30 August 2018 in Scientific Reports
MKMatthijs van Kranenburg
+1
The most widespread technique used to register sets of medical images consists of selecting one image as fixed reference, to which all remaining images are successively registered. This pairwise scheme requires one optimization procedure per pair of images to register. Pairwise mutual information is a common dissimilarity measure applied to a large variety of datasets. Alternative methods, called groupwise registrations, have been presented to register two or more images in a single optimization procedure, without the need of a reference image. Given the success of mutual information in pairwise registration, we adapt one of its multivariate versions, called total correlation, in a groupwise context. We justify the choice of total correlation among other multivariate versions of mutual information, and provide full implementation details. The resulting total correlation measure is remarkably close to measures previously proposed by Huizinga et al. based on principal component analysis. Our experiments, performed on five quantitative imaging datasets and on a dynamic CT imaging dataset, show that total correlation yields registration results that are comparable to Huizinga’s methods. Total correlation has the advantage of being theoretically justified, while the measures of Huizinga et al. were designed empirically. Additionally, total correlation offers an alternative to pairwise mutual information on quantitative imaging datasets.
1 October 2021 in International Journal of Oral and Maxillofacial Surgery
The aim of this study was to evaluate and present an automated method for registration of magnetic resonance imaging (MRI) and computed tomography (CT) or cone beam CT (CBCT) images of the mandibular region for patients with oral squamous cell carcinoma (OSCC). Registered MRI and (CB)CT could facilitate the three-dimensional virtual planning of surgical guides employed for resection and reconstruction in patients with OSCC with mandibular invasion. MRI and (CB)CT images were collected retrospectively from 19 patients. MRI images were aligned with (CB)CT images employing a rigid registration approach (stage 1), a rigid registration approach using a mandibular mask (stage 2), and two non-rigid registration approaches (stage 3). Registration accuracy was quantified by the mean target registration error (mTRE), calculated over a set of landmarks annotated by two observers. Stage 2 achieved the best registration result, with an mTRE of 2.5±0.7mm, which was comparable to the inter- and intra-observer variabilities of landmark placement in MRI. Stage 2 was significantly better aligned compared to all approaches in stage 3. In conclusion, this study demonstrated that rigid registration with the use of a mask is an appropriate image registration method for aligning MRI and (CB)CT images of the mandibular region in patients with OSCC.
6 June 2018 in IEEE Journal of Biomedical and Health Informatics
Multi-channel image registration is an important challenge in medical image analysis. Multi-channel images result from modalities such as dual-energy CT or multispectral microscopy. Besides, multi-channel feature images can be derived from acquired images, for instance, by applying multi-scale feature banks to the original images to register. Multi-channel registration techniques have been proposed, but most of them are applicable to only two multi-channel images at a time. In the present study, we propose to formulate multi-channel registration as a groupwise image registration problem. In this way, we derive a method that allows the registration of two or more multi-channel images in a fully symmetric manner (i.e. all images play the same role in the registration procedure), and therefore has transitive consistency by definition. The method that we introduce is applicable to any number of multi-channel images, any number of channels per image, and it allows to take into account correlation between any pair of images and not just corresponding channels. In addition, it is fully modular in terms of dissimilarity measure, transformation model, regularisation method, and optimisation strategy. For two multimodal datasets, we computed feature images from the initially acquired images, and applied the proposed registration technique to the newly created sets of multi-channel images. MIND descriptors were used as feature images, and we chose total correlation as groupwise dissimilarity measure. Results show that groupwise multi-channel image registration is a competitive alternative to the pairwise multi-channel scheme, in terms of registration accuracy and insensitivity towards registration reference spaces.
12 December 2025 in Applied Spectroscopy
Chromosome characterization is crucial in cytogenetic research and diagnostics, necessitating precise imaging methods to ensure proper analyses. The aim of this project is to identify a reliable method for chromosomal characterization that uses hyperspectral imagery of stained metaphase chromosomes using bright-field microscopy. We analyzed four hyperspectral images of stained chromosomes acquired under bright-field illumination. To address the high dimensionality of the hyperspectral hypercubes, we applied five dimension reduction algorithms based on spectral band selection to determine the most effective approach. A comparative study was conducted between five band selection methods to assess their effectiveness in chromosome characterization. The results indicate that sparse subspace clustering and multi-objective band selection are the most effective methods, outperforming the others in reducing the spectral dimensionality of the hyperspectral data, while preserving key properties essential for accurate chromosomes characterization. This study demonstrates that careful selection of spectral bands can enhance the analysis of spectral hypercubes for chromosome characterization.
3 October 2026 in SSRN Electronic Journal
Accurate segmentation of overlapped chromosomes in hyperspectral imageryremains challenging due to complex spatial interactions and spectral redun-dancy. In this work, we propose a multimodal framework for segmentationthat integrates phasor-based spectral representations with a 3D nnU-Netarchitecture. The proposed framework exploits phasor features as a com-plementary modality processed through a dedicated 2D encoder and fusedwith volumetric hyperspectral representations within a unified 3D decodingpathway. To bridge the dimensional gap between 2D and 3D feature spaces,decoder-level fusion strategies are investigated to enable effective cross-modalinteraction. Experimental results demonstrate consistent improvements overRGB-based, hyperspectral imaging (HSI)-based, and hybrid multimodal ap-proaches. Compared with the Vanilla nnU-Net 3D, the proposed frameworkimproves the IoU and Dice scores by approximately 4 and 2 percentage points,respectively. Compared with the RGB-based baseline methods, it achievesIoU gains of up to 11 percentage points. Statistical analysis confirms thesignificance of the observed improvements (p ≪ 0.001). The results fur-ther suggest that phasor representations provide complementary informationto volumetric hyperspectral features, leading to more accurate and robustchromosome segmentation. These findings highlight the potential of phasor-guided multimodal learning for hyperspectral biomedical image analysis.