Related Experiment Video
Updated: Dec 5, 2025
Author Spotlight: Integrating High-Resolution Intravital Imaging and MRI to Enhance Stereotactic Body Radiation Therapy Planning
Published on: April 12, 2024
Multimodality Imaging and Artificial Intelligence for Tumor Characterization: Current Status and Future Perspective
Jérémy Dana1, Vincent Agnus2, Farid Ouhmich2
1IHU of Strasbourg, Strasbourg, France; Inserm & University of Strasbourg UMR-S1110, Strasbourg, France; Faculty of Medicine, University of Paris, Paris, France.
Abstract:
Research in medical imaging has yet to do to achieve precision oncology. Over the past 30 years, only the simplest imaging biomarkers (RECIST, SUV,…) have become widespread clinical tools. This may be due to our inability to accurately characterize tumors and monitor intratumoral changes in imaging. Artificial intelligence, through machine learning and deep learning, opens a new path in medical research because it can bring together a large amount of heterogeneous data into the same analysis to reach a single outcome. Supervised or unsupervised learning may lead to new paradigms by identifying unrevealed structural patterns across data. Deep learning will provide human-free, undefined upstream, reproducible, and automated quantitative imaging biomarkers. Since tumor phenotype is driven by its genotype and thus indirectly defines tumoral progression, tumor characterization using machine learning and deep learning algorithms will allow us to monitor molecular expression noninvasively, anticipate therapeutic failure, and lead therapeutic management. To follow this path, quality standards have to be set: standardization of imaging acquisition as it has been done in the field of biology, transparency of the model development as it should be reproducible by different institutions, validation, and testing through a high-quality process using large and complex open databases and better interpretability of these algorithms.
Insights
Artificial intelligence (AI) in medical imaging offers new quantitative biomarkers for precision oncology. Machine and deep learning can analyze complex data to noninvasively monitor tumors and guide treatment, improving patient outcomes.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Precision oncology currently lacks advanced imaging biomarkers beyond simple metrics like RECIST and SUV.
- Tumor characterization and monitoring of intratumoral changes remain significant challenges in medical imaging.
Purpose of the Study:
- To explore the potential of artificial intelligence (AI), specifically machine learning (ML) and deep learning (DL), in advancing medical imaging for precision oncology.
- To highlight the development of novel, automated, and reproducible quantitative imaging biomarkers.
Main Methods:
- Utilizing supervised or unsupervised ML/DL algorithms to analyze heterogeneous medical imaging data.
- Identifying and leveraging unrevealed structural patterns within complex datasets.
- Developing human-free, reproducible, and automated quantitative imaging biomarkers.
Main Results:
- AI algorithms can integrate diverse data for comprehensive tumor analysis.
- ML/DL enable noninvasive monitoring of molecular expression and tumor progression.
- AI facilitates anticipation of therapeutic failure and personalized treatment management.
Conclusions:
- AI, particularly DL, promises to revolutionize medical imaging for precision oncology by providing advanced quantitative biomarkers.
- Establishing quality standards, including data standardization, model transparency, and rigorous validation, is crucial for clinical adoption.
- Interpretable AI models are essential for trust and integration into clinical workflows.
