1
Return

Multimodal biomarker AI techniques for early neurocognitive disorder diagnosis: A systematic review

delete2026-03-23
delete0
PRE
AI
F
Feliciana Catino
F
Fabio Castellana
R
Roberta Zupo
V
Viviana Giannoccaro
L
Luisa Lampignano
A
Angelo Michele Petrosillo
F
Francesco Addabbo
G
Giancarlo Sborgia
G
Giuseppe Colacicco
C
Carlo Santoro
G
Giovanni Boero
D
Donato Impedovo
Y
Yalin Zheng
R
Rodolfo Sardone *
DOI：10.1016/j.artmed.2026.103389delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Multimodal AI improves early AD/MCI detection over single-modality models. • Typical AUC gains: ~0.75–0.88 to ~0.85–0.95 with multimodal integration. • Most studies lacked external validation and used homogeneous datasets. • Emerging modalities (e.g., retina + plasma) show strong early-stage signal. • Clinical adoption needs fairness assessment and stronger model interpretability.
Keywords:
Multimodal AI
early detection
neurocognitive disorders
biomarkers
model interpretability

Journal

Artificial Intelligence in Medicine cover
Artificial Intelligence in Medicine
IF:
6.2
Papers:
2.5K
Citations:
7.8K

Organization

U
University of Bari Aldo Moro
Scholars:
1.0K
Papers: 343
Citations: 1.6W
U
University of Liverpool
Scholars:
2.8W
Papers: 2.5W
Citations: 3.5W
L
local health authority of taranto and municipality of taranto
Scholars:
9
Papers: 2
Citations: 0
Cited Papers

Cited Papers

Citing Papers

Citing Papers