MRI has several advantages over other imaging modalities, by providing 3-dimensional, multi-contrast, structural, functional, and metabolic information. The introduction of novel treatments and therapies in the field of medicine has led to the formation of new unique imaging phenomena, challenging conventional radiological interpretation. As such, imaging diagnosis needs to include advanced methods “tailored” to the patient, i.e., to their condition or therapy. The concept of personalized medicine (PM), aiming to tailor therapy to achieve the best response, highest safety margin, and better care, relates currently to the patient's specific genetic mutations and condition. Radiological assessment is currently far from being “personalized”, as it provides only general and non-specific information, mainly based on conventional imaging. In recent years, there are dramatic developments in the field of AI and the use of deep learning tools in different imaging modalities. New tools for automatic assessment and extraction of quantitative measures, show improved diagnosis, follow-up, and prognosis. These tools and the use of multi-contrast MRI seems highly necessary, yet their use in clinical practice is limited. In this talk I will introduce the approach to “personalized” MRI by using advanced imaging methods. I will explain how integrating multi-modalities information combined with clinical and genetic information for personalized strategies can be the key for early detection and prevention, thus improving outcome. Specifically, I will introduce AI methods for image analysis and diagnosis, to enable individual assessment. I will demonstrate the use of this approach in two different clinical scenarios, brain tumors and fetal MRI, and their implementation in our clinical site.

Acknowledgements

No acknowledgement found.

References

No reference found.