Low tensor train and low multilinear rank approximations of 3D tensors for compression and de-speckling of optical coherence tomography Images

Kopriva, Ivica; Shi, Fei; Lai, Mingyig; Štanfel, Marija; Chen, Haoyu; Chen, Xinjian (2023) Low tensor train and low multilinear rank approximations of 3D tensors for compression and de-speckling of optical coherence tomography Images. Physics in medicine and biology, 68 . ISSN 0031-9155

PDF - Accepted Version - article
Restricted to Registered users only until 16 June 2024.
Available under License Creative Commons Attribution Non-commercial No Derivatives.
Download (4MB) | Request a personal copy from author

Official URL: https://iopscience.iop.org/article/10.1088/1361-65...

Abstract

Objective. Many methods for compression and/or de-speckling of 3D optical coherence tomography (OCT) images operate on a slice-by-slice basis and, consequently, ignore spatial relations between the B-scans. Thus, we develop compression ratio (CR)-constrained low tensor train (TT) - and low multilinear (ML) rank approximations of 3D tensors for compression and de-speckling of 3D OCT images. Due to inherent denoising mechanism of low-rank approximation, compressed image is often even of better quality than the raw image it is based on. Approach. We formulate CR-constrained low rank approximations of 3D tensor as parallel non-convex non-smooth optimization problems implemented by alternating direction method of multipliers of unfolded tensors. In contrast to patch- and sparsity-based OCT image compression methods, proposed approach does not require clean images for dictionary learning, enables CR as high as 60:1, and it is fast. In contrast to deep networks based OCT image compression, proposed approach is training free and does not require any supervised data pre-processing. Main results. Proposed methodology is evaluated on twenty four images of a retina acquired on Topcon 3D OCT-1000 scanner, and twenty images of a retina acquired on Big Vision BV1000 3D OCT scanner. For the first dataset, statistical significance analysis shows that for CR<=35, all low ML rank approximations and Schatten-0 (S0) norm constrained low TT rank approximation can be useful for machine learning-based diagnostics by using segmented retina layers. Also for CR<=35, S0-constrained ML rank approximation and S0-constrained low TT rank approximation can be useful for visual inspection-based diagnostics. For the second dataset, statistical significance analysis shows that for CR<=60 all low ML rank approximations as well as S0 and S1/2 low TT ranks approximations can be useful for machine learning-based diagnostics by using segmented retina layers. Also, for CR<=60, low ML rank approximations constrained with Sp, p e {;0, 1/2, 2/3}; and one surrogate of S0 can be useful for visual inspection-based diagnostics. That is also true for low TT rank approximations constrained with Sp, p e {;0, 1/2, 2/3}; for CR<=20. Significance. Studies conducted on datasets acquired by two different types of scanners confirmed capabilities of proposed framework that, for a wide range of CRs, yields de-speckled 3D OCT images suitable for clinical data archiving and remote consultation, for visual inspection-based diagnosis and for machine learning-based diagnosis by using segmented retina layers.

Item Type:

Article

Uncontrolled Keywords:

3D optical coherence tomography ; compression ; de-speckling ; tensor train rank ; multilinear rank ; Schatten p-norm ; surrogate of Schatten-0 norm

Subjects:

BIOMEDICINE AND HEALTHCARE > Clinical Medical Sciences

Divisions:

Division of Electronics

Projects:

Project title	Project leader	Project code	Project type
Strukturne dekompozicije empirijskih podataka za računalno potpomognutu dijagnostiku bolesti	Kopriva, Ivica	IP-2016-06-5235	HRZZ
Napredne metode i tehnologije u znanosti o podatcima i kooperativnim sustavima	Lončarić, Sven	KK.01.1.1.01.0009	EK
8th Chinese-Croatian Inter-governmental S&T Cooperation Project	Kopriva, Ivica	UNSPECIFIED	Bilateralni