Advancements in computed tomography (CT) technology, particularly the emergence of dual-energy CT (DE-CT) and photon-counting detector CT (PCD-CT), can improve detection, characterization, and treatment monitoring of focal liver lesions. DE-CT, through its ability to differentiate tissues with similar densities and produce diverse datasets, has enhanced lesion visibility and diagnostic precision. PCD-CT further advances imaging with superior spatial resolution and material decomposition capabilities, offering potential for complex diagnostic scenarios. This review aimed to highlight the role of CT in hepatic imaging and its application to focal liver lesions. DE-CT improves lesion detectability using low-energy virtual monochromatic images, which enhance iodine contrast and reduce radiation and contrast agent doses. It also facilitates treatment response evaluation after locoregional therapies for hepatocellular carcinoma by quantifying biomarkers, such as the extracellular volume fraction. This review underscores the transformative impact of DE-CT and PCD-CT on liver imaging, emphasizing their complementary roles alongside magnetic resonance imaging. These innovations have paved the way for more precise diagnostics, improved treatment planning, and enhanced patient outcomes in the management of liver diseases.

The integration of artificial intelligence (AI) into clinical practice is limited due to the "black box" nature of deep learning models that do not provide interpretable reasoning. While explainable AI (XAI) methods, such as Grad-CAM, LIME, and SHAP, attempt to visualize decision-related image regions, they frequently lack clinical alignment, limiting their adoption in radiology. To fill in this gap, we propose Explainable AI that provides decision reasons in the Domain Language of Radiologists (XAI-DL), a novel framework designed to replicate radiologists' diagnostic reasoning for hepatocellular carcinoma (HCC) diagnosis in liver MRI. XAI-DL integrates the Liver Imaging Reporting and Data System (LI-RADS) into the AI pipeline, enabling transparent predictions articulated in clinically meaningful terms. The framework consists of modules of tumor segmentation, feature encoding, discriminative image feature gating, LI-RADS feature significance optimization, malignancy likelihood prediction, and reason answering. Evaluated on 291 MRI studies from 105 patients, XAI-DL achieved an average AUC of 0.799 for LI-RADS feature prediction and an AUC of 0.835 for tumor classification, outperforming state-of-the-art deep learning models including VGGNet (0.751) and ResNet (0.727). Beyond this high accuracy, the model provides LI-RADS-based explanations that mirror clinical workflows, allowing radiologists to understand not only the diagnostic outcome but also the reasoning behind it. This structured interpretability bridges the gap between AI algorithms and radiologists' decision making, fostering greater trust and integration into diagnostic workflows. By aligning AI decision-making with domain-specific standards, XAI-DL demonstrates the potential to improve diagnostic accuracy, enhance clinical interpretability, and support radiologists in evidence-based liver cancer diagnosis.

BACKGROUND:The classification of oncological resectability for hepatocellular carcinoma (HCC) has been established, requiring validation of treatment outcomes for hepatectomy and systemic chemotherapy. METHODS:The study evaluated treatment outcomes in 978 patients who underwent hepatectomy and 222 patients with HCC who received first-line systemic chemotherapy (atezolizumab plus bevacizumab, lenvatinib, or durvalumab plus tremelimumab). RESULTS:Among three factors defining patients with borderline resectable 1 (BR1) and 2 (BR2), macrovascular invasion factor was associated with significantly worse prognosis in a hepatectomy group (BR1: 34.2 vs. 63.4 months, p = 0.04; BR2: 14.4 vs. 20.9 months, p = 0.004). In contrast, in the systemic chemotherapy group, none of the three factors affected prognosis in either BR1 or BR2 patients. In BR2 patients undergoing hepatectomy, those with a single risk factor had significantly better outcomes than those with 2-3 factors (20.1 vs. 12.6 months, p < 0.001). Similarly, in the entire systemic chemotherapy cohort, patients with a single risk factor had better outcomes than those with 2-3 (22.6 vs. 11.9 months, p = 0.001). However, among chemotherapy responders (per modified Response Evaluation Criteria in Solid Tumors), prognosis did not significantly differ between those with one factor and those with 2-3 factors (25.4 vs. 24.5 months, p = 0.502). CONCLUSION:Macrovascular invasion significantly impacted prognosis in patients undergoing hepatectomy, for both BR1 and BR2, whereas any of the tumor factors did not affect the prognosis of patients receiving systemic chemotherapy. Tumor burden correlated with prognosis in the entire cohort but not in chemotherapy responders, suggesting effective treatment may overcome poor prognostic indicators.

