Qian Dong

Background: A comprehensive understanding of surgical scenes by computers is a crucial foundation for achieving intelligent surgical assistance and autonomous decision-making. Surgical scene information encompasses coarse-grained data reflecting the overall process and fine-grained details showcasing specific operations. This study aims to construct a standardized, full-grained annotation dataset for laparoscopic radical nephrectomy and develop a deep learning framework for multi-hierarchical granularity integration, providing support for clinical intelligent applications. Materials and methods: We collected 41 multi-center videos and performed comprehensive annotations, including 8 surgical phases, 17 surgical steps, 6 types of instrument segmentation, and 35 surgical action triplets (SATs). Using this dataset, we designed a lightweight framework that enables simultaneous perception of multi-granularity information through shared feature extraction and task-specific decoders. Based on the comprehensive understanding information, we developed a robust intraoperative bleeding prediction model using 123 combinations of machine learning algorithms and generated an automated postoperative surgical summary report. Results: The dataset of this study comprises 141 443 frames with annotations for surgical phases and steps, 8435 frames with instrument segmentation annotations, and 25 305 frames with SATs. Our deep learning framework achieves outstanding performance across tasks of different granularities, with recognition accuracy even surpassing that of six state-of-the-art algorithms developed separately for individual subtasks. Based on full-granularity features, we found that surgical steps involving key anatomical structures such as the renal artery, renal vein, and ureter are critical predictors of bleeding risk. And the post-operative automated reporting system accurately extracts key surgical time points and specific procedural durations. Conclusion: We constructed the first fully annotated dataset for laparoscopic radical nephrectomy, and developed a novel framework that achieves bidirectional compensation and enhancement of both coarse and fine-granularity information. This study not only expands the theoretical boundaries of artificial intelligence in surgical procedures, but also lays a solid foundation for advancing intelligent surgery through the “data–algorithm–application” cycle.

Hepatoblastoma (HB) is the most common malignant liver tumor in children. The expression of TAF9 is frequently upregulated in HB; however, its underlying molecular mechanisms are not yet fully understood, and its potential as a therapeutic target warrants further investigation. Bioinformatic analysis was performed using sequencing datas to evaluate clinical diagnostic and prognostic values of molecules. Biological functions were assessed using in vitro and in vivo experiments. Various techniques, including quantitative PCR, western blotting, immunohistochemistry, RNA immunoprecipitation, RNA pull-down, immunofluorescence, and luciferase reporter assays, were used to investigate the underlying molecular mechanisms. TAF9 was significantly overexpressed in HB tissues and correlated with poor prognosis. Both lncRNA938 and TAF9 promoted HB proliferation and metastasis. Mechanistically, lncRNA938 directly bound TAF9 and regulated its nuclear localization, while TAF9 activated TTK transcription via promoter binding. TTK inhibitors effectively reversed the epithelial-mesenchymal transition and malignant phenotypes induced by TAF9 overexpression. The lncRNA938–TAF9–TTK axis is a critical driver of HB progression. Targeting this axis, particularly through TTK inhibition, represents a novel therapeutic strategy against HB.

This study investigates the use of CT-based radiomics for predicting extrahepatic metastasis in hepatocellular carcinoma (HCC) following hepatectomy. We analyzed data from 374 patients from two centers (277 in the training cohort and 97 in an external validation cohort). Radiomic features were extracted from contrast-enhanced CT scans. Key features were identified using the least absolute shrinkage and selection operator (LASSO) to compute radiomics scores (radscore) for model development. A clinical model based on risk factors was also created. We developed a combined model integrating both radscore and clinical variables, constructing nomograms for personalized risk assessment. Model performance was compared via the Delong test, with calibration curves assessing prediction consistency. Decision curve analysis (DCA) was employed to assess the clinical utility and net benefit of the predictive models across different threshold probabilities, thereby evaluating their potential value in guiding clinical decision-making for extrahepatic metastasis. Radscore based on CT was an independent predictor of extrahepatic disease (p < 0.05). The combined model showed high predictive performance with an AUC of 87.2% (95% CI: 81.8%-92.6%) in the training group and 86.0% (95% CI: 69.4%-100%) in the validation group. Predictive performance of the combined model significantly outperformed both the radiomics and clinical models (p < 0.05). The DCA shows that the combined model has a higher net benefit in predicting extrahepatic metastases of HCC than the clinical model and radiomics model. The combined prediction model, utilizing CT radscore alongside clinical risk factors, effectively forecasts extrahepatic metastasis in HCC patients.

An abstract is unavailable. This article is available as a PDF only.

