Dr. Di Xiao Profile

Dr. Di Xiao

Scientist at Australian e-Health Research Ctr

SPIE Involvement:

Author

Publications (15)

SPIE Journal Paper | 23 April 2021 Open Access

Deep feature loss to denoise OCT images using deep neural networks

Maryam Mehdizadeh, Cara MacNish, Di Xiao, David Alonso-Caneiro, Jason Kugelman, Mohammed Bennamoun

JBO, Vol. 26, Issue 04, 046003, (April 2021) https://doi.org/10.1117/12.10.1117/1.JBO.26.4.046003

KEYWORDS: Optical coherence tomography, Denoising, Neural networks, Speckle, Image denoising, Signal attenuation, Feature extraction, Network architectures, Visualization, Image processing

Read Abstract +

Significance: Speckle noise is an inherent limitation of optical coherence tomography (OCT) images that makes clinical interpretation challenging. The recent emergence of deep learning could offer a reliable method to reduce noise in OCT images. Aim: We sought to investigate the use of deep features (VGG) to limit the effect of blurriness and increase perceptual sharpness and to evaluate its impact on the performance of OCT image denoising (DnCNN). Approach: Fifty-one macula-centered OCT pairs were used in training of the network. Another set of 20 OCT pair was used for testing. The DnCNN model was cascaded with a VGG network that acted as a perceptual loss function instead of the traditional losses of L1 and L2. The VGG network remains fixed during the training process. We focused on the individual layers of the VGG-16 network to decipher the contribution of each distinctive layer as a loss function to produce denoised OCT images that were perceptually sharp and that preserved the faint features (retinal layer boundaries) essential for interpretation. The peak signal-to-noise ratio (PSNR), edge-preserving index, and no-reference image sharpness/blurriness [perceptual sharpness index (PSI), just noticeable blur (JNB), and spectral and spatial sharpness measure (S3)] metrics were used to compare deep feature losses with the traditional losses. Results: The deep feature loss produced images with high perceptual sharpness measures at the cost of less smoothness (PSNR) in OCT images. The deep feature loss outperformed the traditional losses (L1 and L2) for all of the evaluation metrics except for PSNR. The PSI, S3, and JNB estimates of deep feature loss performance were 0.31, 0.30, and 16.53, respectively. For L1 and L2 losses performance, the PSI, S3, and JNB were 0.21 and 0.21, 0.17 and 0.16, and 14.46 and 14.34, respectively. Conclusions: We demonstrate the potential of deep feature loss in denoising OCT images. Our preliminary findings suggest research directions for further investigation.

DOWNLOAD PAPER

Proceedings Article | 24 March 2016 Paper

Detection and measurement of retinal blood vessel pulsatile motion

Di Xiao, Shaun Frost, Janardhan Vignarajan, Dong An, Mei-Ling Tay-Kearney, Yogi Kanagasingam

Proceedings Volume 9785, 97852C (2016) https://doi.org/10.1117/12.2217339

KEYWORDS: Blood, Image processing, Electroluminescence, Motion detection

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Improvement of retinal blood vessel detection by spur removal and Gaussian matched filtering compensation

Di Xiao, Janardhan Vignarajan, Dong An, Mei-Ling Tay-Kearney, Yogi Kanagasingam

Proceedings Volume 9785, 97852E (2016) https://doi.org/10.1117/12.2216799

KEYWORDS: Blood vessels, Eye, Retina, Image segmentation, Blood, Electroluminescence, Binary data, Image enhancement, Image processing

Read Abstract +

Proceedings Article | 28 February 2013 Paper

Region-based multi-step optic disk and cup segmentation from color fundus image

Di Xiao, Jane Lock, Javier Moreno Manresa, Janardhan Vignarajan, Mei-Ling Tay-Kearney, Yogesan Kanagasingam

Proceedings Volume 8670, 86702H (2013) https://doi.org/10.1117/12.2006738

KEYWORDS: Optical discs, Image segmentation, Blood vessels, Visualization, RGB color model, Image processing, Image enhancement, Image filtering, Channel projecting optics, Optic nerve

Read Abstract +

Proceedings Article | 23 February 2012 Paper

Retinal image enhancement and registration for the evaluation of longitudinal changes

Di Xiao, Shaun Frost, Janardhan Vignarajan, Jane Lock, Mei-Ling Tay-Kearney, Yogesan Kanagasingam

Proceedings Volume 8315, 83152O (2012) https://doi.org/10.1117/12.911180

KEYWORDS: Image enhancement, Image registration, Blood vessels, Blood, Image segmentation, Image processing, Image quality, Image analysis, Medical imaging, Retina

Read Abstract +

Proceedings Article | 11 March 2011 Paper

An accurate 3D shape context based non-rigid registration method for mouse whole-body skeleton registration

Di Xiao, David Zahra, Pierrick Bourgeat, Paula Berghofer, Oscar Acosta Tamayo, Catriona Wimberley, Marie Gregoire, Olivier Salvado

Proceedings Volume 7962, 79620V (2011) https://doi.org/10.1117/12.877915

KEYWORDS: Image registration, Computed tomography, Error analysis, 3D modeling, 3D acquisition, Target detection, 3D image processing, Image restoration, Shape analysis, Skull

