Project duration:
t.b.a.
multimodality imaging with algorithm-assisted image analysis using machine learning
Project summary:
New hybrid imaging devices such as SPECT/CT, PET/CT and PET/MRI yield comprehensive multimodal information about pathophysiology on a molecular, functional and morphological level. This wealth of information promises a more accurate diagnosis of a disease, an improved selection of the appropriate therapy, an earlier assessment of therapy efficacy and a more accurate estimation of a patient’s prognosis. However, the vast amount of images acquired in the different stages of treatment is becoming more and more difficult to assess by the interpreting physician. Simply reviewing thousands of images, easily the numbers to deal with, is time-consuming and error-prone. Even more, cross-correlating all images is an enormous task that may well leave valuable information hidden in the data. In addition to images, text-based information such as prior reports and letters of referring physicians need to be taken into account for full apprehension of imaging data.
Funding:
WWU, Münster