Background Technetium-99m-diethylenetriamine-penta-acetic acid-galactosyl human serum albumin (99mTc-GSA) scintigraphy is a useful method for assessing liver function and its heterogeneity. We evaluated its predictive usefulness for post-hepatectomy liver failure (PHLF) in liver resection for biliary tract cancer. Methods Between 2013 and 2024, 95 patients underwent major hepatectomy with bile duct resection for biliary tract cancer and had preoperative 99mTc-GSA scintigraphy. The GSA-K value was defined using established reduction formulas of indocyanine green plasma clearance rate (ICG-K) values based on LHL15 value (99mTc-GSA uptake ratio of the liver to the liver plus heart at 15 min) from 99mTc-GSA scintigraphy. Functional remnant liver volume (f-RLV) ratio was estimated by dividing scintillation counts of the future remnant liver by total counts of the whole liver. We compared volumetric (ICG-Krem: ICG-K x RLV ratio) and functional (GSA-Krem: GSA-K x f-RLV ratio) assessment methods of the future remnant liver for predicting PHLF. Results PHLF was observed in 34 patients (35.8%) and a receiver operating characteristic curve revealed cut-off values for predicting PHLF of 0.088 for ICG-Krem and 0.086 for GSA-Krem. Positive predictive values for PHLF were 53.2% (ICG-Krem) and 87.6% (GSA-Krem). In patients with portal vein embolization (PVE), GSA-Krem showed 89.1% of positive predictive value for PHLF, and ICG-Krem was very low (53.3%). The multivariate analysis identified GSA-Krem as one of independent predictors of PHLF. Conclusions GSA-Krem derived from 99mTc-GSA scintigraphy is a valuable predictor of PHLF.

To investigate the frequency of visually ill-demarcated small pancreatic ductal adenocarcinomas (PDACs) and compare their clinicopathologic and MRI characteristics with those of well-demarcated tumors. This multicenter retrospective study enrolled 210 patients with surgically confirmed small PDACs (≤ 20 mm) who underwent preoperative unenhanced MRI. Clinical and pathological data were collected. Two radiologists independently evaluated the following MRI features: (a) tumor delineation (well-demarcated or ill-demarcated), (b) tumor signal intensity, and (c) secondary signs, including distal parenchymal signal intensity, parenchymal atrophy, main pancreatic duct (MPD) stenosis and dilatation, and small retention cyst. Clinicopathologic features and MRI findings were compared between well-demarcated and ill-demarcated PDACs using the Mann–Whitney and Fisher’s exact tests. Multivariate logistic regression analysis was performed to identify independent factors associated with ill-demarcated PDAC. Visually ill-demarcated tumors accounted for 46.7

This study aimed to compare image quality between photon-counting detector computed tomography (PCD-CT) and third-generation dual-source energy-integrating detector CT (EID-CT) in multiphasic contrast-enhanced abdominal imaging using intraindividual analysis. This retrospective study included 88 patients who underwent both arterial phase (AP) and portal venous phase (PVP) abdominal CT with PCD-CT and EID-CT. Virtual monoenergetic images (VMIs) from 50 to 90 keV were reconstructed in PCD-CT, and 120 kVp images were applied in EID-CT. Quantitative image analysis evaluated CT attenuation, image noise, and contrast-to-noise ratio (CNR) in abdominal organs and vessels. Qualitative image quality, including vessel and parenchymal contrast, noise, and overall image quality, was assessed by two radiologists using a 5-point Likert scale. Compared with EID-CT, VMIs at 50–70 keV obtained from PCD-CT demonstrated significantly higher CT attenuation values (P < 0.001) across all evaluated abdominal organs and vessels. CNRs at 70 keV were significantly greater than those in EID-CT for all objects (P < 0.001), with increases of 13.3‒41.6% on AP and 40.4‒49.3% on PVP. Image noise in 70 keV VMI was lower than EID-CT (P < 0.001). Optimal image quality was achieved using VMIs at 60–70 keV, which achieved significantly higher qualitative scores (P < 0.001) than those from EID-CT. Substantial interobserver agreement was observed (κ = 0.60–0.79). The CT dose volume index was slightly lower with PCD-CT compared with EID-CT (P < 0.001), whereas dose-length product values remained similar. PCD-CT significantly improved image quality in multiphasic abdominal CT compared with third-generation dual-source EID-CT under equivalent radiation dose conditions. The superior iodine enhancement and higher CNR in VMIs at 60–70 keV highlight the advantages of PCD-CT.

