The Annotated Quantitative Phase Microscopy Cell Dataset of Various Adherent Cell Lines for Segmentation Purposes comprising manually labelled images suitable for the task of robust adherent cell segmentation for quantitative phase microscopy images. Alongside this labeled dataset, they provided unlabelled data for self-supervised pretraining, contributing to the broader research community. In the authors’ paper, a novel U-Net-based method was designed and optimized. They meticulously crafted and assessed four specific post-processing pipelines. To enhance the method’s applicability to diverse cell types, the authors employed non-deep learning transfer with adjustable parameters in the post-processing phase.

Quantitative phase imaging (QPI) stands as a potent tool for label-free live cell microscopy, offering images with exceptional properties for automated image processing. The authors noted that various QPI techniques have evolved over time, enabling the time-lapse observation of subtle changes in quantitative phase dynamics within cells. These dynamic changes in cell properties, as measured by QPI, have been shown to be indicative of specific cell behaviors and find applications in diverse areas such as cell motility assessment, analysis of cell content homogeneity, and evaluation of cell mass distribution. Notably, these phase-related changes can be observed without the need for cell fixation, labeling, or harvesting, preserving the inherent characteristics of the cells under study.

The authors utilized a diverse set of adherent cell lines for their research, encompassing various origins, tumorigenic potential, and morphologies, including PC-3, PNT1A, 22Rv1, DU145, LNCaP, A2058, A2780, Fadu, G361, and HOB. These cell lines were cultured in specific media and conditions tailored to their requirements, and the authors maintained them under controlled environmental conditions for microscopy acquisition. The QPI data was acquired using a coherence-controlled holographic microscope, with images taken at different time intervals to capture cells at varying stages of morphological changes. Please note, that cells 22Rv1, A2058, A2780, DU145, Fadu, G361, HOB and LNCaP are in the unlabelled part of the dataset

PC-3, PNT1A, 22Rv1, DU145, LNCaP, A2780, and G361 cell lines were cultured in RPMI-1640 medium, A2058, FaDu, and HOB cell lines were cultured in DMEM-F12 medium, all supplemented with antibiotics (penicillin 100 U/ml and streptomycin 0.1 mg/ml), and with 10% fetal bovine serum (FBS). Prior to microscopy acquisition, the cells were maintained at 37°C in a humidified (60%) incubator with 5% CO² (Sanyo, Japan). For acquisition purposes, the cells were cultivated in the Flow chamber µ-Slide I Luer Family (Ibidi, Martinsried, Germany). To maintain standard cultivation conditions during time-lapse experiments, cells were placed in the gas chamber H201 - for Mad City Labs Z100/Z500 piezo Z-stage (Okolab, Ottaviano NA, Italy). For the acquisition of QPI, a coherence-controlled holographic microscope (Telight, Q-Phase) was used. Objective Nikon Plan 10×/0.3 was used for hologram acquisition with a CCD camera (XIMEA MR4021MC). Holographic data were numerically reconstructed with the Fourier transform method. QPI datasets used in this paper were acquired during various experimental setups and treatments. In most cases, experiments were conducted with the time-lapse acquisition. The final dataset contains images acquired at least three hours apart.