Open Source Infrastructure for Automatic Cell Segmentation

TL;DR

Open-source infrastructure is presented, utilizing the UNet model, a deep-learning architecture noted for its effectiveness in image segmentation tasks, integrated into the open-source DeepChem package, enhancing accessibility and usability for researchers and practitioners.

Abstract

Automated cell segmentation is crucial for various biological and medical applications, facilitating tasks like cell counting, morphology analysis, and drug discovery. However, manual segmentation is time-consuming and prone to subjectivity, necessitating robust automated methods. This paper presents open-source infrastructure, utilizing the UNet model, a deep-learning architecture noted for its effectiveness in image segmentation tasks. This implementation is integrated into the open-source DeepChem package, enhancing accessibility and usability for researchers and practitioners. The resulting tool offers a convenient and user-friendly interface, reducing the barrier to entry for cell segmentation while maintaining high accuracy. Additionally, we benchmark this model against various datasets, demonstrating its robustness and versatility across different imaging conditions and cell types.

Figures (4)

• Figure 1: Block diagram of the UNet model architecture. The numbers show the number of channels in the image, we can see that our input and output are both 3-channel images.
• Figure 2: Overview of the cell segmentation pipeline using DeepChem. The pipeline includes data loading, pre-processing, model training, evaluation, and inference.
• Figure 3: This is the DeepChem implementation of the Image Segmentation Pipeline as seen in Fig. 2.
• Figure 4: The above image compares the UNet model's predictions with the true segmentation mask. We've compared 1 random sample from the Fluo-N2DH-GOWT1, Fluo-C2DL-MSC and PhC-C2DL-PSC datasets each.