eduzhai > Applied Sciences > Engineering >

A Preliminary Exploration into an Alternative CellLineNet An Evolutionary Approach

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 15 pages

Abstract: Within this paper, the exploration of an evolutionary approach to analternative CellLineNet: a convolutional neural network adept at theclassification of epithelial breast cancer cell lines, is presented. Thisevolutionary algorithm introduces control variables that guide the search ofarchitectures in the search space of inverted residual blocks, bottleneckblocks, residual blocks and a basic 2x2 convolutional block. The promise ofEvoCELL is predicting what combination or arrangement of the feature extractingblocks that produce the best model architecture for a given task. Therein, theperformance of how the fittest model evolved after each generation is shown.The final evolved model CellLineNet V2 classifies 5 types of epithelial breastcell lines consisting of two human cancer lines, 2 normal immortalized lines,and 1 immortalized mouse line (MDA-MB-468, MCF7, 10A, 12A and HC11). TheMulticlass Cell Line Classification Convolutional Neural Network extends ourearlier work on a Binary Breast Cancer Cell Line Classification model. Thispaper presents an on-going exploratory approach to neural network architecturedesign and is presented for further study.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×