Related Experiment Video
Updated: Jun 25, 2025
07:49
Evaluation of Tumor-infiltrating Leukocyte Subsets in a Subcutaneous Tumor Model
Published on: April 13, 2015
20.1K
Enhancing AI Research for Breast Cancer: A Comprehensive Review of Tumor-Infiltrating Lymphocyte Datasets
Alessio Fiorin1,2,3, Carlos López Pablo4,5,6, Marylène Lejeune7,8,9
1Oncological Pathology and Bioinformatics Research Group, Institut d'Investigació Sanitària Pere Virgili (IISPV), C/Esplanetes no 14, 43500, Tortosa, Spain. alessio.fiorin@estudiants.urv.cat.
Journal of Imaging Informatics in Medicine
|May 28, 2024
Summary
This review examines publicly available datasets for tumor-infiltrating lymphocytes (TILs) to aid in training computer-assisted pathology (CAP) tools. It addresses the scarcity of annotated data for deep learning models in breast cancer immunology.
Area of Science:
- Immunology
- Computational Pathology
- Bioinformatics
Background:
- Tumor-infiltrating lymphocytes (TILs) are crucial in breast cancer prognosis and treatment response.
- Computer-assisted pathology (CAP) tools, utilizing deep learning, are increasingly used for TIL quantification.
- Training these AI models requires extensive annotated datasets, which are currently scarce and time-consuming to create.
Purpose of the Study:
- To review and evaluate publicly accessible datasets for TIL assessment.
- To provide a valuable resource for researchers and developers in the TIL and CAP fields.
- To facilitate the training and validation of current and future CAP tools for TIL analysis.
Main Methods:
- Systematic review of publicly available online datasets related to TILs.
- Inspection and evaluation of dataset accessibility, annotation quality, and suitability for AI model training.
- Analysis of the current landscape of TIL datasets for computer-assisted pathology.
Main Results:
- Identification and assessment of existing public datasets relevant to TIL quantification.
- Highlighting the challenges associated with dataset scarcity and annotation efforts.
- Providing an overview of resources available for advancing CAP tools in breast cancer immunology.
Conclusions:
- Publicly available datasets are essential for advancing AI-driven TIL assessment in breast cancer.
- A comprehensive review of these datasets can significantly streamline the development and validation of CAP tools.
- Further efforts in curating and annotating TIL datasets will accelerate progress in computational pathology and cancer immunology.
