Liver transplantation following two conversions in a patient with huge hepatocellular carcinoma and portal vein invasion: A case report.

BACKGROUND: Surgical resection and liver transplantation (LT) are the most effective curative options for hepatocellular carcinoma (HCC). However, few patients with huge HCC (> 10 cm in diameter), especially those with portal vein tumor thrombus (PVTT), can receive these treatments. Selective internal radiation therapy (SIRT) can be used as a conversion therapy for them because it has the dual benefit of shrinking tumors and increasing residual hepatic volume. However, in patients with huge HCC, high lung absorbed dose often prevents them from receiving SIRT. CASE SUMMARY: A 35-year-old man was admitted because of emaciation and pain in the hepatic region for about 1 month. The computed tomography scan showed a 20.2 cm × 19.8 cm tumor located in the right lobe-left medial lobes with right portal vein and right hepatic vein invasion. After the pathological type of HCC was confirmed by biopsy, two conversions were presented. The first one was drug-eluting bead transarterial chemoembolization plus hepatic arterial infusion chemotherapy and lenvatinib and sintilimab, converted to SIRT, and the second one was sequential SIRT with continued systemic treatment. The tumor size significantly decreased from 20.2 cm × 19.8 cm to 16.2 cm × 13.8 cm, then sequentially to 7.8 cm × 6.8 cm. In the meantime, the ratio of spared volume to total liver volume increased gradually from 34.4% to 55.7%, then to 62.9%. Furthermore, there was visualization of the portal vein, indicating regression of the tumor thrombus. Finally, owing to the new tumor in the left lateral lobe, the patient underwent LT instead of resection without major complications. CONCLUSION: Patients with inoperable huge HCC with PVTT could be converted to SIRT first and accept surgery sequentially.

A flexible deep learning framework for liver tumor diagnosis using variable multi-phase contrast-enhanced CT scans.

BACKGROUND: Liver cancer is a significant cause of cancer-related mortality worldwide and requires tailored treatment strategies for different types. However, preoperative accurate diagnosis of the type presents a challenge. This study aims to develop an automatic diagnostic model based on multi-phase contrast-enhanced CT (CECT) images to distinguish between hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICC), and normal individuals. METHODS: We designed a Hierarchical Long Short-Term Memory (H-LSTM) model, whose core components consist of a shared image feature extractor across phases, an internal LSTM for each phase, and an external LSTM across phases. The internal LSTM aggregates features from different layers of 2D CECT images, while the external LSTM aggregates features across different phases. H-LSTM can handle incomplete phases and varying numbers of CECT image layers, making it suitable for real-world decision support scenarios. Additionally, we applied phase augmentation techniques to process multi-phase CECT images, improving the model's robustness. RESULTS: The H-LSTM model achieved an overall average AUROC of 0.93 (0.90, 1.00) on the test dataset, with AUROC for HCC classification reaching 0.97 (0.93, 1.00) and for ICC classification reaching 0.90 (0.78, 1.00). Comprehensive validation in scenarios with incomplete phases was performed, with the H-LSTM model consistently achieving AUROC values over 0.9. CONCLUSION: The proposed H-LSTM model can be employed for classification tasks involving incomplete phases of CECT images in real-world scenarios, demonstrating high performance. This highlights the potential of AI-assisted systems in achieving accurate diagnosis and treatment of liver cancer. H-LSTM offers an effective solution for processing multi-phase data and provides practical value for clinical diagnostics.

Artificial intelligence in predicting recurrence after first-line treatment of liver cancer: a systematic review and meta-analysis.

