The important progress in oncology observed over the last years has opened a new way of treating cancer thanks to the development of targeted therapies. These treatments allow the targeting of molecular changes identified in the tumor of the patient. By taking into consideration not only the phenotypic aspect of cancer, the evolution of this individualization has revolutionized the treatment of certain cancers, such as metastatic lung cancer. At the same time, radiology procedures have become generalized, and systematic. This allows the patient to benefit from successive, precise, exams of morphology (CT-scan) or function (PET-scan). Given the frenetic rhythm of this progress, it becomes clear that therapeutic decisions risk to be more and more difficult for the doctor to make. The multiplication of potential targets and their different impact on prognosis combined with the therapeutic history (treatment administered, efficacity, side effects, etc.) will create exponential increases in the difficulty for clinicians to choose the right treatment for the right patient, at the right time. The arrival of immunotherapies will make the panorama even more complex. Precise quantitative data on a number of genes, complex clinical histories, and morphologic and functional imaging data all need to be interpreted. In this talk, I will present some example of integration of multi-modal data in oncology for the diagnosis and the follow-up of the patient.