Abstract:Radiotherapy and chemotherapy are essential for preoperative and postoperative treatment of cancer patients. Chemotherapy drugs destroy cancer cells and inhibit their proliferation mainly by promoting cancer cell apoptosis. The efficacy of anticancer drugs is measured by their ability to recognize cancer cells and selectively promote their apoptosis. The Drug Sensitivity Test (DST) is a method to determine the most effective drug for tumor treatment according to the sensitivity. Tumors may be resistant to one or more drugs, or show sensitivity to multiple drugs for their different genotype and pathogenesis. Therefore, the detection of drug-induced apoptosis in anticancer drug sensitivity test is of great significance for reducing drug resistance, improving the efficiency of drug sensitivity test and achieving more effective personalized treatment. At present, the main methods to detect apoptosis are to detect the changes of cell morphology and surface markers related to apoptosis. However, the commonly used methods like flow cytometry, membrane protein, TUNEL analysis, have poor specificity in the detection of cell apoptosis. The typical morphological changes exhibited in the process of apoptosis which have become a reliable basis for the identification of apoptosis. Digital holographic microscopy provides a non-invasive quantitative phase imaging method for living cells. It can meet the requirements of label-free, long-term imaging, and evaluation of cell morphological and kinetic parameters under different treatments. In this paper, digital holographic microscopy is used to record images of label-free cancer cells during apoptosis process after adding drugs. Firstly, a Mach-Zehnder digital holographic microscopic system with an off-axis configuration is used to capture the wave-field of cancer cells. This system can realize complex object wavefront reconstruction with a single camera exposure. In the process of cell imaging, the cells adhere to the bottom of the cell culture dish, and the culture dish is filled with cell culture solution to ensure the normal growth of cells. The camera records a hologram every 1min for a total of 9 hours after adding drugs into the cell culture solution. Then, the phase images of cancer cells are numerically reconstructed. Two pre-processing operations are implemented, consisting of hologram apodization and spatial filtering and then the angle spectrum reconstruction algorithm is employed to implement the numerical propagation, keeping the object image size constant whatever the propagation distance. To obtain an in-focus and sharp object image, an optimal propagation distance needs to be found by automatic focus method. Besides, Numerical Parametric Lenses (NPL) method is employed to compensate the phase aberrations in the phase image. Due to the phase value reconstructed from the hologram constrained between- pi and pi, the continuous phase map of the object can be retrieved by phase unwrapping. From the reconstructed phase image of cells after 9 hours of drug treatment, it can be clearly seen that most of the cells have broken and died, while other cells that have not broken have also shrunk significantly, and the cell height has increased significantly. Furthermore, we select 8 cells from the dead cells for further analysis of their complete death process. Single-cell phase images are segmented from the phase images. And, the morphological change of cell apoptosis process is characterized morphologically by the average phase shift and dry mass. It can be seen from the change carves of these two parameters during the cell apoptosis, there is no obvious change of cell dry mass before apoptosis. But at the moment of the cells lose membrane integrity and release their intracellular contents, cells' dry mass decreased sharply. At the meantime, the average phase shift continuously increases before the cells lose membrane integrity, indicating that the cells have contracted, which is consistent with the conclusion in previous study. These results show that there are significant differences in phase images and morphological parameters between growing cancer cells and apoptotic and dead cells. Therefore, the method in this paper can distinguish apoptotic cells and dead cells without fluorescent labeling. And it can provide a more economical and convenient detection method for determining and selecting the most effective chemotherapeutic drugs and determining their effective dose for in vitro drug sensitivity test in individualized treatment.

AI-based Apoptosis Cell Classification Using Phase-contrast Images of K562 Cells

Apoptosis detection for adherent cell populations in time-lapse phase-contrast microscopy images.

Abstract 356: Developing a multiplex assay to monitor apoptosis and necrosis using the Cellaca® PLX Image Cytometer

A Multiplex Assay to Simultaneously Monitor Apoptosis and Necrosis Using the Cellaca® PLX Image Cytometer

A guide to automated apoptosis detection: How to make sense of imaging flow cytometry data

Automated biophysical classification of apoptotic pancreatic cancer cell subpopulations by using machine learning approaches with impedance cytometry

Feasibility study of stain-free classification of cell apoptosis based on diffraction imaging flow cytometry and supervised machine learning techniques

Quantitative Monitoring of Morphological Change of Cancer Cells Apoptosis by Digital Holographic Microscopy(Invited)

Transformer-based spatial–temporal detection of apoptotic cell death in live-cell imaging

Machine Learning Classifies Ferroptosis and Apoptosis Cell Death Modalities with TfR1 Immunostaining

Abstract 2311: Precise assessment of cancer cell growth and survival by artificial intelligence

A Real‐Time Image‐Based Approach to Distinguish and Discriminate Apoptosis from Necrosis

Evaluating Cellularity Estimation Methods: Comparing AI Counting with Pathologists' Visual Estimates

Image Analysis Algorithms for Immunohistochemical Assessment of Cell Death Events and Fibrosis in Tissue Sections

D-MAINS: A Deep-Learning Model for the Label-Free Detection of Mitosis, Apoptosis, Interphase, Necrosis, and Senescence in Cancer Cells

Study of the Relationship Between Apoptosis and Intracellular Ph in Single Living Cells Using a Three-Channel Real-Time Fluorescence Imaging Method

Learning deep features for dead and living breast cancer cell classification without staining

AI-Enhanced Blood Cell Recognition and Analysis: Advancing Traditional Microscopy with the Web-Based Platform IKOSA

Comparison of FDA-Approved vs. Mayo Clinic’s Ki-67 Immunostaining Protocols for Breast Cancer Specimens: A Quantitative Analysis Using Artificial Intelligence (AI) Based Image Analysis Algorithms

Evaluation of a Deep Learning Based Approach to Computational Label Free Cell Viability Quantification

AIM-CICs: an automatic identification method for cell-in-cell structures based on convolutional neural network