Handcrafted vs. Deep Radiomics vs. Fusion vs. Deep Learning: A Comprehensive Review of Machine Learning -Based Cancer Outcome Prediction in PET and SPECT Imaging

2025-07-21

AI Summary

Methodological heterogeneity and insufficient standardization hinder the clinical translation of radiomics in PET/SPECT-based cancer prognosis prediction. Method: We conducted a systematic review of 226 studies (2020–2025) and established, for the first time, a unified evaluation framework comprising 59 metrics. Performance of handcrafted radiomic features (HRF), deep radiomic features (DRF), end-to-end deep learning (DL), and hybrid fusion models was rigorously assessed using ANOVA, IBSI compliance analysis, and multidimensional metrics (e.g., AUC, accuracy). Contribution/Results: Hybrid fusion models achieved the highest AUC (0.861), while DRF attained peak accuracy (0.862). PET-based studies constituted 95% of the corpus and demonstrated superior performance. Only 48% of studies adhered to IBSI standards; prevalent issues included class imbalance and missing data. The study identifies methodological inconsistency as the primary translational bottleneck and underscores the critical need for standardized preprocessing pipelines and interpretable AI frameworks to enable robust clinical deployment.

Technology Category

• Machine Learning: Ensemble Methods
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success

Abstract

Machine learning (ML), including deep learning (DL) and radiomics-based methods, is increasingly used for cancer outcome prediction with PET and SPECT imaging. However, the comparative performance of handcrafted radiomics features (HRF), deep radiomics features (DRF), DL models, and hybrid fusion approaches remains inconsistent across clinical applications. This systematic review analyzed 226 studies published from 2020 to 2025 that applied ML to PET or SPECT imaging for outcome prediction. Each study was evaluated using a 59-item framework covering dataset construction, feature extraction, validation methods, interpretability, and risk of bias. We extracted key details including model type, cancer site, imaging modality, and performance metrics such as accuracy and area under the curve (AUC). PET-based studies (95%) generally outperformed those using SPECT, likely due to higher spatial resolution and sensitivity. DRF models achieved the highest mean accuracy (0.862), while fusion models yielded the highest AUC (0.861). ANOVA confirmed significant differences in performance (accuracy: p=0.0006, AUC: p=0.0027). Common limitations included inadequate handling of class imbalance (59%), missing data (29%), and low population diversity (19%). Only 48% of studies adhered to IBSI standards. These findings highlight the need for standardized pipelines, improved data quality, and explainable AI to support clinical integration.

Problem

• Compare ML methods for cancer prediction in PET/SPECT imaging
• Evaluate performance differences among radiomics and deep learning approaches
• Identify limitations in current ML-based outcome prediction studies

Innovation

• Deep radiomics features for high accuracy
• Hybrid fusion models for best AUC
• Standardized pipelines for clinical integration