1
|
Subbakrishna Adishesha A, Vanselow DJ, La Riviere P, Cheng KC, Huang SX. Sinogram domain angular upsampling of sparse-view micro-CT with dense residual hierarchical transformer and attention-weighted loss. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2023; 242:107802. [PMID: 37738839 PMCID: PMC11158828 DOI: 10.1016/j.cmpb.2023.107802] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 06/08/2023] [Revised: 07/23/2023] [Accepted: 09/04/2023] [Indexed: 09/24/2023]
Abstract
Reduced angular sampling is a key strategy for increasing scanning efficiency of micron-scale computed tomography (micro-CT). Despite boosting throughput, this strategy introduces noise and extrapolation artifacts due to undersampling. In this work, we present a solution to this issue, by proposing a novel Dense Residual Hierarchical Transformer (DRHT) network to recover high-quality sinograms from 2×, 4× and 8× undersampled scans. DRHT is trained to utilize limited information available from sparsely angular sampled scans and once trained, it can be applied to recover higher-resolution sinograms from shorter scan sessions. Our proposed DRHT model aggregates the benefits of a hierarchical- multi-scale structure along with the combination of local and global feature extraction through dense residual convolutional blocks and non-overlapping window transformer blocks respectively. We also propose a novel noise-aware loss function named KL-L1 to improve sinogram restoration to full resolution. KL-L1, a weighted combination of pixel-level and distribution-level cost functions, leverages inconsistencies in noise distribution and uses learnable spatial weight maps to improve the training of the DRHT model. We present ablation studies and evaluations of our method against other state-of-the-art (SOTA) models over multiple datasets. Our proposed DRHT network achieves an average increase in peak signal to noise ratio (PSNR) of 17.73 dB and a structural similarity index (SSIM) of 0.161, for 8× upsampling, across the three diverse datasets, compared to their respective Bicubic interpolated versions. This novel approach can be utilized to decrease radiation exposure to patients and reduce imaging time for large-scale CT imaging projects.
Collapse
Affiliation(s)
| | - Daniel J Vanselow
- Penn State College of Medicine, The Jake Gittlen Laboratories for Cancer Research, Hershey, 17033, PA, USA; Penn State College of Medicine, Division of Experimental Pathology, Department of Pathology, Hershey, 17033, PA, USA.
| | - Patrick La Riviere
- University of Chicago, Department of Radiology, Chicago, 60637, IL, USA.
| | - Keith C Cheng
- Penn State College of Medicine, The Jake Gittlen Laboratories for Cancer Research, Hershey, 17033, PA, USA; Penn State College of Medicine, Division of Experimental Pathology, Department of Pathology, Hershey, 17033, PA, USA.
| | - Sharon X Huang
- Pennsylvania State University, College of Information Sciences and Technology, University Park, 16802, PA, USA.
| |
Collapse
|
2
|
Väänänen V, Christensen MM, Suhonen H, Jernvall J. Gene expression detection in developing mouse tissue using in situ hybridization and µCT imaging. Proc Natl Acad Sci U S A 2023; 120:e2301876120. [PMID: 37279266 PMCID: PMC10268296 DOI: 10.1073/pnas.2301876120] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/17/2023] [Accepted: 05/07/2023] [Indexed: 06/08/2023] Open
Abstract
High resolution and noninvasiveness have made soft-tissue X-ray microtomography (µCT) a widely applicable three-dimensional (3D) imaging method in studies of morphology and development. However, scarcity of molecular probes to visualize gene activity with µCT has remained a challenge. Here, we apply horseradish peroxidase-assisted reduction of silver and catalytic gold enhancement of the silver deposit to in situ hybridization in order to detect gene expression in developing tissues with µCT (here called GECT, gene expression CT). We show that GECT detects expression patterns of collagen type II alpha 1 and sonic hedgehog in developing mouse tissues comparably with an alkaline phosphatase-based detection method. After detection, expression patterns are visualized with laboratory µCT, demonstrating that GECT is compatible with varying levels of gene expression and varying sizes of expression regions. Additionally, we show that the method is compatible with prior phosphotungstic acid staining, a conventional contrast staining approach in µCT imaging of soft tissues. Overall, GECT is a method that can be integrated with existing laboratory routines to obtain spatially accurate 3D detection of gene expression.
Collapse
Affiliation(s)
- Vilma Väänänen
- Institute of Biotechnology, University of Helsinki, HelsinkiFI-00014, Finland
| | - Mona M. Christensen
- Institute of Biotechnology, University of Helsinki, HelsinkiFI-00014, Finland
| | - Heikki Suhonen
- Department of Physics, University of Helsinki, HelsinkiFI-00014, Finland
| | - Jukka Jernvall
- Institute of Biotechnology, University of Helsinki, HelsinkiFI-00014, Finland
- Department of Geosciences and Geography, University of Helsinki, HelsinkiFI-00014, Finland
| |
Collapse
|
3
|
Zebrafish Models to Study Ectopic Calcification and Calcium-Associated Pathologies. Int J Mol Sci 2023; 24:ijms24043366. [PMID: 36834795 PMCID: PMC9967340 DOI: 10.3390/ijms24043366] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/09/2022] [Revised: 01/20/2023] [Accepted: 02/01/2023] [Indexed: 02/11/2023] Open
Abstract
Ectopic calcification refers to the pathological accumulation of calcium ions in soft tissues and is often the result of a dysregulated action or disrupted function of proteins involved in extracellular matrix mineralization. While the mouse has traditionally been the go-to model organism for the study of pathologies associated with abnormal calcium deposition, many mouse mutants often have exacerbated phenotypes and die prematurely, limiting the understanding of the disease and the development of effective therapies. Since the mechanisms underlying ectopic calcification share some analogy with those of bone formation, the zebrafish (Danio rerio)-a well-established model for studying osteogenesis and mineralogenesis-has recently gained momentum as a model to study ectopic calcification disorders. In this review, we outline the mechanisms of ectopic mineralization in zebrafish, provide insights into zebrafish mutants that share phenotypic similarities with human pathological mineralization disorders, list the compounds capable of rescuing mutant phenotypes, and describe current methods to induce and characterize ectopic calcification in zebrafish.
Collapse
|