quality of the reconstructed images. In this paper, we propose an adaptive spatial filtering approach based on convolutional neural networks (CNN) to automatically extracts the optimal shape of frequency components. For achieving robust and precise recognition performance, the net model is trained by using the tens of thousands of frequency spectrums with a variety of specimens and imaging conditions. The experimental results demonstrate that the trained network produce an adaptive spatial filtering window which can accurately select the frequency components of the object term and eliminate the frequency components of the interference terms, especially the coherent noise that overlaps with the object term in the spatial frequency domain. We find that the proposed approach has a fast, robust, and outstanding frequency filtering capability without any manual intervention and initial input parameters compared to previous techniques. Furthermore, the applicability of the proposed method in off-axis DHM for dynamic analysis is demonstrated by real-time monitoring the morphologic changes of living MLO-Y4 cells that are constantly subject to Fluid Shear Stress (FSS).

show abstract

Deep convolutional neural network for binary regression of three-dimensional objects using information retrieved from digital Fresnel holograms

Mahesh

Nelleri

2022

Appl. Phys. B

View full text Add to dashboard Cite

Focus classification in digital holographic microscopy using deep convolutional neural networks

Abstract: In digital holographic microscopy, one often obtains an in-focus image of the sample by applying a focus metric to a stack of numerical reconstructions. We present an alternative approach using a deep convolutional neural network.

Cited by 7 publications

References 9 publications

Focus prediction in digital holographic microscopy using deep convolutional neural networks

Focus prediction in digital holographic microscopy using deep convolutional neural networks

Adaptive frequency filtering based on convolutional neural networks in off-axis digital holographic microscopy

Deep convolutional neural network for binary regression of three-dimensional objects using information retrieved from digital Fresnel holograms

Contact Info

Product

Resources

About