Related Experiment Video
Updated: May 8, 2026
Transcutaneous Microcirculatory Imaging in Preterm Neonates
Published on: December 31, 2015
The microcirculation image quality score: development and preliminary evaluation of a proposed approach to grading
Michael J Massey1, Ethan Larochelle, Gabriel Najarro
1Department of Emergency Medicine, Beth Israel Deaconess Medical Center, Boston, MA.
Purpose:
Side-stream dark-field microscopy is currently used to directly visualize sublingual microcirculation at the bedside. Our experience has found inherent technical challenges in the image acquisition process. This article presents and assesses a quality assurance method to rate image acquisition quality before analysis.
Materials And Methods:
We identified 6 common image capture and analysis problem areas in sublingual side-stream dark-field videos: illumination, duration, focus, content, stability, and pressure. We created the "Microcirculation Image Quality Score" by assigning a score of optimal (0 points), suboptimal but acceptable (1 point), or unacceptable (10 points) to each category (for further details, go to http://www.MicroscanAnalysis.blogspot.com). We evaluated 59 videos from a convenience sample of 34 unselected, noncritically ill emergency department patients to create a test set. Two raters, blinded to each other, implemented the score. Any video with a cumulative score of 10 or higher (range, 0-60) was considered unacceptable for further analysis.
Results:
We created the Microcirculation Image Quality Score and applied it to 59 videos. For this particular set of 59 videos, the mean (SD) passing quality score was 1.68 (0.90), and the mean (SD) failing quality score was 15.74 (6.19), with 27 of 59 passing the quality score less than 10. Highest failure occurred from pressure artifact. The interrater agreement for acceptability was assessed using Cohen κ for each category: illumination (κ = 1.0), duration (κ = 1.0), focus (κ = 0.91), content (κ = 0.76), stability (κ = 0.71), and pressure (κ = 0.82) and overall pass-fail rates (score >10) (κ = 0.66).
Conclusion:
Our Microcirculation Image Quality Score addresses many of the common areas where video quality can degrade. The criteria introduced are an objective way to assess the quality of image acquisition, with the goal of selecting videos of adequate quality for analysis. The interrater reliability results in our preliminary study suggest that the Microcirculation Image Quality Score is reasonably repeatable between reviewers. Further assessment is warranted.
Insights
A new scoring system, the Microcirculation Image Quality Score, objectively assesses bedside sublingual microcirculation videos. This method ensures adequate image quality for analysis, improving diagnostic reliability.
Area of Science:
- Medical Imaging
- Microcirculation Research
- Point-of-Care Diagnostics
Background:
- Side-stream dark-field microscopy enables direct bedside visualization of sublingual microcirculation.
- Technical challenges in image acquisition can compromise the quality and reliability of these videos.
- Standardized quality assessment is needed to ensure accurate analysis of microcirculation data.
Purpose of the Study:
- To present and assess a novel quality assurance method for rating image acquisition quality in sublingual side-stream dark-field microscopy videos.
- To establish objective criteria for evaluating video quality before performing microcirculation analysis.
- To improve the consistency and validity of bedside microcirculation imaging.
Main Methods:
- A "Microcirculation Image Quality Score" was developed, evaluating 6 key areas: illumination, duration, focus, content, stability, and pressure.
- Each category was scored as optimal (0 points), acceptable (1 point), or unacceptable (10 points).
- 59 videos from 34 emergency department patients were assessed by two blinded raters; a cumulative score of 10 or higher indicated unacceptable quality.
Main Results:
- The Microcirculation Image Quality Score was applied to 59 videos, with mean passing and failing scores of 1.68 and 15.74, respectively.
- 27 of 59 videos met the quality criteria (score < 10), with pressure artifact being the most common failure point.
- High interrater agreement (Cohen κ ranging from 0.66 to 1.0) was observed across all categories and overall pass-fail rates.
Conclusions:
- The developed Microcirculation Image Quality Score effectively addresses common video quality degradation issues.
- The criteria provide an objective method for assessing image acquisition quality, ensuring adequate video selection for analysis.
- Preliminary results indicate good interrater reliability, suggesting the score is repeatable and warrants further validation.
