Evaluation of Deep Learning architectures for complex immunofluorescence nuclear image segmentation

F. Kromp, L. Fischer, E. Bozsaky, I. Ambros, W. Dörr, K. Beiske, P. Ambros, A. Hanbury, S. Taschner-Mandl. Evaluation of Deep Learning architectures for complex immunofluorescence nuclear image segmentation. IEEE Transactions on Medical Imaging, DOI 10.1109/TMI.2021.3069558, 3, 2021.

Autoren
  • Florian Kromp
  • Lukas Fischer
  • Eva Bozsaky
  • Inge M. Ambros
  • Wolfgang Dörr
  • Klaus Beiske
  • Peter F. Ambros
  • Allan Hanbury
  • Sabine Taschner-Mandl
TypArtikel
JournalIEEE Transactions on Medical Imaging
DOI10.1109/TMI.2021.3069558
Monat3
Jahr2021
Abstract

Separating and labeling each nuclear instance (instance-aware segmentation) is the key challenge in nuclear image segmentation. Deep Convolutional Neural Networks have been demonstrated to solve nuclear image segmentation tasks across different imaging modalities, but a systematic comparison on complex immunofluorescence images has not been performed. Deep learning based segmentation requires annotated datasets for training, but annotated fluorescence nuclear image datasets are rare and of limited size and complexity. In this work, we evaluate and compare the segmentation effectiveness of multiple deep learning architectures (U-Net, U-Net ResNet, Cellpose, Mask R-CNN, KG instance segmentation) and two conventional algorithms (Iterative h-min based watershed, Attributed relational graphs) on complex fluorescence nuclear images of various types. We propose and evaluate a novel strategy to create artificial images to extend the training set. Results show that instance-aware segmentation architectures and Cellpose outperform the U-Net architectures and conventional methods on complex images in terms of F1 scores, while the U-Net architectures achieve overall higher mean Dice scores. Training with additional artificially generated images improves recall and F1 scores for complex images, thereby leading to top F1 scores for three out of five sample preparation types. Mask R-CNN trained on artificial images achieves the overall highest F1 score on complex images of similar conditions to the training set images while Cellpose achieves the overall highest F1 score on complex images of new imaging conditions. We provide quantitative results demonstrating that images annotated by under-graduates are sufficient for training instance-aware segmentation architectures to efficiently segment complex fluorescence nuclear images.