Deep learning for image analysis of breast and prostate cancer cell cultures

ORAL

Abstract

We present a machine-learning analysis of phase-contrast microscope images of breast cancer (MDA-MB-231) and prostate cancer (PC3) cell cultures. Semantic segmentation (cell detection) was performed using several variants of the U-net [1] convolutional neural network architecture. Best results were obtained using an Attention U-net [2] with a binary focal loss function. Instance segmentation (cell labeling) was performed using the watershed method. Geometrical properties of each cell (area, solidity, and eccentricity) were computed and their statistics were plotted as a function of time, in order to quantify cell growth under different conditions.

*This research was supported by the National Science Foundation under NSF EPSCoR Track-1 CooperativeAgreement OIA #1946202. This work used resources of the Center for Computationally Assisted Science and Technology (CCAST) at North Dakota State University, which were made possible in part by NSF MRI Award No. 2019077. This work used advanced cyberinfrastructure resources provided by the University of North Dakota Computational Research Center. Imaging studies were conducted in the UND Imaging Core facility supported by NIH grant P20GM113123, DaCCoTA CTR NIH grant U54GM128729, and UNDSMHS funds.

–

Publication: 1. O. Ronneberger, P. Fischer, and T. Brox, U-Net: Convolutional Networks for Biomedical Image Segmentation, in MICCAI 2015, pp. 234–241 (2015)
2. O. Oktay et al., Attention U-Net: Learning Where to Look for the Pancreas, ArXiv 1804:03999 (2018)

Presenters

  • Aliakbar Sepehri

    • University of North Dakota

Authors

  • Aliakbar Sepehri

    • University of North Dakota

  • Ian Bergerson

    • University of North Dakota

  • Yen Lee Loh

    • University of North Dakota

  • Lucas Bierscheid

    • North dakota university state

  • John Wilkinson

    • North Dakota State University