DIGITAL PATHOLOGY NETWORK AND ARTIFICIAL INTELLIGENCE IN SAPIENZA'S PATHOLOGY UNIT.

Anno
2021
Proponente Carlo Della Rocca - Professore Ordinario
Sottosettore ERC del proponente del progetto
LS3_1
Componenti gruppo di ricerca
Componente Categoria
Claudio Di Cristofano Componenti strutturati del gruppo di ricerca
Martina Leopizzi Componenti strutturati del gruppo di ricerca
Componente Qualifica Struttura Categoria
Massimiliano Mancini Dirigente medico Azienda Ospedaliero-Universitaria Sant'Andrea Altro personale aggregato Sapienza o esterni, titolari di borse di studio di ricerca
Caterina Chiappetta Dirigente biologo Azienda Ospedaliero-Universitaria Policlinico Umberto I Altro personale aggregato Sapienza o esterni, titolari di borse di studio di ricerca
Fabrizio De Lorenzo Informatico Sapienza, Università di Roma Altro personale aggregato Sapienza o esterni, titolari di borse di studio di ricerca
Abstract

Light microscopy is still considered the gold standard in pathology even though acceleration of technical progress in diagnostic disciplines, has enormously increased during recent years. Digital pathology (DP) is the integrated use of technology to assist in the creation, sharing, and exchange of information, including data and images, and to support the complex workflow, which ranges from receipt of study material to submission of the final report. DP can also improve diagnostic reliability and reproducibility: in certain cases, computers can fulfill tedious tasks more objectively and more rapidly. DP can be considered as one of the very first attempt in the field of applied artificial intelligence (AI). AI has a great potential as a data-driven DP approach, indeed its use in DP is referred to as the third major revolution in pathology after the introduction of immunohistochemistry in the 1980s and molecular pathology in the 1990s. Aim of our project is to build a DP network among four Pathology Units. The DP network will support second opinion consultation between pathologists from the four Units making available different expertise. The archival library of digital slides will permit remote learning both for medical students and pathology residents; it could help in the scientific exchange with colleagues in multidisciplinary tumor board (MTB) conferences. Real time frozen section intraoperative consultation from specialists at other institutions could also be possible thus reducing the risk of diagnostic errors and pitfalls. Archival digitized cases could provide an excellent source of datasets for the development and training of AI.
DP will open up to new diagnostic possibilities and with real-time integration in existing clinical infrastructures will allow telemedical applications, automated image evaluations with benefits both in the patient care and in medical education.

ERC
LS2_13
Keywords:
ANATOMIA PATOLOGICA, INTELLIGENZA ARTIFICIALE, BIOLOGIA COMPUTAZIONALE

© Università degli Studi di Roma "La Sapienza" - Piazzale Aldo Moro 5, 00185 Roma