Background: Spinal metastasis is the third most common site for metastatic localization, following the lung and liver. Manual detection through imaging modalities such as CT, MRI, PET, and bone scintigraphy can be costly and inefficient. Preliminary artificial intelligence (AI) techniques and computer-aided detection (CAD) systems have attempted to improve lesion detection, segmentation, and treatment response in oncological imaging. The objective of this review is to evaluate the current applications of AI across multimodal imaging techniques in the diagnosis of spinal metastasis. Methods: Databases like PubMed, Scopus, Web of Science Advance, Cochrane, and Embase (Ovid) were searched using specific keywords like ‘spine metastases’, ‘artificial intelligence’, ‘machine learning’, ‘deep learning’, and ‘diagnosis’. The screening of studies adhered to the PRISMA guidelines. Relevant variables were extracted from each of the included articles such as the primary tumor type, cohort size, and prediction model performance metrics: area under the receiver operating curve (AUC), accuracy, sensitivity, specificity, internal validation and external validation. A random-effects meta-analysis model was used to account for variability between the studies. Quality assessment was performed using the PROBAST tool. Results: This review included 39 studies published between 2007 and 2024, encompassing a total of 6267 patients. The three most common primary tumors were lung cancer (56.4%), breast cancer (51.3%), and prostate cancer (41.0%). Four studies reported AUC values for model training, 16 for internal validation, and five for external validation. The weighted average AUCs were 0.971 (training), 0.947 (internal validation), and 0.819 (external validation). The risk of bias was the highest in the analysis domain, with 22 studies (56%) rated high risk, primarily due to inadequate external validation and overfitting. Conclusions: AI-based approaches show promise for enhancing the detection, segmentation, and characterization of spinal metastatic lesions across multiple imaging modalities. Future research should focus on developing more generalizable models through larger and more diverse training datasets, integrating clinical and imaging data, and conducting prospective validation studies to demonstrate meaningful clinical impact.

Background: Spinal metastases can cause significant impairment of neurological function and quality of life. Hence, personalized clinical decision-making based on prognosis and likely outcome is desirable. The effectiveness of AI in predicting complications and treatment outcomes for patients with spinal metastases is assessed. Methods: A thorough search was carried out through the PubMed, Scopus, Web of Science, Embase, and Cochrane databases up until 27 January 2025. Included were studies that used AI-based models to predict outcomes for adult patients with spinal metastases. Three reviewers independently extracted the data, and screening was conducted in accordance with PRISMA principles. AUC results were pooled using a random-effects model, and the PROBAST program was used to evaluate the study’s quality. Results: Included were 47 articles totaling 25,790 patients. For training, internal validation, and external validation, the weighted average AUCs were 0.762, 0.876, and 0.810, respectively. The Skeletal Oncology Research Group machine learning algorithms (SORG-MLAs) were the ones externally validated the most, continuously producing AUCs > 0.84 for 90-day and 1-year mortality. Models based on radiomics showed promise in preoperative planning, especially for outcomes of radiation and concealed blood loss. Most research concentrated on breast, lung, and prostate malignancies, which limited its applicability to less common tumors. Conclusions: AI models have shown reasonable accuracy in predicting mortality, ambulatory status, blood loss, and surgical complications in patients with spinal metastases. Wider implementation necessitates additional validation, data standardization, and ethical and regulatory framework evaluation. Future work should concentrate on creating multimodal, hybrid models and assessing their practical applications.

Background: Spinal metastases (SMs) are associated with poor prognosis and significant morbidity. We hypothesize that artificial intelligence (AI) models can enhance the identification and clinical utility of genetic and molecular signatures associated with SMs, improving diagnostic accuracy and enabling personalized treatment strategies. Methods: A systematic review of five databases was conducted to identify studies that used AI to predict genetic alterations and SMs outcomes. Accuracy, area under the receiver operating curve (AUC), and sensitivity were used for comparison. Data analysis was performed in R. Results: Eleven studies met the inclusion criteria, covering three different primary tumor origins, comprising a total of 2211 patients with an average of 201 ± 90 patients (range: 76–359 patients) per study. EGFR, Ki-67, and HER-2 were studied in ten (90.9%), two (18.1%), and one (9.1%) study, respectively. The weighted average AUC is 0.849 (95% CI: 0.835–0.863) and 0.791 (95% CI: 0.738–0.844) for internal and external validation of the established models, respectively. Conclusions: AI, through radiomics and machine learning, shows strong potential in predicting molecular markers in SMs. Our study demonstrates that AI can predict molecular markers in SMs with high accuracy.

Introduction: Three-dimensional deep learning offers promise for automated accurate brain tumor segmentation and survival prediction but requires robust validation across multiple MRI modalities to be effectively implemented in clinical practice. Methods: This study presents a comprehensive 3D deep learning framework using 369 cases from the BraTS2020 dataset. A 3D U-Net architecture was developed for tumor segmentation utilizing combined imaging data and optimized for computational efficiency and memory. The final 3D U-Net model segmentations were used to build machine learning 6-month and 12-month survival classifiers. Segmentation models were evaluated using multiple metrics, including the Dice Similarity Coefficient, Hausdorff Distance, and Cohen’s d. The classification models were evaluated using AUC-ROC and balanced accuracy. Results: Segmentation achieved a modest, but promising, performance across 30 epochs and with 295 training patients, achieving the best mean validation Dice = 0.8388 and a final-epoch mean Dice of 0.8263. Survival classification with a hybrid clinical and imaging logistic regression showed promising results, with 12-month prediction achieving AUC = 0.746 and 69% accuracy. The top contributing features for the 12-month prediction classifier were extent of resection, T1 contrast-enhanced tumor median, and FLAIR tumor median. Conclusions: This comprehensive framework demonstrates that a multi-modal approach provides meaningful performance gains, while segmentation-derived features show a promising ability to enable survival prediction.

Multi-sequence MRI is the standard for glioma assessment, but complete or contrast-enhanced protocols are not always available (withheld gadolinium, low-field MRI, incomplete archives). We asked whether tumor localization improves grade classification from T2-weighted MRI alone. In this retrospective study of the BraTS2020 training cohort (369 patients; legacy labels 293 high-grade [HGG], 76 low-grade [LGG]), 74 patients (15 LGG) were held out at the patient level for internal testing. A 2D U-Net trained in five development folds supplied out-of-fold whole-tumor masks for training 3D CNN classifiers; nnU-Net served as a segmentation benchmark. Classifiers received T2 alone or T2 plus a mask, which at test time came from the 2D U-Net, nnU-Net, the ground truth (oracle) or random noise. Analyses used five seeds, DeLong tests with Holm correction and development-set cross-validation. Whole-tumor Dice was 0.807 (95% CI 0.773–0.840) for the 2D U-Net and 0.877 (0.857–0.895) for nnU-Net. The T2-only CNN reached an ROC-AUC of 0.814 (95% CI 0.702–0.913). Mask-guided variants reached 0.677–0.742, none above T2-only (Holm-adjusted p = 1.00 for all four confirmatory comparisons of informative masks). Cross-validation reproduced this ranking (0.782 vs. 0.707). With LGG as the positive class, PR-AUC was 0.571 (no-skill 0.203). Tumor-mask guidance did not improve T2-only grading at any level of mask quality, including ground truth. The results concern a legacy grade label in one public cohort and require external testing.