Output details
11 - Computer Science and Informatics
Manchester Metropolitan University
Automatically identifying and annotating mouse embryo gene expression patterns
<28>Work (funded by EU ADMIRE project, FP7 ICT 215024) describes first attempt to automatically annotate Mouse embryo gene expression patterns (a significantly challenging example of multi-cellular organisation). Goes beyond existing work that focuses on relatively simple organisms (e.g., fruit fly). Experimental results with real mouse embryo images show that our method has better classification performance than other known algorithms. Investigated images originally misclassified by our method, and found several cases where the original annotation was not correct. This shows our method is robust against this kind of noise. Work provided the foundation for our BBSRC AGILE project (BB/K004077/1).