Role of Three-Dimensional Convolution Neural Networks (3D-CNN) in Image Processing and Recognition in Oncology: A Systematic Review and Meta-Analysis.

Three-dimensional convolutional neural networks (3D CNNs) have transformed oncology imaging, excelling in tumor detection, classification, segmentation, and prognosis prediction. Unlike traditional two-dimensional CNNs, 3D CNNs effectively analyze volumetric medical imaging data, enhancing spatial feature extraction and diagnostic accuracy across modalities, including CT, MRI, PET, and ultrasound. This systematic review and meta-analysis evaluates the diagnostic performance and clinical utility of 3D CNNs across 22 studies, of which 11 were eligible for quantitative synthesis. Pooled sensitivity, specificity, and AUC were 0.72, 0.73, and 0.77, respectively, with a diagnostic odds ratio of 10.38, indicating favorable discriminative ability. Subgroup analyses demonstrated superior accuracy in lung cancer and CT-based models, with DenseNet and ResNet architectures outperforming traditional CNNs. Technical innovations-including multi-modal fusion, spatial context integration, and explainable AI techniques-enhance model robustness and clinician trust. However, substantial heterogeneity (I² > 95%) across studies, attributable to differences in imaging protocols, dataset quality, and model design, underscores the need for standardized methodologies. Persistent challenges include computational demands, annotation variability, and generalization limitations. Future directions should prioritize the integration of explainable AI, PACS-compatible user interfaces, and federated learning frameworks to bridge institutional gaps. This review highlights the considerable promise of 3D CNNs in advancing precision oncology, while also identifying the infrastructural and methodological refinements necessary for widespread clinical adoption.