BACKGROUND: The aim of this study was to conduct a systematic review and meta-analysis to comprehensively evaluate the performance and methodological quality of artificial intelligence (AI) in predicting recurrence after single first-line treatment for liver cancer. METHODS: A rigorous and systematic evaluation was conducted on the AI studies related to recurrence after single first-line treatment for liver cancer, retrieved from the PubMed, Embase, Web of Science, Cochrane Library, and CNKI databases. The area under the curve (AUC), sensitivity (SENC), and specificity (SPEC) of each study were extracted for meta-analysis. RESULTS: Six percutaneous ablation (PA) studies, 16 surgical resection (SR) studies, and 5 transarterial chemoembolization (TACE) studies were included in the meta-analysis for predicting recurrence after hepatocellular carcinoma (HCC) treatment, respectively. Four SR studies and 2 PA studies were included in the meta-analysis for recurrence after intrahepatic cholangiocarcinoma (ICC) and colorectal cancer liver metastasis (CRLM) treatment. The pooled SENC, SEPC, and AUC of AI in predicting recurrence after primary HCC treatment via PA, SR, and TACE were 0.78, 0.90, and 0.92; 0.81, 0.77, and 0.86; and 0.73, 0.79, and 0.79, respectively. The values for ICC treated with SR and CRLM treated with PA were 0.85, 0.71, 0.86 and 0.69, 0.63,0.74, respectively. CONCLUSION: This systematic review and meta-analysis demonstrates the comprehensive application value of AI in predicting recurrence after a single first-line treatment of liver cancer, with satisfactory results, indicating the clinical translation potential of AI in predicting recurrence after liver cancer treatment.

Endoscopic ultrasound-guided vascular interventions: A review (with videos).

Endoscopic ultrasound (EUS) has evolved from a diagnostic to an interventional modality, allowing precise vascular access and therapy. EUS-guided vascular access of the portal vein has received increasing attention in recent years as a diagnostic and therapeutic tool. EUS-guided portal pressure gradient directly measures the hepatic vein portal pressure gradient and is crucial for understanding of liver function and prognostication of liver disease. EUS facilitates the sampling of portal venous blood to obtain circulating tumor cells (CTCs) in pancreatobiliary malignancies. This technique aids in the diagnosis and staging of cancers. EUS-guided interventions have a substantial potential for diagnosing portal vein tumor thrombus (PVTT) in patients with hepatocellular carcinoma. EUS-guided coil and glue embolization have higher efficacy for the treatment of gastric varices than direct endoscopic glue. Pseudoaneurysm (PsA), a rare vascular complication of acute and chronic pancreatitis, is typically managed with interventional radiology (IR)-guided embolization and surgery. EUS is increasingly used in specialized centers for non-variceal gastrointestinal bleeding, particularly for pseudoaneurysm-related bleeding. There is limited data on EUS-guided intervention for bleeding ectopic varices, rectal varices and Dieulafoy lesions, but it is becoming more widely accepted. In this extensive review, we evaluated both current and potential future applications of EUS-guided vascular interventions, including EUS-guided gastric variceal bleed therapy, rectal and ectopic varices, pseudoaneurysmal bleeding, splenic artery embolization, portal pressure gradient measurement, portal vein sampling for CTCs, fine needle aspiration of PVTT, intrahepatic portosystemic shunt placement, liver tumor ablation and EUS-guided cardiac intervention.

Multiparametric MRI based deep learning model for prediction of early recurrence of hepatocellular carcinoma after SR following TACE.

BACKGROUND: Surgical resection (SR) following transarterial chemoembolization (TACE) is a promising treatment for unresectable hepatocellular carcinoma (uHCC). However, biomarkers for the prediction of postoperative recurrence are needed. PURPOSE: To develop and validate a model combining deep learning (DL) and clinical data for early recurrence (ER) in uHCC patients after TACE. METHODS: A total of 511 patients who received SR following TACE were assigned to derivation (n = 413) and validation (n = 98) cohorts. Deep learning features were taken from the largest tumor area in liver MRI. A nomogram using DL signatures and clinical data was made to forecast early recurrence risk in uHCC patients. Model performance was evaluated using area under the curve (AUC). RESULTS: A total of 2278 subsequences and 31,346 slices multiparametric MRI including contrast-enhanced T1WI, T2WI and DWI were input in the DL model simultaneously. Multivariable analysis identified three independent predictors for the development of the nomogram: tumor number (hazard ratio [HR]:3.42, 95% confidence interval [CI]: 2.75-4.31, P = 0.003), microvascular invasion (HR: 9.21, 6.24-32.14; P < 0.001), and DL scores (HR: 17.46, 95% CI: 12.94-23.57, P < 0.001). The AUC of the nomogram was 0.872 and 0.862 in two cohorts, significantly outperforming single-subsequence-based DL mode and clinical model (all, P < 0.001). The nomogram provided two risk strata for cumulative overall survival in two cohorts, showing significant statistical results (P < 0.001). CONCLUSIONS: The DL-based nomogram is essential to identify patients with uHCC suitable for treatment with SR following TACE and may potentially benefit personalized decision-making.

