Oncologic therapy of laryngeal cancer may necessitate a total excision of the larynx which results in loss of voice. Voice rehabilitation can be achieved using mucosal tissue vibrations at the upper part of the esophagus which serves as substitute voice generating element (PE segment). The quality of the substitute voice is closely related to vibratory characteristics of the PE segment. By means of a high-speed camera the dynamics of the PE segment can be recorded in real-time. Using image processing the deformations of the PE segment are extracted from the image series as deforming contours. Commonly, the characterization of PE dynamics bases on the spectral analysis of the time varying contour area. However, this constitutes an integral approach which masks most of the specific dynamics of PE deformations. We present an algorithm that automatically registers one segmented contour in a frame of the video sequence to the contour in the next frame to derive discrete 2-D trajectories of PE vibrations. By concatenation of the obtained transformations this approach provides a total registration of PE segment contours. We suggest a mixed-integer programming formulation for the problem that combines an advanced outlier and deformation handling with the introduction of dummy points in regions that newly open up, and that includes normal information in the objective function to avoid unwanted deformations. Numerical experiments show that the implemented alternate convex search algorithm produces robust results which is demonstrated at the example of five high-speed recordings of laryngectomee subjects.

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.media.2007.12.001DOI Listing

Publication Analysis

Top Keywords

registration segment
8
substitute voice
8
contour frame
8
segment
5
contour
4
segment contour
4
deformations
4
contour deformations
4
deformations digital
4
digital high-speed
4

Similar Publications

Synthetic data have emerged as an attractive option for developing machine-learning methods in human neuroimaging, particularly in magnetic resonance imaging (MRI)-a modality where image contrast depends enormously on acquisition hardware and parameters. This retrospective paper reviews a family of recently proposed methods, based on synthetic data, for generalizable machine learning in brain MRI analysis. Central to this framework is the concept of domain randomization, which involves training neural networks on a vastly diverse array of synthetically generated images with random contrast properties.

View Article and Find Full Text PDF

Timelapse imaging using high-resolution peripheral quantitative computed tomography (HR-pQCT) has emerged as a noninvasive method to quantify bone (re)modelling. However, there is no consensus on how to perform the procedure. As part of the ASTEROID phase-2b multicenter trial, we used 29 same-day repeated scans from adults with osteogenesis imperfecta (OI) to identify a method that minimized measurement error.

View Article and Find Full Text PDF

Background: Most breast cancers are detected at an early stage in which case conservative surgery is indicated. An accurate preoperative localization technique is essential for conservative surgery of non-palpable breast lesions. Currently, the gold standard technique is wire localization (WL).

View Article and Find Full Text PDF

Purpose: To evaluate the impact of Implantable Collamer Lens (ICL) implantation on anterior chamber angle parameters and posterior segment structures in highly myopic eyes and explore potential correlations between these changes. The study aimed to assess alterations in superficial and deep vessel density (SVD, DVD), foveal avascular zone (FAZ) area, and retinal nerve fiber layer (RNFL) thickness to clarify the safety profile of ICL implantation.

Methods: Prospective observational study, included 36 highly myopic eyes undergoing ICL implantation in surgery group and 23 non-surgical control eyes in non-surgery group.

View Article and Find Full Text PDF

ReIU: an efficient preliminary framework for Alzheimer patients based on multi-model data.

Front Public Health

January 2025

Engineering Research Center of Photoelectric Detection and Perception Technology, Yunnan Normal University, Kunming, China.

The rising incidence of Alzheimer's disease (AD) poses significant challenges to traditional diagnostic methods, which primarily rely on neuropsychological assessments and brain MRIs. The advent of deep learning in medical diagnosis opens new possibilities for early AD detection. In this study, we introduce retinal vessel segmentation methods based on U-Net ad iterative registration Learning (ReIU), which extract retinal vessel maps from OCT angiography (OCT-A) facilities.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!