Jove
Visualize
Contact Us

Related Experiment Videos

An algorithm for fast adaptive image binarization with applications in radiotherapy imaging.

Torbjørn Sund1, Karsten Eilertsen

  • 1Telenor Research and Development, B2F, N-1331 Fornebu, Norway. torbjorn.sund@telenor.com

IEEE Transactions on Medical Imaging
|April 22, 2003
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Erroneous age assessment of young asylum seekers.

Tidsskrift for den Norske laegeforening : tidsskrift for praktisk medicin, ny raekke·2022
Same author

Favorable outcomes in locally advanced and node positive prostate cancer patients treated with combined pelvic IMRT and androgen deprivation therapy.

Radiation oncology (London, England)·2015
Same author

Dose painting by numbers in a standard treatment planning system using inverted dose prescription maps.

Acta oncologica (Stockholm, Sweden)·2015
Same author

A gel tumour phantom for assessment of the accuracy of manual and automatic delineation of gross tumour volume from FDG-PET/CT.

Acta oncologica (Stockholm, Sweden)·2012
Same author

Adaptive radiotherapy based on contrast enhanced cone beam CT imaging.

Acta oncologica (Stockholm, Sweden)·2010
Same author

The performance of multileaf collimators evaluated by the stripe test.

Medical dosimetry : official journal of the American Association of Medical Dosimetrists·2009
Same journal

PIPA: Prior-Driven Prompting with Diagnosis-Oriented Retrieval-Augmentation for 3D Radiology Report Generation.

IEEE transactions on medical imaging·2026
Same journal

DiffGeo-AOR: Diffusion-Optimized Medical Grading via Geometric Priors enhanced Autoregressive Ordinal Regression.

IEEE transactions on medical imaging·2026
Same journal

UniOCTSeg++: Refined Hierarchical Prompt Strategy and Bi-directional Progressive Consistency Learning for Universal Retinal Layer Segmentation in OCT.

IEEE transactions on medical imaging·2026
Same journal

Volumetric Functional Ultrasound Imaging in Macaques.

IEEE transactions on medical imaging·2026
Same journal

MUST: Multi-style virtual staining with incomplete pairs.

IEEE transactions on medical imaging·2026
Same journal

BrainCL: Transformer-Based Brain Network Contrastive Learning with Multi-Order Topology and Salience Masking.

IEEE transactions on medical imaging·2026
See all related articles
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

This article introduces a new, high-speed computer method for converting complex medical images into simpler black-and-white formats. By improving how the computer processes local image data, the authors enable faster and more accurate identification of bone structures in radiotherapy scans. This advancement helps clinicians perform routine image analysis more efficiently without sacrificing quality.

Area of Science:

  • Radiotherapy imaging outcomes research within medical physics
  • Computational image binarization techniques for diagnostic analysis

Background:

No prior work had resolved the computational bottleneck associated with standard sliding-window image processing in clinical settings. Existing approaches often struggle to balance processing speed with the high precision required for medical diagnostics. That uncertainty drove the need for a more efficient way to handle local thresholding tasks. Prior research has shown that adaptive methods are effective for detecting specific anatomical landmarks in portal imaging. However, these traditional routines frequently demand excessive time for routine clinical workflows. This gap motivated the development of a faster, more reliable alternative for real-time image analysis. Researchers have long sought to optimize these calculations without losing the fine details necessary for accurate radiotherapy planning. The current study addresses these limitations by proposing a novel criterion for incremental updates during image scanning.

Purpose Of The Study:

The aim of this study is to develop a faster, more efficient algorithm for locally adaptive image binarization. This research addresses the significant time constraints associated with standard sliding-window processing in clinical environments. The authors seek to create a new thresholding criterion that supports incremental updates to improve computational performance. A primary motivation is to enhance the detection of bone ridges within radiotherapy portal images. These images are often difficult to process due to their complex nature and low contrast. The researchers intend to provide a tool that is suitable for routine use in medical imaging workflows. They aim to demonstrate that their method outperforms existing public adaptive thresholding routines in both speed and accuracy. This work addresses the urgent need for rapid image analysis techniques that do not sacrifice diagnostic quality.

Keywords:
medical image processingthresholding algorithmsbone ridge detectioncomputational efficiency

Frequently Asked Questions

The researchers propose an incremental update mechanism within a sliding window. This approach avoids the high computational cost of recalculating thresholds at every pixel position, which is a common limitation in standard adaptive binarization routines.

The authors utilize a specific thresholding criterion designed for rapid, incremental updates. This mathematical tool allows the system to adjust to local image intensity variations efficiently, unlike the standard Otsu algorithm which often relies on interpolation between fixed tiles.

A small window size is necessary to achieve superior speed performance. The authors demonstrate that when using these smaller dimensions, their routine executes faster than the adaptive Otsu implementation while maintaining equivalent image quality for bone ridge detection.

Related Experiment Videos

Main Methods:

The review approach focuses on the development and validation of a novel thresholding criterion for local image analysis. Researchers designed an algorithm capable of performing incremental updates during the sliding-window process. This methodology avoids the redundant calculations typical of conventional adaptive thresholding techniques. The team evaluated their routine using complex radiotherapy portal images to assess performance in real-world clinical scenarios. They compared the efficiency and accuracy of their method against several publicly available adaptive thresholding tools. The study also benchmarked the new approach against an adaptive implementation of the Otsu algorithm. Investigators utilized interpolation between fixed tiles as a baseline for comparing image quality and processing speed. The experimental design prioritized achieving high-speed execution for small window sizes without compromising the clarity of the resulting images.

Main Results:

The strongest finding is that the new algorithm provides superior detection of bone ridges in difficult portal images compared to existing public routines. The researchers report that their method is faster than the adaptive Otsu implementation when using small window sizes. The study confirms that the resulting images are of equal quality to those produced by traditional interpolation techniques. This performance gain is achieved through a novel thresholding criterion that allows for efficient incremental updates. The data indicate that the routine successfully handles the computational demands of sliding-window processing. By reducing the time required for these tasks, the algorithm becomes practical for routine clinical use. The authors provide evidence that their approach maintains high fidelity in image output while significantly accelerating the processing pipeline. These results demonstrate a clear improvement over standard methods currently available for adaptive thresholding in radiotherapy.

Conclusions:

The authors demonstrate that their proposed thresholding criterion significantly enhances the efficiency of local image processing. Their approach provides superior detection of anatomical features in challenging portal images compared to existing public routines. The researchers report that their method maintains high image quality while drastically reducing the time required for computation. For smaller window sizes, the new algorithm outperforms the adaptive Otsu method in terms of speed. The authors state that the resulting image outputs are comparable in quality to those produced by traditional interpolation techniques. This work provides a practical solution for integrating advanced image binarization into routine radiotherapy clinical practice. The findings suggest that incremental updates are a viable strategy for overcoming standard computational delays in medical imaging. These results offer a robust framework for future developments in rapid, locally adaptive image analysis.

The authors use radiotherapy portal images to validate their method. This data type is critical for testing the algorithm's ability to identify bone ridges, which are often difficult to distinguish in noisy, low-contrast clinical scans.

The researchers measure the speed and accuracy of bone ridge detection. They compare their results against publicly available adaptive thresholding routines, showing that their method provides better detection performance on difficult images while remaining faster for small window configurations.

The authors claim that their algorithm is suitable for routine clinical use. They suggest that by overcoming the time-consuming nature of previous methods, their approach facilitates the practical application of adaptive binarization in radiotherapy imaging workflows.