Ionizing radiation produces characteristic chromosome changes. The altered chromosomes contain two central constrictions, termed centromeres, instead of one (known as dicentric chromosomes [DCs]). Chromosome biodosimetry studies are approved by the IAEA for occupational radiation exposure, radiation emergencies, or monitoring long term exposures. In emergency responses to a range of doses, labs need efficient methods that identify DCs.
Cytognomix has developed a novel approach to find DCs that is independent of chromosome length, shape and structure from different laboratories (paper: TBME). The Automated Dicentric Chromosome Identifier and Dose Estimator (ADCI) software works on multiple platforms and uses images produced by any of the existing automated metaphase capture systems found in most cytogenetic laboratories.
ADCI is now available for purchase. Contact us for details (pricing, online user manual: wiki). Ask about discount for early adopters!
ADCI uses machine learning based algorithms with high sensitivity and specificity that distinguish monocentric and dicentric chromosomes. Together with our novel segmentation approaches, this has been translated into a fully functional cytogenetic biodosimetry system. The ADCI* takes images from all types of commercial metaphase scanning systems, selects high quality cells for analysis, identifies dicentric chromosomes (removing false positives), builds biodosimetry calibration curves, and estimates exposures. ADCI fulfills the criteria established by the IAEA for accurate triage biodosimetry of a sample in less than an hour. The accuracy is comparable to an experienced cytogeneticist.
The early development of ADCI was supported by the US Public Health Service 5U01AI091173-02. Working with IBM Canada and Western University, we have developed a high performance computing version, ADCI-BG/Q has demonstrated adequate throughput to handle the workload from a simulated mass casualty event (1000 samples in 1.4 hr; Li et al. IEEE Intnl Conf on Automation for Sustainability, 2012. Pp. 30 – 35 DOI 10.1109/ICIAFS.2012.6420039 (paper: ICIAFS); and paper: Rogan et al. Radiat Prot Dosimetry, 2014).
Presentations and articles about ADCI:
Yanxin Li, Joan H. Knoll, Ruth Wilkins, Farrah Flegal, and Peter K. Rogan. Automated Discrimination of Dicentric and Monocentric Chromosomes by Machine Learning-based Image Processing. EPR Biodosimetry International conference. Dartmouth College Oct 4-8, 2015. Now published in Microscopy Research and Technique (link). Manuscript (pdf): Li et al 2016. Companion on-line software, the Dicentric Chromosome Analyzer is at: cytobiodose.cytognomix.com
Peter K Rogan, Yanxin Li, Ruth Wilkins, Farrah Flegal, Joan H. Knoll. Radiation dose estimation by automated chromosome biodosimetry. EPR Biodosimetry International conference. Dartmouth College Oct 4-8, 2015. Now published in Radiation Protection Dosimetry (link). Manuscript (pdf): Rogan et al. 2016.
*ADCI is covered by US Patent 8,605,981, German Patent 11 2011103687, and others pending.
- February 27, 2017. CytoGnomix finalizes contract with Government of Canada
- Jan. 28, 2017. New version of F1000Research paper on chemotherapy response in breast cancer
- January 25, 2017. Comment from the Transforming Genetic Medicine Initiative Blog
- Jan. 23, 2017. Automated interpretation of digital pathology images is currently at an embryonic stage of development