PHAROS Training Series - Course 7 "Quantitative Pathologic Assessment using AI-based Whole-Slide Image Analysis"
PHAROS AI Factory announces the 7th Course of its Training Series, under the title "Quantitative Pathologic Assessment using AI-based Whole-Slide Image Analysis", under the specialisation "Computational Pathology, Deep Learning, Whole-Slide Imaging (WSI), Spatial Biology, Tumor Microenvironment (TME), Digital Pathology, Precision Medicine", held online via Zoom.
Date: March 20th, 2026, at 12:00 EET
Location: Online via Zoom
Presentation Language: Greek
Description: Digital pathology has revolutionized the field of histopathological analysis, enabling unprecedented opportunities for quantitative assessment of tissue specimens. This course introduces a comprehensive AI-driven infrastructure designed to transform conventional pathology workflows into precise, reproducible, and scalable analytical pipelines. We will explore cutting-edge deep learning methodologies for automated tissue segmentation, enabling accurate delineation of tumor regions, stroma, necrosis, and other morphologically distinct areas within whole-slide images (WSIs). Building upon this foundation, we present advanced cell classification algorithms capable of identifying and categorizing diverse cellular populations with high accuracy and throughput. A significant focus will be placed on quantitative feature extraction from both traditional Hematoxylin and Eosin (H&E) stained sections and multiplex immunohistochemistry (mIHC) images. These computational approaches enable the derivation of morphometric, spatial, and contextual features that capture the complex tumor microenvironment architecture. Furthermore, we will demonstrate practical applications of deployed AI models for clinically relevant predictions, including molecular mutation status inference directly from histopathology images, patient survival outcome stratification, and treatment response prediction through pathologic complete response (pCR) assessment. These predictive models leverage the extracted quantitative features to provide actionable insights that can guide therapeutic decision-making. Throughout the presentation, emphasis will be placed on infrastructure design, model validation strategies, and clinical integration considerations essential for translating AI-based pathology tools from research settings to routine diagnostic practice.
Audience:
This course is suitable for:
- Pathologists and histopathology professionals
- Computational biologists and data scientists
- Oncology researchers and clinicians
- Medical imaging specialists
- Pharmaceutical/biotech R&D professionals
Learning Objectives:
By participating in this webinar, attendees will:
- Understand the principles of AI-based whole-slide image analysis and the infrastructure for digital pathology workflows.
- Describe deep learning approaches for automated tissue segmentation and cell classification in histopathology.
- Identify quantitative features extractable from H&E and multiplex immunohistochemistry images to characterize tumor microenvironment.
- Evaluate predictive AI models for mutation inference, survival stratification, and treatment response prediction.
- Recognize key considerations for validating and deploying AI-based pathology tools into clinical practice.
Learning Outcomes:
After completing the course, participants will be able to understand:
- AI infrastructure for whole-slide image analysis
- Tissue segmentation and cell classification AI methods
- Quantitative feature extraction from H&E and mIHC images using AI and image analysis techniques
- How predictive models work for mutations, survival, and pCR prediction
- AI deployment strategies in clinical pathology
Instructor's profile:
Georgios Manikis is a Marie Skłodowska-Curie Postdoctoral Fellow at the University of Cyprus and his fellowship research is dedicated to AI-driven computational pathology and multimodal predictive modeling in oncology. He is a collaborating researcher at the Computational BioMedicine Laboratory (CBML), FORTH and the Karolinska Institutet, Departent of Oncology-Pathology. His research interests lie in the areas of medical image analysis, machine and deep learning analysis.
Note: Please enter your institutional/corporate email when registering.