The fifth edition of the Ingenio al femminile Award, a celebration of female talent, has selected UniTS PhD student Giulia Saccomano as winner of the ‘Best doctoral thesis’ category for her work ‘From pixels to diagnosis: applications of X-ray Virtual Histology (XVH) in clinical pathology’.

Supported by the Italian National Council of Engineers and designed to foster women’s talent and representation in engineering, this year’s award is inspired by the theme ‘Artificial Intelligence meets the challenges of 2050’.

The motivation reads:

‘Traditional histopathological examination provides two-dimensional images due to histological samples’ cut plane. XVH, on the other hand, is a technique that allows high-resolution, non-destructive three-dimensional observations while preserving the structure of the organ under examination. The research work involved the integration of advanced deep learning algorithms to manage and analyse large XVH datasets. Automated organ segmentation in XVH images improves the identification of critical characteristics such as cell architecture and the margins of a tumour mass, while also being able to accurately calculate the values of prognostic markers without the need to physically dissect the affected part. The research has led to the integration between clinical diagnostic imaging of the Department of Pathological Anatomy of Trieste’s hospital and physical-experimental imaging of Elettra Sincrotrone Trieste, and the application of AI algorithms, in collaboration with the Computational Pathology Group at Radboud UMC (Netherlands), to improve the automatic segmentation of tumour masses and overcome the limits of traditional histopathology.’