Systematic Review and Meta-Analysis of AI-Assisted Mammography and the Systemic Immune-Inflammation Index in Breast Cancer: Diagnostic and Prognostic Perspectives

<i>Background and Objectives</i>: Breast cancer remains a significant global health burden, demanding continuous innovation in diagnostic and prognostic tools. This meta-analysis and systematic review aims to synthesize evidence from 2015 to 2025 regarding the diagnostic utility of artif...

Full description

Saved in:
Bibliographic Details
Main Authors: Sebastian Ciurescu, Maria Ciupici-Cladovan, Victor Bogdan Buciu, Diana Gabriela Ilaș, Cosmin Cîtu, Ioan Sas
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Medicina
Subjects:
Online Access:https://www.mdpi.com/1648-9144/61/7/1170
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:<i>Background and Objectives</i>: Breast cancer remains a significant global health burden, demanding continuous innovation in diagnostic and prognostic tools. This meta-analysis and systematic review aims to synthesize evidence from 2015 to 2025 regarding the diagnostic utility of artificial intelligence (AI) in mammography and the prognostic value of the Systemic Immune-Inflammation Index (SII) in breast cancer patients. <i>Materials and Methods:</i> A systematic literature search was conducted in PubMed, Google Scholar, EMBASE, Web of Science, and Scopus. Studies evaluating AI performance in mammographic breast cancer detection and those assessing the prognostic significance of SII (based on routine hematologic parameters) were included. The risk of bias was assessed using QUADAS-2 and the Newcastle–Ottawa Scale. Meta-analyses were conducted using bivariate and random-effects models, with subgroup analyses by clinical and methodological variables. <i>Results:</i> Twelve studies were included, five assessing AI and seven assessing SII. AI demonstrated high diagnostic accuracy, frequently matching or surpassing that of human radiologists, with AUCs of up to 0.93 and notable reductions in radiologist reading times (17–91%). Particularly in dense breast tissue, AI improved detection rates and workflow efficiency. SII was significantly associated with poorer outcomes, including reduced overall survival (HR ~1.97) and disease-free survival (HR ~2.07). However, variability in optimal cut-off values for SII limits its immediate clinical standardization. <i>Conclusions:</i> AI enhances diagnostic precision and operational efficiency in mammographic screening, while SII offers a cost-effective prognostic biomarker for systemic inflammation in breast cancer. Their integration holds promise for more personalized care. Nevertheless, challenges persist regarding prospective validation, standardization, and equitable access, which must be addressed through future translational research.
ISSN:1010-660X
1648-9144