Related Experiment Video
Updated: Jul 16, 2026
Assessment of GFP Expression and Viability Using the Tali Image-Based Cytometer
Published on: November 17, 2011
Single-cell viability assessment using YOLOv8n object detection
Felix Peyre1, Allan Sauvat1, Marion Leduc1
1Centre de Recherche des Cordeliers, Inserm UMRS 1138, Sorbonne Université, Université Paris Cité, Équipe labellisée par la Ligue contre le Cancer, Institut Universitaire de France, Paris, France; INSERM US23/CNRS UAR 3655, Metabolomics and Cell Biology Platforms, Institut Gustave Roussy, Université Paris-Saclay, Villejuif, France.
Abstract:
Assessing cellular viability is a fundamental step for high-throughput screening (HTS) in drug development, particularly for evaluating the unintended adverse toxicity of standard medications or, conversely, the therapeutic efficacy of oncological agents, in which inducing cell death is the desired outcome. Conventional methods typically rely on ATP-dependent enzymatic substrate conversion assays to measure cellular metabolic activity as a proxy for viability. While effective, these plate reader-based methods provide only bulk measurements, limiting the level of detail in the analysis. Recent advancements in convolutional neural networks for biological image analysis have introduced promising alternatives, allowing the assessment of cellular viability at single-cell resolution. This approach provides more precise and detailed viability assessments than bulk assays. In this study, we present a methodology employing a trained YOLOv8n detection model, developed by Ultralytics, for endpoint and kinetic assays. This method offers rapid and precise toxicity predictions, making it a valuable tool in drug discovery.
Insights
This study introduces a new AI method using YOLOv8n for precise cellular viability assessment in drug discovery. It enables faster and more detailed toxicity predictions than traditional assays.
Area of Science:
- Biotechnology
- Drug Discovery
- Artificial Intelligence in Medicine
Background:
- Cellular viability assessment is crucial for high-throughput screening (HTS) in drug development.
- Conventional methods (e.g., ATP assays) provide bulk measurements, limiting detailed analysis.
- Adverse drug toxicity and therapeutic efficacy (oncology) require accurate viability data.
Purpose of the Study:
- To present a novel methodology for assessing cellular viability using AI.
- To enable precise and detailed viability assessments at single-cell resolution.
- To offer rapid toxicity predictions for drug discovery applications.
Main Methods:
- Utilized a trained YOLOv8n detection model for image analysis.
- Applied the model to endpoint and kinetic cellular assays.
- Leveraged convolutional neural networks for biological image analysis.
Main Results:
- Achieved single-cell resolution for cellular viability assessment.
- Demonstrated rapid and precise toxicity predictions.
- Provided more detailed viability data compared to bulk assays.
Conclusions:
- The YOLOv8n-based methodology offers a valuable advancement for drug discovery.
- AI-driven image analysis enhances the precision of cellular viability assessment.
- This approach supports more effective evaluation of drug toxicity and efficacy.
Related Concept Videos
01:21Cell Culture
01:16Cell Lines