Read Abstract +

Proceedings Article | 9 March 2011 Paper

Mouse whole-body organ mapping by non-rigid registration approach

Di Xiao, David Zahra, Pierrick Bourgeat, Paula Berghofer, Oscar Acosta Tamayo, Heather Green, Marie Claude Gregoire, Olivier Salvado

Proceedings Volume 7965, 79650E (2011) https://doi.org/10.1117/12.877921

KEYWORDS: Skin, Lung, Image registration, Computed tomography, Kidney, 3D acquisition, Brain, Associative arrays, Tumors, 3D image processing

Read Abstract +

Proceedings Article | 13 March 2010 Paper

An improved 3D shape context registration method for non-rigid surface registration

Di Xiao, David Zahra, Pierrick Bourgeat, Paula Berghofer, Oscar Acosta Tamayo, Catriona Wimberley, Marie-Claude Gregoire, Olivier Salvado

Proceedings Volume 7623, 76232Q (2010) https://doi.org/10.1117/12.844058

KEYWORDS: Image registration, 3D modeling, 3D image processing, Error analysis, Computed tomography, Image processing, 3D acquisition, Compact discs, Positron emission tomography, Detection and tracking algorithms

Read Abstract +

Proceedings Article | 9 March 2010 Paper

A non-rigid registration method for mouse whole body skeleton registration

Di Xiao, David Zahra, Pierrick Bourgeat, Paula Berghofer, Oscar Acosta Tamayo, Catriona Wimberley, Marie Claude Gregoire, Olivier Salvado

Proceedings Volume 7626, 76261N (2010) https://doi.org/10.1117/12.844067

KEYWORDS: Image registration, Computed tomography, 3D modeling, 3D acquisition, Error analysis, Detection and tracking algorithms, Image processing, Sensors, Scanners, Tumors

Read Abstract +

Proceedings Article | 27 March 2009 Paper

Non-rigid registration of small animal skeletons from micro-CT using 3D shape context

Di Xiao, Pierrick Bourgeat, Jurgen Fripp, Oscar Acosta Tamayo, Marie Claude Gregoire, Olivier Salvado

Proceedings Volume 7259, 72591G (2009) https://doi.org/10.1117/12.812461

KEYWORDS: Image registration, Computed tomography, Positron emission tomography, X-ray computed tomography, 3D image processing, 3D acquisition, Image restoration, Bone, Error analysis, Image analysis

Read Abstract +

Proceedings Article | 14 March 2009 Paper

Method for optic nerve intensity characterization in mice from Mn2+ enhanced MR images

D. Xiao, J. Ribic, H. Butzkueven, K. Fang, T. Kilpatrick, G. Egan, O. Salvado

Proceedings Volume 7258, 72584J (2009) https://doi.org/10.1117/12.813351

KEYWORDS: Optic nerve, Magnetic resonance imaging, Retina, Manganese, Visualization, Image segmentation, Signal detection, Animal model studies, Interference (communication), 3D modeling

Read Abstract +

SPIE Journal Paper | 1 July 2005

Multigradient field-active contour model for multilayer boundary detection of ultrasound rectal wall image

Di Xiao, Wan Ng, Udantha Abeyratne, Charles Tsang

JEI, Vol. 14, Issue 03, 033010, (July 2005) https://doi.org/10.1117/12.10.1117/1.1992500

KEYWORDS: Ultrasonography, Detection and tracking algorithms, Rectum, Tumors, Surgery, Image processing, 3D image reconstruction, 3D image processing, Improvised explosive devices, Object recognition

Read Abstract +

Proceedings Article | 9 May 2002 Paper

Prototype of a rectal wall ultrasound image analysis system

Di Xiao, Wan Ng, Udantha Abeyratne, Charles Tsang

Proceedings Volume 4684, (2002) https://doi.org/10.1117/12.467108

KEYWORDS: 3D image processing, 3D displays, Image processing, 3D visualizations, Tumors, Ultrasonography, Image display, Detection and tracking algorithms, Java, 3D modeling

Read Abstract +

Proceedings Article | 9 May 2002 Paper

Rectal tumor boundary detection by unifying active contour model

Di Xiao, Wan Ng, Udantha Abeyratne, Charles Tsang

Proceedings Volume 4684, (2002) https://doi.org/10.1117/12.467107

KEYWORDS: Image segmentation, Expectation maximization algorithms, Tumors, Image processing algorithms and systems, Statistical modeling, 3D modeling, Statistical analysis, Ultrasonography, Data modeling, Image processing

Read Abstract +

Proceedings Article | 9 October 1998 Paper

Complementary data fusion in guidance and control of robot compliant motion

Di Xiao, Bijoy Ghosh, Ning Xi, Tzyh-Jong Tarn

Proceedings Volume 3523, (1998) https://doi.org/10.1117/12.327006

KEYWORDS: Cameras, Sensors, Visualization, Data fusion, Imaging systems, Calibration, Algorithm development, Computer programming, Image fusion, Space robots

Read Abstract +

Showing 5 of 15 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years