The purpose of this study is to identify and validate new imaging biomarkers allowing an individual phenotyping of patient with venous thrombo-embolism (VTE), mainly in terms of recurrence risk assessment and to distinguish provoked from unprovoked VTE. To do so, the investigators will create a retrospective imaging database including multiple imaging modalities, performed at diagnosis of the VTE.

Using a large imaging data in VTE (including ultrasound, CT-scan and Ventilation/Perfusion lung scan), the investigators will search for association between imaging data and VTE phenotypes, mainly in terms of recurrence risk assessment and to distinguish provoked from unprovoked VTE. Then, using a radiomic approach, the investigators will try to identify new imaging biomarkers able to provide an individual phenotyping in these patients.

These imaging biomarkers will be validated using an internal cross validation technique.

Inclusion Criteria:

• 18 years or older.
• Confirmed VTE diagnosis (deep vein thrombosis or pulmonary embolism).
• Inclusion in the EDITH (Etude des Déterminants et Interaction de la THrombose veineuse) between November 2009 and November 2019.
• No opposition to be included in the present study.

Exclusion Criteria:

• Patients under 18 years old.
• Patients under judicial protection.
• Patient physically or cognitively unable to give consent.
• Refusal to participate.

Every patients included in the EDITH cohort between November 2009 and Novembre 2019, with a confirmed diagnosis of VTE.