Exploring transparent reporting and data availability in systematic reviews to identify subgroup evidence: imaging for suspected hepatocellular carcinoma in the non-cirrhotic liver.

We aim to illustrate the role of complete and transparent reporting coupled with access to data sourced from published systematic reviews, especially assisting in the identification of evidence for subgroups within the context of a rare disease. To accomplish this principle, we provide a real-world example encountered during the revision of the Dutch clinical practice guideline for hepatocellular carcinoma. Specifically, we retrieved insights from two Cochrane reviews to identify direct evidence concerning the diagnostic test accuracy of computed tomography and magnetic resonance imaging for detecting hepatocellular carcinomas in suspected patients without liver cirrhosis. Through reusing the Cochrane review authors' efforts already undertaken in their exhaustive literature search and selection, we successfully identified relevant direct evidence for this subgroup of suspected patients without cirrhosis and performed an evidence synthesis within the constraints of limited resources for the guideline revision. This approach holds the potential for replication in other subgroups in the context of rare diseases, contingent on the transparent and complete reporting of systematic reviews, as well as the availability and accessibility of their extracted data. Consequently, we underscore the importance of adhering to established reporting guidelines for systematic reviews, while simultaneously advocating for increased availability and accessibility to data. Such practices would not only increase the transparency and reproducibility of systematic reviews but could also increase reusability of their data. In turn, the increased reusability could result in reduced resource utilization in other sectors such as the guideline developing community as we show in our example.

Radio-opaque contrast agents for liver cancer targeting with KIM during radiation therapy (ROCK-RT): an observational feasibility study.

BACKGROUND: This observational study aims to establish the feasibility of using x-ray images of radio-opaque chemoembolisation deposits in patients as a method for real-time image-guided radiation therapy of hepatocellular carcinoma. METHODS: This study will recruit 50 hepatocellular carcinoma patients who have had or will have stereotactic ablative radiation therapy and have had transarterial chemoembolisation with a radio-opaque agent. X-ray and computed tomography images of the patients will be analysed retrospectively. Additionally, a deep learning method for real-time motion tracking will be developed. We hypothesise that: (i) deep learning software can be developed that will successfully track the contrast agent mass on two thirds of cone beam computed tomography (CBCT) projection and intra-treatment images (ii), the mean and standard deviation (mm) difference in the location of the mass between ground truth and deep learning detection are ≤ 2 mm and ≤ 3 mm respectively and (iii) statistical modelling of study data will predict tracking success in 85% of trial participants. DISCUSSION: Developing a real-time tracking method will enable increased targeting accuracy, without the need for additional invasive procedures to implant fiducial markers. TRIAL REGISTRATION: Registered to ClinicalTrials.gov (NCT05169177) 12th October 2021.

Construction of in-situ self-assembled agent for NIR/PET dual-modal imaging and photodynamic therapy for hepatocellular cancer.

Hepatocellular cancer (HCC) remained a life-threatening carcinoma. Agents for HCC imaging and therapy were expected to possess different intratumoral retention time. To construct an agent with different intratumoral retention time when applied for tumor imaging or therapy remained great values. A lasialoglycoprotein receptor (ASGPR) targeted lactobionic acid derivative (LABO) was constructed for fluorescent imaging and photodynamic therapy of HCC. 18F labeled LABO (18F-LABO) was developed for PET imaging of HCC. LABO and 18F-LABO showed similar molecular structure. LABO exhibited characteristic of viscosity and concentration-induced intratumoral in-situ self-assembly to expand the intratumoral retention. LABO was non-fluorescent at free stage, but emitted NIR fluorescence and generated irradiation-induced ROS after self-assembly for fluorescent imaging and photodynamic therapy. ASGPR specificity of LABO and 18F-LABO was confirmed using HepG2 cell. Biodistribution and fluorescent imaging confirmed the different tumor retention time of LABO and 18F-LABO when used for photodynamic therapy and PET imaging. PET imaging and photodynamic therapy were performed on HepG2 tumor bearing mice, which revealed that 18F-LABO/LABO could specifically accumulated in the HepG2 tumor for tumor location/inhibition. LABO/18F-LABO with excellent HCC specificity but different intratumoral behaviors showed great values for the PET/NIR imaging and photodynamic therapy for HCC.