Background: Severe traumatic liver rupture in children has a high mortality rate. The computer-assisted surgery (CAS) system is an effective medical image simulation tool, which can display the adjacent relationship between the liver and surrounding tissues (especially compressed blood vessels) in a three-dimensional (3D), dynamic and complete way, and assist in precise liver resection. It provides important guidance for preoperative planning and intraoperative navigation. This chapter reports the individualized computer-assisted surgical planning and progress of a case of complex pediatric abdominal trauma. Case Description: A 3-year-old girl was admitted to the hospital due to a severe abdominal crush injury caused by a car accident. Contrast-enhanced computed tomography (CT) showed grade V liver injuries [2018 American Association for the Surgery of Trauma-Organ Injury Scale (2018 AAST-OIS)], as well as splenic and renal contusion. Emergency CAS was performed to repair liver contusion. Biliary fluid was drained from the chest cavity after the operation, and contrast-enhanced CT showed diaphragmatic rupture and intrahepatic pseudoaneurysm. The ruptured diaphragm was repaired by laparotomy with the assistance of a Hisense CAS system, and the intrahepatic pseudoaneurysm was treated by interventional therapy. The child responded well to the comprehensive treatment, and no complications such as bile leakage and infection were found after the operation. Regular imaging and laboratory tests confirmed that the child recovered stably, and the child displayed satisfactory physical development and growth during the follow-up period. Conclusions: The CAS system can predict surgical risk, and has important clinical value for in the treatment of children with multiple traumas.

This study aimed to measure the normal renal volume in children and adolescents aged 0–18 years from the eastern coastal region of China using the computer-assisted surgery (CAS) system. Accurate measurement of kidney volume is important as a marker for follow-up in patients with congenital anomalies. We established reference ranges for renal volume based on age, height, and weight, and developed predictive models using commonly available clinical data. A total of 803 children and adolescents participated, with clinical data including age, gender, height, weight, body mass index (BMI) and body surface area (BSA) collected. The Hisense CAS system was employed to perform 3D reconstructions of CT images, allowing precise measurement of left and right kidney volumes. Statistical analysis revealed a significant difference in size between the left and right kidneys (P < 0.01), with the left kidney being larger in 79.08% of cases. Age, height, weight, BMI, and BSA were all significantly correlated with kidney volume, with BSA showing the strongest correlation. We developed predictive formulas based on height and weight with good accuracy (R² = 0.896 for left kidney and R² = 0.891 for right kidney). These findings provide reference values for renal volume and offer useful tools for early detection and monitoring of renal abnormalities in the pediatric population.

Objective: This study aims to identify potential biomarkers for Hepatoblastoma (HB) using bioinformatics and machine learning, and to explore their underlying mechanisms of action.Methods: We analyzed the datasets GSE131329 and GSE133039 to perform differential gene expression analysis. Single-sample gene set enrichment analysis (ssGSEA) and weighted gene co-expression network analysis (WGCNA) were utilized to identify gene modules linked to gene set activity. Protein-protein interaction (PPI) networks were constructed to identify hub genes, while random forest and support vector machine models were employed to screen for key diagnostic genes. Survival and immune infiltration analyses were conducted to assess the prognostic significance of these genes. Additionally, the expression levels, biological functions, and mechanisms of action of the selected genes were validated in HB cells through relevant experimental assays.Results: We identified 1,377 and 1,216 differentially expressed genes in datasets GSE131329 and GSE133039, respectively. ssGSEA and WGCNA analyses identified 234 genes significantly linked to gene set activity. PPI analysis identified 20 core Hub genes. Machine learning highlighted three key diagnostic genes: CDK1, CCNA2, and MAD2L1. Studies have demonstrated that MAD2L1 is significantly overexpressed in HB and is associated with prognosis. WGCNA revealed that MAD2L1 is enriched in gene sets related to E2F_ TARGETS and G2M_CHECKPOINT. Experimental assays demonstrated that MAD2L1 knockdown significantly inhibits the proliferation, migration, and invasion of HB cell lines, and that MAD2L1 promotes cell cycle progression through the regulation of E2F.Conclusion: Our study identifies MAD2L1 as a novel potential biomarker for HB, providing new strategies for early diagnosis and targeted therapy in HB.

Hepatoblastoma (HB) is a common paediatric liver malignancy characterised by significant intratumoral heterogeneity and a complex tumour microenvironment (TME). Using single‐cell RNA sequencing (scRNA‐seq), we analysed 43,592 cells from three tumour regions and adjacent normal tissue of an HB patient. Our study revealed distinct cellular compositions and varying degrees of malignancy across different tumour regions, with the T1 region showing the highest malignancy and overexpression of HMGB2 and TOP2A. Survival analysis demonstrated that high HMGB2 expression is associated with poor prognosis and increased recurrence, suggesting its potential as a prognostic marker. Additionally, we identified a diverse immune microenvironment enriched with regulatory T cells (Tregs) and CD8⁺ effector memory T cells (Tem), indicating potential immune evasion mechanisms. Notably, CTLA‐4 and PD‐1 were highly expressed in Tregs and Tem cells, highlighting their potential as immunotherapy targets. Myeloid cells, including Kupffer cells and dendritic cells, also exhibited distinct functional roles in different tumour regions. This study provides the first comprehensive single‐cell atlas of HB, revealing critical insights into its intratumoral heterogeneity and immune microenvironment. Our findings not only advance the understanding of HB biology but also offer new directions for precision medicine, including the development of targeted therapies and immunotherapeutic strategies to improve patient outcomes.