Acute kidney injury (AKI) is a common and serious complication after cardiac surgery, affecting 10%-40% of patients. It worsens patient outcomes and consumes significant healthcare resources. Its pathophysiology is complex and involves ischemia-reperfusion injury, inflammatory responses, and endothelial dysfunction. Artificial intelligence (AI) offers considerable potential to improve the management of this condition. AI models can integrate multimodal data, including preoperative clinical profiles, intraoperative hemodynamics, and postoperative laboratory values, thereby enabling early prediction of AKI. By identifying distinct clinical subtypes, AI may support personalized therapeutic strategies. Furthermore, it may improve prognostic assessments, allowing more precise risk stratification for both cardiac and renal outcomes. However, current applications face challenges, including inconsistent data quality, limited model interpretability, and high implementation costs. Existing models are also constrained by the range of variables they incorporate. Future technological advances may enable the analysis of a broader array of variables, potentially revealing novel biomarkers and clinically useful combinations of indicators. Such progress could advance precision medicine in this field, ultimately improving patient care and optimizing clinical workflows.

Artificial intelligence is rapidly reshaping healthcare, with computer vision enabling automated interpretation of imaging across a wide range of clinical applications. Given its heavy reliance on imaging across perioperative settings, spine surgery is particularly well-suited for the integration of computer vision technologies. At the same time, spine surgery is among the fastest-growing and most resource-intensive specialties, with rising demands that underscore the need for technologies capable of enhancing precision, efficiency, and safety. In this narrative review, we will synthesize current applications of computer vision across the spine surgery workflow, outline key barriers for implementation, and discuss future directions for translating these tools into widespread clinical practice. Computer vision tools span a spectrum of maturity, with some already achieving clinical deployment for spinopelvic parameter measurement, pathology detection, and surgical planning, while intraoperative applications represent the most actively developing frontier. These innovations may redefine the standard of care in spine surgery, enabling a new era of surgical performance and data-informed decision-making.

Aim: To develop a deep learning radiomics (DLR) model based on Gd-EOB-DTPA- enhanced magnetic resonance imaging for preoperative prediction of dual-phenotype hepatocellular carcinoma (DPHCC) and patient prognosis.

Methods: This study included 331 patients from two centers: 228 in the training cohort (center I, 93 DPHCC, 135 non-DPHCC) and 103 in the external validation cohort (center II, 47 DPHCC, 56 non-DPHCC). Conventional radiological features were analyzed to build a radiological model. In total, 1,316 radiomics features and 768 deep learning (DL) features were extracted from the volumes of interest in arterial phase (AP), portal venous phase (PP), and hepatobiliary phase images, respectively. Three machine learning classifiers were applied to develop radiomics, DL, and DLR models using single-phase and combined-phase (CP) data for DPHCC prediction. Recurrence-free survival (RFS) was assessed by the Kaplan-Meier method.

Results: In the external validation cohort, rim AP hyperenhancement and non-smooth tumor margins were independent predictors of DPHCC (both P < 0.05), yielding a radiological model with an area under the curve (AUC) of 0.607 [95% confidence interval (CI), 0.511-0.704]. Among radiomics and DL models, the PP-DLR model achieved the highest discrimination (AUC: 0.805; 95%CI, 0.722-0.887), outperforming the radiological model, PP-Radiomics (AUC: 0.714; 95%CI, 0.615-0.813), CP-Radiomics (AUC: 0.639; 95%CI, 0.532-0.746), and CP-DLR models (AUC: 0.650; 95%CI, 0.543-0.758) (all P < 0.05). Patients classified as DPHCC by the PP-DLR model had significantly shorter median RFS than those classified as non-DPHCC (P < 0.05).

Conclusion: The PP-DLR model, which integrates radiomics and DL features, may aid in the preoperative prediction of DPHCC. DPHCC may be associated with poorer postoperative outcomes.

Aim: To evaluate deep learning models for anatomical structure and peritoneal metastasis (PM) detection and segmentation during staging laparoscopy (SL) using a phase-independent dataset, and to quantify how annotation strategy and spatial representation relate to predictive performance.

Methods: A checklist covering 25 anatomical structures, one surgical instrument, and PM was defined. Detection models (YOLOv9, Co-DETR) and segmentation models (SegFormer, Mask2Former) were trained under two label configurations. Videos were split at the video level (60/20/20). To quantify annotation distribution and spatial representation, two class-level descriptors were derived from the training set: object count and area fraction (percentage of image area occupied by each class). Class-level associations between these descriptors and test-set performance [F1-score, Intersection over Union (IoU)] were evaluated using Spearman correlation.

Results: Thirty SL videos yielded 2,309 annotated frames (1,304/433/572 for training/validation/testing). YOLOv9 reached mean mAP@50 of 0.52 and 0.61; Mask2Former achieved mean IoU of 0.51 and 0.61 and F1-scores of 0.65 and 0.73 for Sets A and B, respectively. Despite 4,094 annotations, PM remained difficult to segment (IoU 0.29-0.30; F1-score 0.45-0.46), due to low area fraction and high heterogeneity. For IoU, area fraction showed stronger correlations with performance than object count (ρ up to 0.66 vs. 0.48). Similar differences were observed for F1-score.

Conclusions: Anatomical detection and segmentation during SL are feasible but limited by small-target representation and heterogeneous intra-abdominal context. Spatial representation is more closely associated with segmentation performance than annotation frequency, supporting annotation strategies that address sparse pixel coverage in phase-independent intra-abdominal models.

Musculoskeletal disorders represent a leading cause of chronic pain and long-term disability worldwide, with conventional orthopedic practices often limited by one-size-fits-all management strategies. This scoping review maps the current landscape of evidence on artificial intelligence (AI)-integrated digital twin (DT) systems in orthopedics, focusing on their conceptual frameworks, clinical applications in personalized diagnosis, surgical simulation and prognostic management, and existing translational gaps. A systematic literature search was conducted across six electronic databases (PubMed, Embase, Web of Science, IEEE Xplore, ACM Digital Library and Scopus) up to 10 February 2026. Fourteen studies published between 2021 and 2026 met the inclusion criteria, with data charted across four predefined core dimensions: patient-specific DT construction, AI-driven biomechanical modeling, diagnostic stratification and surgical simulation, and individualized prognostic modeling. Accumulating preliminary evidence indicates that AI enables orthopedic DTs to evolve from static anatomical replicas to semi-dynamic predictive frameworks, with proof-of-concept applications validated across joint arthroplasty, spinal rehabilitation, foot and ankle surgery, and trauma care. The convergence of AI and DT technologies suggests preliminary technical feasibility and promising translational potential for personalized orthopedic care. However, the field remains in an early translational phase, limited by small sample sizes, single-center designs, insufficient large-scale clinical validation, and heterogeneous methodologies. Further multicenter clinical trials and standardized construction pipelines are required to confirm the clinical effectiveness, safety, and generalizability of AI-powered DT systems in routine orthopedic practice.