Detection of hepatocellular carcinoma feeding vessels: MDCT angiography with 3D reconstruction versus digital subtraction angiography.

BACKGROUND: Accurate detection of Hepatocellular carcinoma (HCC) feeding vessels during transcatheter arterial chemoembolization (TACE) is important for an effective treatment, while limiting non-target embolization. This study aimed to investigate the feasibility and accuracy of pre-TACE three dimensional (3D) CT angiography for tumor-feeding vessels detection compared to DSA. METHODS: Sixty-nine consecutive patients referred for TACE from May 2022 to May 2023 were included. (3D) CT images were reconstructed from the pre-TACE diagnostic multiphasic contrast enhanced CT images and compared with non-selective digital subtraction angiography (DSA) images obtained during TACE for detection of HCC feeding vessels. A "Ground truth" made by consensus between observers after reviewing all available pre-TACE CT images, and DSA and CBCT images during TACE to detect the true feeding vessels was the gold standard. Sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), accuracy and ROC curve with AUC were calculated for each modality and compared. RESULTS: A total of 136 active HCCs were detected in the 69 consecutive patients included in the study. 185 feeding arteries were detected by 3D CT and DSA and included in the analysis. 3D CT detection of feeding arteries revealed mean sensitivity, specificity, PPV, NPV and accuracy of 91%, 71%, 98%, 36%, and 90%, respectively, with mean AUC = 0.81. DSA detection of feeding arteries revealed mean sensitivity, specificity, PPV, NPV, and accuracy of 80%, 58%, 96.5%, 16.5% and 78%, respectively, with mean AUC = 0.69. CONCLUSIONS: Pre-TACE 3D CT angiography has shown promise in improving the detection of HCC feeding vessels compared to DSA. However, further studies are required to confirm these findings across different clinical settings and patient populations. TRIAL REGISTRATION: This study was prospectively registered at Clinicaltrials.gov with ID NCT05304572; Date of registration: 2-4-2022.

An initial report of robotic-assisted anatomical liver resection with indocyanine green fluorescence navigation using the ultrasound-guided preoperative positive staining technique.

INTRODUCTION: Robotic-assisted surgery has become increasingly popular because of its potential benefits. Anatomical liver resection (ALR) is a valuable strategy in hepatocellular carcinoma (HCC) management. ALR with indocyanine green (ICG) fluorescence navigation was reported as an effective solution for segment identification. We reported a simple and convenient "preoperative positive staining technique" for laparoscopic ALR to overcome some limitations. To our knowledge, this is the first report of robotic-assisted surgery in which ALR was performed using this technique. MATERIALS AND SURGICAL TECHNIQUE: A 69-year-old man presented with a 12-mm HCC in segment 8. Preoperative three-dimensional simulation images showed that the fourth-order branch of the portal vein was a tumor-bearing portal pedicle. After anesthesia induction, 1 mL of 0.025 mg/mL ICG was injected percutaneously into this branch under B-mode ultrasound guidance before pneumoperitoneum. A robotic laparoscope was inserted. The preoperative positive staining area was clearly stained on the liver surface with the Firefly mode on the da Vinci Xi system. Based on the demarcation line, the liver parenchymal resection was started. The ICG fluorescence staining area was checked frequently on the resected side of the liver transection plane. Subsequently, the fourth-order portal branch was identified with the ICG fluorescence technique and ligated. Finally, the specimen was resected. The operation took 352 min, with 10 mL of blood loss, and was completed without any operative problems. DISCUSSION: Although many cases are required, the proposed preoperative positive staining technique appears useful for accurate and precise surgery given the increasing application of robotic-assisted hepatectomy.

[Research progress and prospects of the application of radiographic imaging in the precise diagnosis and treatment of hepatocellular carcinoma].

Hepatocellular carcinoma (HCC) is a highly heterogeneous kind of malignant tumor with a high recurrence rate and low five-year survival rate, which has become one of the major public health issues in China. Currently, HCC is the only solid tumor that can be solely diagnosed based on epidemiological history and typical imaging features without preoperative pathological confirmation. The paradigm for HCC imaging diagnosis has shifted in recent years from anatomy to function, from macroscopic to microscopic, and from diagnosis to prediction in the context of precision medicine, making it possible to study the microscopic processes such as HCC genes and their metabolic laws from the perspective of qualitative and quantitative imaging, thereby providing more accurate biological and imaging information for elucidating the occurrence, development, and clinical treatment decisions of HCC.This paper reviews the research progress of HCC imaging in recent years, demonstrating the rapid horizontal development and enormous potential of imaging in the vertical follow-up of HCC precision diagnosis and treatment. Simultaneously, it also puts forward the shortcomings of current HCC imaging research and looks forward to future development directions in order to be more accurately used in clinical decision support systems.

[Research progress of radiomics in hepatocellular carcinoma].

Primary liver cancer is a common malignant digestive system tumor, with hepatocellular carcinoma being the most common pathological type. Radiomics significantly boosts the efficiency of predictions by accurately capturing the intrinsically heterogeneous features of tumors that are difficult to discern with the human eye in imaging images. This article outlines the background and concepts of radiomics, introduces its latest research progress in various aspects, such as diagnosis and differential diagnosis, prediction of pathological molecular subtypes, efficacy evaluation, and survival prediction, and further discusses its limitations and prospects in HCC.

[Application and prospects of magnetic resonance imaging techniques in the diagnosis and evaluation of hepatocellular carcinoma].

Hepatocellular carcinoma (HCC) is the most common primary malignant tumor of the liver. MRI has become an important imaging method for non-invasive diagnosis and evaluation of HCC in clinics because of its advantageous aspects, such as its non-radiative nature, superior detection, and qualitative accuracy over CT and ultrasound. Various MRI techniques, including hepatobiliary-specific contrast agents, magnetic resonance elastography, diffusion-weighted imaging, and others, can diagnose HCC or evaluate its malignant biological behavior from different dimensions such as blood supply, cell function, tissue hardness, and water molecule diffusion. This article introduces the current status and prospects of various MRI techniques for HCC diagnosis and evaluation.

[Comparative study of ultra-high field diffusion-weighted MRI imaging between hepatocellular carcinoma and paracancerous, distant cancerous, and background liver tissues].

Objective: To investigate the differences in multi-b-value apparent diffusion coefficient (ADC) and exponential apparent diffusion coefficient (eADC) between hepatocellular carcinoma (HCC) and paracancerous liver tissue, distant cancerous liver tissue, and background liver tissues by ultra-high field 3.0T diffusion-weighted (DWI) MRI imaging. Methods: Sixty-eight consecutive HCC cases confirmed by surgical pathology from January 2018 to October 2021 were enrolled and divided into a cirrhosis (n=39) and a non-cirrhosis group (n=29) according to the presence or absence of cirrhosis.The average ADC and eADC of liver tissues of paracancerous (including proximal and distal), distant cancerous, and background were measured by DWI images with diffusion sensitivity factors (b) of 50, 100, 400, 600 s/mm2, and 1 000 s/mm2, respectively. The Kruskal-Wallis H test and Bonferroni method were used to test the differences between the measured values of the five tissues. The statistical differences were used to evaluate the diagnostic efficacy of the five tissues by parametric receiver operating characteristic (ROC) curve and area under the curve (AUC). Results: The comparison of average ADC and eADC among five types of tissues in the liver cirrhosis group showed that the average ADC and eADC measured at b values of 50, 100, 400, and 600 s/mm2 had statistically significant differences (adjusted P<0.005) between cancerous and proximal paracancerous, distal paracancerous, distant cancerous, and background liver tissue, as well as the average ADC measured at b=1 000 s/mm2 between cancerous and proximal paracancerous tissue. The average ADC and eADC in the non-cirrhosis group had statistically significant differences (adjusted P<0.005) between cancerous and proximal paracancerous, distant paracancerous, distant cancerous, and background liver tissue measured at b values of 50, 100, and 400 s/mm2, respectively. The average ADC and eADC measured at b=600 s/mm2 showed statistically significant differences (adjusted P<0.005) between cancerous and proximal paracancerous, distal paracancerous, and distant cancerous liver tissue, as well as the average ADC measured at b=1 000 s/mm2 between cancerous and distal paracancerous, and distant cancerous liver tissue. The average ADC and eADC in the cirrhosis group had no statistically significant difference between the proximal paracancerous and the distant cancerous, as well as the background liver tissue measured at b-values of 50, 100, 400, 600, and 1 000 s/mm2, respectively (adjusted P>0.005), while there were statistically significant differences (adjusted P<0.005) in the average ADC values in the non-cirrhosis group between the proximal paracancerous and the distant paracancerous and background liver tissues at b=50 s/mm2, as well as the average ADC and eADC values between the proximal paracancerous and the distant liver tissues at b=100 s/mm2. The average ADC and eADC values measured in the cirrhosis group and non-cirrhosis group had no statistically significant difference between the distant paracancerous, distant cancerous, and background liver tissue (adjusted P>0.005). The efficacy of average ADC and eADC in distinguishing five types of tissues (cancerous and proximal paracancerous, distant paracancerous, distant cancerous, and background liver tissue) showed that in the cirrhosis group, the diagnostic efficacy was best at b=50 s/mm2. The area under the ROC curve (AUC) of average ADC was 0.815, 0.828, 0.855, and 0.855, respectively, and the AUC of average eADC was 0.815, 0.830, 0.856, and 0.855, respectively. The diagnostic efficacy was best in the non cirrhosis group at b=100 s/mm2, with average ADC AUCs of 0.787, 0.823, 0.841, and 0.821, and average eADC AUCs of 0.836, 0.874, 0.893, and 0.873, respectively. The AUC of the average ADC in the non-cirrhosis group for distinguishing between proximal paracancerous and distant cancerous liver tissues, as well as proximal paracancerous and background liver tissues, with b=50 s/mm2, were 0.605 and 0.604, respectively. The average AUC of ADC and eADC for distinguishing between proximal paracancerous and distant liver tissues with b=100 s/mm2 were 0.619 and 0.620, respectively. Conclusion: The average ADC and eADC measured by multiple b-values are helpful in distinguishing HCC from proximal paracancerous, distal paracancerous, distant-cancerous, and background liver tissues in patients with cirrhosis and non-cirrhosis, while the average ADC and eADC at b=50 s/mm2 and 100 s/mm2 exhibit differences between the proximal paracancerous from the distant cancerous liver tissue and background liver tissue in patients with non-cirrhosis.

The reproducibility and predictivity of radiomic features extracted from dynamic contrast-enhanced computed tomography of hepatocellular carcinoma.

PURPOSE: To assess the reproducibility of radiomic features (RFs) extracted from dynamic contrast-enhanced computed tomography (DCE-CT) scans of patients diagnosed with hepatocellular carcinoma (HCC) with regards to inter-observer variability and acquisition timing after contrast injection. The predictive ability of reproducible RFs for differentiating between the degrees of HCC differentiation is also investigated. METHODS: We analyzed a set of DCE-CT scans of 39 patients diagnosed with HCC. Two radiologists independently segmented the scans, and RFs were extracted from each sequence of the DCE-CT scans. The same lesion was segmented across the DCE-CT sequences of each patient's scan. From each lesion, 127 commonly used RFs were extracted. The reproducibility of RFs was assessed with regard to (i) inter-observer variability, by evaluating the reproducibility of RFs between the two radiologists; and (ii) timing of acquisition following contrast injection (inter- and intra-imaging phase). The reproducibility of RFs was assessed using the concordance correlation coefficient (CCC), with a cut-off value of 0.90. Reproducible RFs were used for building XGBoost classification models for the differentiation of HCC differentiation. RESULTS: Inter-observer analyses across the different contrast-enhancement phases showed that the number of reproducible RFs was 29 (22.8%), 52 (40.9%), and 36 (28.3%) for the non-contrast enhanced, late arterial, and portal venous phases, respectively. Intra- and inter-sequence analyses revealed that the number of reproducible RFs ranged between 1 (0.8%) and 47 (37%), inversely related with time interval between the sequences. XGBoost algorithms built using reproducible RFs in each phase were found to be high predictive ability of the degree of HCC tumor differentiation. CONCLUSIONS: The reproducibility of many RFs was significantly impacted by inter-observer variability, and a larger number of RFs were impacted by the difference in the time of acquisition after contrast injection. Our findings highlight the need for quality assessment to ensure that scans are analyzed in the same physiologic imaging phase in quantitative imaging studies, or that phase-wide reproducible RFs are selected. Overall, the study emphasizes the importance of reproducibility and quality control when using RFs as biomarkers for clinical applications.

Aggregation-induced emission: Application in diagnosis and therapy of hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is a serious health issue due to its low early diagnosis rate, resistance to chemotherapy, and poor five-year survival rate. Therefore, it is crucial to explore novel diagnostic and therapeutic approaches tailored to the characteristics of HCC. Aggregation-induced emission (AIE) is a phenomenon where the luminescence of certain molecules, typically non-luminescent or weakly luminescent in solution, is significantly enhanced upon aggregation. AIE has been extensively applied in bioimaging, biosensors, and therapy. Fluorophore materials based on AIE (AIEgens) have a wide range of application scenarios and potential for clinical translation. This review focuses on recent advances in AIE-based strategies for diagnosing and treating HCC. First, the specific functional mechanism of AIE is described. Next, we summarize recent progress in the application of AIE for multimodal imaging, biosensor detection, and phototherapy. Finally, prospects and challenges for the AIE-based application in the diagnosis and therapy of HCC are discussed.

CT-based radiomics nomogram to predict proliferative hepatocellular carcinoma and explore the tumor microenvironment.

BACKGROUND: Proliferative hepatocellular carcinomas (HCCs) is a class of aggressive tumors with poor prognosis. We aimed to construct a computed tomography (CT)-based radiomics nomogram to predict proliferative HCC, stratify clinical outcomes and explore the tumor microenvironment. METHODS: Patients with pathologically diagnosed HCC following a hepatectomy were retrospectively collected from two medical centers. A CT-based radiomics nomogram incorporating radiomics model and clinicoradiological features to predict proliferative HCC was constructed using the training cohort (n = 184), and validated using an internal test cohort (n = 80) and an external test cohort (n = 89). The predictive performance of the nomogram for clinical outcomes was evaluated for HCC patients who underwent surgery (n = 201) or received transarterial chemoembolization (TACE, n = 104). RNA sequencing data and histological tissue slides from The Cancer Imaging Archive database were used to perform transcriptomics and pathomics analysis. RESULTS: The areas under the receiver operating characteristic curve of the radiomics nomogram to predict proliferative HCC were 0.84, 0.87, and 0.85 in the training, internal test, and external test cohorts, respectively. The radiomics nomogram could stratify early recurrence-free survivals in the surgery outcome cohort (hazard ratio [HR] = 2.25; P < 0.001) and progression-free survivals in the TACE outcome cohort (HR = 2.21; P = 0.03). Transcriptomics and pathomics analysis indicated that the radiomics nomogram was associated with carbon metabolism, immune cells infiltration, TP53 mutation, and heterogeneity of tumor cells. CONCLUSION: The CT-based radiomics nomogram could predict proliferative HCC, stratify clinical outcomes, and measure a pro-tumor microenvironment.

Artificial intelligence for ultrasonographic detection and diagnosis of hepatocellular carcinoma and cholangiocarcinoma.

The effectiveness of ultrasonography (USG) in liver cancer screening is partly constrained by the operator's expertise. We aimed to develop and evaluate an AI-assisted system for detecting and classifying focal liver lesions (FLLs) from USG images. This retrospective study incorporated 26,288 USG images from 5444 patients to train YOLOv5 model for FLLs detection and classification of seven different types of FLLs, including hepatocellular carcinoma (HCC), cholangiocarcinoma (CCA), focal fatty infiltration, focal fatty sparing (FFS), cyst, hemangioma, and regenerative nodules. AI model performance was assessed for detection and diagnosis of the FLLs on a per-image and per-lesion basis. The AI achieved an overall FLLs detection rate of 84.8% (95%CI:83.3-86.4), with consistent performance for FLLs ≤ 1 cm and > 1 cm. It also exhibited sensitivity and specificity for distinguishing malignant FLLs from other benign FLLs at 97.0% (95%CI:95. 9-98.2) and 97.0% (95%CI:95.9-98.1), respectively. Among specific FLL types, CCA detection rate was at 92.2% (95%CI:88.0-96.4), followed by FFS at 89.7% (95%CI:87.1-92.3), and HCC at 82.3% (95%CI:77.1-87.5). The specificities and NPVs for regenerative nodules were 100% and 99.9% (95%CI:99.8-100.0), respectively. Our AI model can potentially assist physicians in FLLs detection and diagnosis during USG examinations. Further external validation is needed for clinical application.