Background Deep learning (DL) models have shown promise in diagnosing pancreatic cancer (PC); however, models that simultaneously detect both direct and indirect imaging findings associated with PC are lacking. Purpose To develop and evaluate DL models that detect direct and indirect imaging findings on noncontrast CT (NCCT) and contrast-enhanced CT (CECT) images for PC diagnosis. Materials and Methods This retrospective study from August 2007 to December 2022 included patients with PC and control patients. Two DL models were developed using NCCT and CECT to detect direct (pancreatic mass) and indirect (parenchymal atrophy, main pancreatic duct [MPD] dilatation, and MPD stenosis) imaging findings and diagnose PC based on these findings. For training and validation, CT scans from multiple institutions were manually annotated. Model evaluation was performed using two external test sets (CECT and NCCT sets). Receiver operating characteristic curve analysis was used to assess diagnostic performance. Model performance in detecting imaging findings and PC was compared with the performance of six physicians. The reference standard for PC diagnosis was histopathologic confirmation. Results This study included 2251 patients (mean age, 66 years ± 13.3 [SD]; age range, 20-96 years; 850 men). DL models demonstrated area under the receiver operating characteristic curve (AUC) values of 0.94, 0.90, 0.94, and 0.94 in the CECT set and 0.88, 0.88, 0.95, and 0.93 in the NCCT set for detecting pancreatic masses, parenchymal atrophy, MPD dilatation, and MPD stenosis, respectively. For PC diagnosis, DL models performed similarly to or better than the mean of six readers in the CECT (AUC, 0.99 vs 0.99; P = .84) and NCCT (AUC, 0.93 vs 0.91; P = .03) sets. For PCs that were 20 mm or smaller, the DL models demonstrated higher sensitivity than the reader mean in both the CECT (98% vs 82.6%; P < .001) and NCCT (86% vs 41.1%; P < .001) sets. Conclusion DL models detected direct and indirect imaging findings on CT images and diagnosed PC with performance comparable to or better than that of physicians, particularly for small PCs. © RSNA, 2026 Supplemental material is available for this article. See also the editorial by Bhayana and Rajpurkar in this issue.

Although digital subtraction angiography remains the gold standard for the diagnosis and treatment planning of intracranial arteriovenous shunts (AVS), including arteriovenous malformations and arteriovenous fistulas, non-invasive imaging is increasingly sought for comprehensive evaluation. Magnetic resonance imaging (MRI) plays a crucial role in AVS detection, identification of feeding arteries and draining veins, localization of shunt points, classification of subtypes, assessment of venous reflux and congestion, and evaluation of post-treatment residual or recurrent lesions. Clinical techniques such as time-of-flight MR angiography (MRA), contrast-enhanced time-resolved MRA, susceptibility-weighted imaging, and arterial spin labeling (ASL) are established for these assessments. Recent advances have expanded MRI capabilities: Ultrashort echo time MRA can overcome turbulent flow-related signal loss and susceptibility artifacts, improving visualization of complex nidus architecture; compressed sensing substantially accelerates three-dimensional and four-dimensional (4D)-MRA while maintaining diagnostic quality; ASL-based 4D-MRA provides high-temporal-resolution dynamic evaluation without contrast, with vessel-selective techniques enabling independent assessment of individual vascular territories; and high-resolution vessel wall imaging shows promise for risk stratification. Emerging artificial intelligence applications enable automated AVS segmentation and characterization, with potential to enhance image quality and reduce scan times. This review summarizes current MRI techniques, recent innovations, and future perspectives in the non-invasive assessment of intracranial AVS.

Purpose: To compare accelerated T2-weighted turbo spin-echo imaging with deep learning reconstruction (DLR-TSE) with conventional T2-weighted TSE (conv-TSE) and accelerated TSE without DLR (non-DLR-TSE), and to evaluate image quality and diagnostic performance of Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1)-based T2 scoring in prostate MRI. Methods: This single-center retrospective study included 60 patients who underwent prostate MRI with all 3 T2-weighted image sets acquired in the same examination. Qualitative image quality was independently assessed by 2 radiologists using 6 parameters on a 5-point Likert scale. Quantitative metrics included apparent SNR (aSNR), apparent contrast-to-noise ratio (aCNR), and contrast ratio (CR). Diagnostic performance of PI-RADS T2 scores for transition zone lesions was evaluated using receiver operating characteristic (ROC) analysis, sensitivity, specificity, and accuracy. Noninferiority of DLR-TSE relative to conv-TSE was tested with predefined margins. Inter-reader agreement was assessed using weighted kappa statistics. Results: DLR-TSE demonstrated noninferiority to conv-TSE for all qualitative parameters and quantitative metrics for both readers. Both DLR-TSE and conv-TSE showed significantly higher image quality scores and quantitative values than non-DLR-TSE. For PI-RADS T2 scoring, DLR-TSE achieved diagnostic performance comparable to conv-TSE. For reader 1, the area under the ROC curve (AUC) was identical for DLR-TSE and conv-TSE (0.83; 95%CI 0.69-0.95), and significantly higher than for non-DLR-TSE (0.70; 95%CI 0.56-0.83). Specificity and overall accuracy were markedly reduced with non-DLR-TSE for both readers, whereas sensitivity did not differ significantly among methods. Inter-reader agreement was substantial to almost perfect for DLR-TSE and conv-TSE, and lower for non-DLR-TSE. DLR-TSE reduced acquisition time by approximately 60% compared with conv-TSE. Conclusion: Accelerated T2-weighted imaging with DLR allows substantial reduction in scan time while maintaining image quality, diagnostic performance, and inter-reader agreement comparable to those of conventional T2-weighted imaging. DLR-TSE may serve as a practical option for improving examination efficiency in clinical prostate MRI.

