AI-Powered Quantitative Phenotyping of Organoids & 3D Spheroids for Drug Screening & Response Prediction
Transform 3D cell models into actionable drug-response insights using quantitative imaging, phenotypic analysis and AI-driven prediction.
Skills you will gain:
About Workshop:
This 3-day mentor-led workshop provides a practical introduction to the quantitative analysis of organoids and 3D spheroids for drug screening and response prediction. Participants will learn how microscopy images of 3D biological models can be processed, segmented and converted into measurable phenotypic features such as size, morphology, intensity, compactness and structural heterogeneity.
The workshop connects image-based phenotyping with dose–response analysis and machine learning to evaluate treatment effects and predict drug sensitivity. Through guided hands-on activities using Google Colab and open-source Python tools, participants will build a complete computational workflow from raw microscopy images to interpretable drug-response predictions.
Aim:
To equip participants with practical knowledge and computational skills for quantitative phenotyping of organoids and 3D spheroids, enabling them to analyse drug-treatment effects, generate dose–response profiles and apply AI-based models for drug-response prediction.
Workshop Objectives:
- Understand the biological and experimental significance of organoids and 3D spheroids in drug discovery.
- Learn the principles of microscopy-based quantitative phenotyping.
- Perform image preprocessing, segmentation and morphological feature extraction.
- Quantify organoid and spheroid size, shape, intensity and structural characteristics.
- Understand experimental design for 3D drug-screening studies.
- Analyse treatment-dependent phenotypic changes across drug concentrations.
- Generate and interpret dose–response curves and IC50 estimates.
- Integrate image-derived phenotypic features with drug-response data.
- Apply machine-learning approaches for drug-response classification and prediction.
- Evaluate predictive models using appropriate performance metrics.
- Identify important phenotypic features associated with drug sensitivity or resistance.
- Explore explainable AI approaches for interpreting drug-response predictions.
- Understand the potential of organoid-based computational phenotyping in precision medicine and preclinical drug development.
What you will learn?
Day 1 — Organoid & 3D Spheroid Imaging and Quantitative Phenotyping
Focus: Understanding 3D culture models and extracting measurable phenotypic features from microscopy images.
Topics Covered
- Organoids vs 3D spheroids: biological and experimental differences
- Applications in cancer research, toxicology and drug discovery
- Brightfield and fluorescence imaging of 3D cultures
- Image preprocessing: normalization, denoising and contrast enhancement
- Segmentation of organoids and spheroids
- Quantitative phenotypic features:
- Area and diameter
- Circularity and eccentricity
- Solidity and compactness
- Intensity-based features
- Growth and structural heterogeneity
- Quality control and common imaging artefacts
Hands-on:
Analyse a microscopy-image dataset in Google Colab, segment individual spheroids/organoids and extract quantitative morphological features.
Tools: Google Colab, Python, OpenCV, scikit-image, pandas, matplotlib
Day 2 — Drug Screening & Dose–Response Analysis in 3D Models
Focus: Translating phenotypic measurements into drug-response information.
Topics Covered
- Experimental design for 3D drug screening
- Control vs treated organoids/spheroids
- Viability and growth-inhibition measurements
- Phenotypic changes following drug treatment
- Dose–response relationships
- IC50 and EC50 concepts
- Normalization against untreated controls
- Comparing multiple compounds or concentrations
- Phenotypic signatures of sensitive and resistant models
- Limitations of conventional viability-only assays
Hands-on:
Analyse a multi-dose drug-treatment dataset and integrate image-derived phenotypic features with drug concentration and response measurements.
Tools: Google Colab, pandas, SciPy, scikit-learn, matplotlib
Day 3 — AI-Based Drug Response Prediction & Phenotypic Classification
Focus: Using machine learning to classify treatment response and identify important phenotypic biomarkers.
Topics Covered
- Preparing phenotypic datasets for machine learning
- Feature scaling and feature selection
- Responders vs non-responders
- Classification and regression approaches
- Random Forest and XGBoost concepts
- Predicting drug sensitivity from image-derived features
- Model evaluation:
- Accuracy
- Precision/Recall
- ROC-AUC
- RMSE for regression
- Feature importance and biological interpretation
- SHAP-based explainability
- Translational relevance for personalized drug screening
- Limitations, reproducibility and validation
Hands-on:
Build an AI model using quantitative organoid/spheroid features to predict drug response and identify the most informative phenotypic characteristics.
Tools: Google Colab, scikit-learn, XGBoost, SHAP, pandas, matplotlib
Mentor Profile
Fee Plan
Important Dates
13 Oct 2026 Indian Standard Timing 4:30 PM
13 Oct 2026 to 15 Oct 2026 Indian Standard Timing 5:00 PM
Get an e-Certificate of Participation!
Intended For :
- Undergraduate and postgraduate students
- PhD scholars and research fellows
- Researchers in biotechnology, biomedical sciences and life sciences
- Cancer biology and regenerative medicine researchers
- Pharmacology and pharmaceutical sciences students and professionals
- Drug discovery and preclinical research professionals
- Cell biology and tissue-engineering researchers
- Bioinformatics and computational biology researchers
- Researchers working with organoids, spheroids or 3D cell-culture models
Career Supporting Skills
Workshop Outcomes
- Differentiate between organoids and 3D spheroids and understand their applications in drug screening.
- Process brightfield or fluorescence microscopy images of 3D cellular models.
- Segment organoids and spheroids using computational image-analysis workflows.
- Extract quantitative morphological and intensity-based phenotypic features.
- Compare untreated and drug-treated 3D cultures using quantitative measurements.
- Analyse concentration-dependent treatment responses.
- Generate dose–response curves and estimate drug-response parameters.
- Create structured phenotypic datasets suitable for machine learning.
- Build classification or regression models for drug-response prediction.
- Evaluate the performance of predictive models.
- Identify key phenotypic features associated with treatment response.
- Interpret AI predictions using feature importance and explainability approaches.
- Develop an end-to-end workflow connecting 3D imaging, phenotyping, drug screening and AI-based response prediction.