We present AIDA (AI-Driven Intelligent Diagnostics and Analytics), an ongoing initiative aimed at advancing the field of digital pathology through the integration of artificial intelligence. The aim of AIDA is twofold: 1) to create a collaborative repository of annotated data on rare pathologies, where specialists from around the world contribute expert insights through both text and graphical explanations of complex cases; and 2) to leverage this annotated data to train machine learning tools, with an emphasis on Visual Language Models (VLMs), that can enhance and accelerate the diagnosis of such conditions. The project adopts a human-in-the-loop paradigm supported by retrieval and exploration of large-scale, high-dimensional medical data. This position paper outlines the current status of the AIDA project, highlights key challenges, and explores possible directions for its ongoing and future evolution.

AIDA: AI-Driven Intelligent Diagnostics and Analytics

Ruan, Zanxi;Gobbo, Stefano;Cima, Luca;Sharifi, Shakiba;Munari, Enrico;Scarpa, Aldo;Giachetti, Andrea;Setti, Francesco;Wang, Yiming;Cristani, Marco
2026-01-01

Abstract

We present AIDA (AI-Driven Intelligent Diagnostics and Analytics), an ongoing initiative aimed at advancing the field of digital pathology through the integration of artificial intelligence. The aim of AIDA is twofold: 1) to create a collaborative repository of annotated data on rare pathologies, where specialists from around the world contribute expert insights through both text and graphical explanations of complex cases; and 2) to leverage this annotated data to train machine learning tools, with an emphasis on Visual Language Models (VLMs), that can enhance and accelerate the diagnosis of such conditions. The project adopts a human-in-the-loop paradigm supported by retrieval and exploration of large-scale, high-dimensional medical data. This position paper outlines the current status of the AIDA project, highlights key challenges, and explores possible directions for its ongoing and future evolution.
2026
9783032113801
Digital Pathology; Human in the Loop; VLMs
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11562/1181868
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact