Early and accurate plant disease detection is crucial for sustainable agriculture and crop productivity. This review critically evaluates the artificial intelligence (AI) techniques, including machine learning (ML), deep neural networks (DNNs), and computer vision (CV) for automated disease detection across diverse horticultural crops. By analysing and synthesizing implementations in strawberries, blueberries, tomatoes, grapes, apples, squash, and others, we identify transferable knowledge and benchmarks applicable to raspberry ( Rubus idaeus ) monitoring within Western Norway's “FutuRaPS” project. Our study introduces following primary novel contributions: (a) the conceptualization and quantification of the ‘Algorithmic and Data Lag’ for raspberries disease detection, revealing a significant knowledge gap despite rapid DL advances and diverse imaging modalities (RGB, multispectral, hyperspectral for pre-symptomatic detection); (b) a tailored ‘knowledge transferability matrix’ that maps potential AI architectures and sensing strategies directly to raspberry-specific challenges; and (c) the ‘first actionable research roadmap’ for building intelligent, autonomous raspberry disease-monitoring framework applicable to Nordic and global environments. Our synthesis highlights promising AI-driven robotics and edge computing for real-time, in-field monitoring and targeted interventions, offering pathways to overcome persistent challenges like dataset limitations and environmental variability. By enabling scalable, field-validated solutions, this work provides a strategic contribution to the UN Sustainable Development Goals (SDGs 2, 12, and 13), fostering resilient and sustainable berry production in Nordic regions and globally. This systematic, PRISMA-compliant review rigorously maps progress and gaps, to guide operational Agriculture 4.0 solutions for raspberry horticulture worldwide. Here is a set of complementary highlights for this review article. • AI monitoring of raspberry diseases: gap analysis and cross-crop synthesis. • Benchmarks DL models (YOLO, CNN, ViT) and HSI for efficient field deployment. • Identifies critical data deficit and algorithmic lag in current raspberry research. • Proposes integrated FutuRaPS roadmap for resilient, edge-AI robotic monitoring. • Enables sustainable raspberry production using actionable AI framework.
Patil et al. (Sun,) studied this question.