The creation of breast lesion models for mammographic virtual clinical trials: a topical review

Abstract Simulated breast lesion models, including microcalcification clusters and masses, have been used in several studies. Realistic models are required for virtual clinical trials to be representative of performance. Multiple methods exist generate with various levels realism depending on the application. First, can obtained using mathematical methods, such as approximating a 3D geometric shapes or algorithmic techniques iterative processes grow lesion. On other hand, based patient data. They either created starting from characteristics real lesions they replica by segmenting cancer cases. Next, approaches embed these into structures create tumour The simplest method, typically calcifications, is intensity scaling. Two common hybrid total simulation which model inserted image model, respectively. In addition, artificial intelligence-based directly images. This article provides review literature available development insert them background their applications, optimisation studies, performance evaluation software education.