We introduce a methodology for assessment of microscopy settings and for predicting biological cell segmentation accuracy based on image quality descriptors. Figure 1 illustrates the relationship between data quality and microscopy instrument settings as well as segmentation accuracy.
High Content Screening (HCS) is an automated microscopy technique enabling the evaluation of spatial and temporal effects on cells for drug discovery and other applications. With advancements in hardware and software, the throughput of HCS can achieve hundreds of cell images per second that have to be acquired and processed to capture transient morphological effects. Higher throughput can impact image quality by, for instance, reducing exposure times, and can affect accuracy of image post-processing, for example, segmentation accuracy. The motivation of our work is to understand the relationship triangle between image quality, microscopy settings and accuracy of image post-processing.
To have confidence in the knowledge gleaned from data, it is essential to verify the quality of the data sources and to have a means of quantifying the potential uncertainties due to the quality of data for making critical, intelligent decisions. In computational biology, two areas where data quality is significant are quality of measured images and quality of the corresponding reference data (e.g., manual segmentations or cell colony labels).
The objective of the data quality component is to collect and develop a repository of image quality descriptors and analyze its sensitivity in improving data quality upstream (microscope conditions, sample preparation) and downstream (recommending segmentation methods, predicting segmentation accuracy).
Developing automated methods to assess image quality improves quality of the biological analysis. Automated image analysis ensures objectivity of the results. However, the quality of the image directly affects the accuracy of the image analysis (segmentation) and has been proven to impact the accuracy of the research findings such as drug effectiveness and optimal dosage. This is true especially in HCS applications which enable the evaluation of spatial and temporal effects on cells for drug discovery using an automated microscopy.
In addition to the quality of measured images, inconsistent reference labels that correspond to the measured images can degrade the ability to derive biologically meaningful classes or clusters. Inconsistent labels imply that the experts did not agree when determining the reference label, and hence, it is difficult to create rules for classification or clustering. The project aims at understanding the impact of reference data quality on clustering and classification uncertainty.
The paper: Ya-Shian Li-Baboud, Antonio Cardone, Joe Chalfoun, Peter Bajcsy, John Elliot, "Understanding the impact of image quality on segmentation accuracy," SPIE Newsroom, 6 August 2013, SPIE Newsroom. DOI: 10.1117/2.1201307.004996, https://spie.org/x102666.xml?pf=true&ArticleID=x102666 (download pdf)
Ya-Shian Li-Baboud, Antonio Cardone, Joe Chalfoun, Peter Bajcsy, John Elliot, "Understanding the impact of image quality on segmentation accuracy," SPIE Newsroom, 6 August 2013, SPIE Newsroom. DOI: 10.1117/2.1201307.004996, https://spie.org/x102666.xml?pf=true&ArticleID=x102666 (download pdf)