Introduction: As the integration of Artificial Intelligence (AI) in healthcare continues to advance, the need for rigorous study design and research protocols tailored to diagnostic and prognostic studies becomes paramount. Aim: The primary objective of this work is to highlight the biostatistician’s point of view about the key points of the research protocol involving AI. Methods: Assessing the current state-of-the-art guidelines, we outline the methodological challenges faced by biostatisticians when collaborating on research protocols in the era of AI-driven medical research. Results: The proposed overview on research protocol involving AI elucidates key considerations in study design, encompassing evaluations of data quality, analysis of biases, methodological approaches, determination of sample size, and validation strategies tailored specifically to AI applications. This position paper underscores the pivotal role of strong statistical frameworks in ensuring the reliability, validity, and applicability of findings derived from AI-based diagnostic and prognostic models. Moreover, the paper seeks to highlight the critical importance of incorporating transparent reporting standards to enhance the reproducibility and clarity of AI-driven studies. Conclusions: By offering a comprehensive biostatistician’s viewpoint, this paper strives to significantly contribute to the methodological progression of diagnostic and prognostic studies in the era of Artificial Intelligence.

Study Design and Research Protocol for diagnostic or prognostic studies in the Age of Artificial Intelligence: A Biostatistician’s Perspective

Scotti L;Franchi M;VILLANI S.
2024-01-01

Abstract

Introduction: As the integration of Artificial Intelligence (AI) in healthcare continues to advance, the need for rigorous study design and research protocols tailored to diagnostic and prognostic studies becomes paramount. Aim: The primary objective of this work is to highlight the biostatistician’s point of view about the key points of the research protocol involving AI. Methods: Assessing the current state-of-the-art guidelines, we outline the methodological challenges faced by biostatisticians when collaborating on research protocols in the era of AI-driven medical research. Results: The proposed overview on research protocol involving AI elucidates key considerations in study design, encompassing evaluations of data quality, analysis of biases, methodological approaches, determination of sample size, and validation strategies tailored specifically to AI applications. This position paper underscores the pivotal role of strong statistical frameworks in ensuring the reliability, validity, and applicability of findings derived from AI-based diagnostic and prognostic models. Moreover, the paper seeks to highlight the critical importance of incorporating transparent reporting standards to enhance the reproducibility and clarity of AI-driven studies. Conclusions: By offering a comprehensive biostatistician’s viewpoint, this paper strives to significantly contribute to the methodological progression of diagnostic and prognostic studies in the era of Artificial Intelligence.
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/11571/1495696
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact