Partitioning a medical image

Abstract

Disclosed is a computer-implemented method for partitioning a medical image which encompasses acquiring a medical image of a surface of a patient and a surface of at least one object (embodied by image data), for example, by means of a 3D scanning device. Furthermore, a thermal image of the surface of the patient and the surface of the object is acquired (embodied by thermal data), for example, by means of a thermal camera. The medical image and the thermal image are fused (based on registration data), such that the image data is associated with the thermal data (embodied by association data). By analyzing the association data with regard to a specified condition (embodied by condition data), for example a condition related to a temperature threshold, a subset of the association data which fulfills the condition and describes a part of the medical image is determined (embodied by condition compliance data). For example, the object is colder than the skin of the patient, so by applying the temperature threshold, the part of the image representing the object may be identified. Accordingly, a part of the medical image showing a surface of the patient may be distinguished from a part of the medical image showing a surface of the object. Furthermore, and for example, one of those two parts may be defined to be a region of interest, and a positional shift of the region of interest (e.g. a face of the patient) may be determined by tracking the region of interest.