Thermal ablation (TA), including microwave ablation, radiofrequency ablation, and cryoablation, is increasingly used as a surgical alternative for T1a renal cell carcinoma (RCC), particularly in lesions measuring ≤ 3 cm. In this review, we aimed to summarize current evidence on the efficacy, safety, and long-term outcomes of TA in RCC and to explore future directions in this evolving field. A comprehensive review of the literature was conducted, focusing on the indications for TA, considering tumor size and location, patient age, and comorbidity status (assessed using Charlson Comorbidity Index). Recent studies have reported favorable long-term outcomes of TA for RCC with follow-up exceeding 10 years. Among the different TA modalities, no significant differences in therapeutic efficacy or safety were observed, particularly for tumors measuring ≤ 3 cm. Compared to partial nephrectomy, the standard treatment for small renal tumors, TA offers advantages in preserving renal function and reducing the risk of complications; however, it is associated with a higher local recurrence rate. Nevertheless, favorable cancer-specific survival rates comparable to those with surgical resection have been achieved with additional TA for local residual tumors. Adjunctive techniques, including preprocedural transarterial embolization and hydrodissection, may further improve the safety and efficacy of TA. In the future, introducing advanced technologies, including robotic systems and artificial intelligence, is expected to further enhance TA’s safety and efficacy in RCC management.

BACKGROUND:Signal intensity on the hepatobiliary phase of gadolinium-ethoxybenzyl-diethylenetriamine (Gd-EOB-DTPA)-magnetic resonance imaging (MRI) reflects Wnt/β-catenin signaling activity in hepatocellular carcinoma (HCC), and has been associated with poor response to immune monotherapy. However, whether such imaging findings predict resistance to combination immunotherapy has not been prospectively validated. METHODS:This multicenter prospective study enrolled 152 patients with unresectable HCC treated with atezolizumab plus bevacizumab across 12 Japanese centers. All had Child-Pugh class A liver function and underwent Gd-EOB-DTPA-MRI prior to treatment. Hyperintensity was defined as a relative enhancement ratio ≥ 0.9. Treatment outcomes, including objective response rate (ORR), progression-free survival (PFS), overall survival (OS), and time to partial response (TtPR), were compared between patients with and without hyperintense nodules. RESULTS:Of the 152 patients, 82 received immunotherapy as the first-line and 70 as a later-line therapy. Hyperintense nodules were identified in 57 (37.5%) patients. The hyperintense group showed a higher ORR (36.8% vs. 22.1%) and comparable PFS (8.9 vs. 7.9 months) and OS (16.7 vs. 23.9 months), though not statistically significant. The TtPR was significantly shorter in the hyperintense group (median 26.0 months vs. not reached, p = 0.033), although mainly influenced by prior systemic therapy. Overall, hyperintense lesions were not associated with reduced efficacy and tended to show more favorable responses. CONCLUSION:This prospective multicenter study demonstrates that hepatobiliary hyperintensity on Gd-EOB-DTPA-enhanced MRI-an imaging surrogate of Wnt/β-catenin activation-does not indicate resistance to atezolizumab plus bevacizumab. These findings support the use of combination immunotherapy in molecularly "immune-cold" HCC.

BACKGROUND/AIM:This study evaluated the clinical significance of baseline lymphocyte-to-monocyte ratio (LMR) and its early dynamic changes as an on-treatment biomarker in patients with advanced hepatocellular carcinoma (HCC) receiving atezolizumab plus bevacizumab (Ate/Bev), with stratification by alpha-fetoprotein (AFP) status. PATIENTS AND METHODS:We retrospectively reviewed 108 patients with advanced HCC treated with first-line Ate/Bev. Baseline LMR and LMR at six weeks (LMR 6w) were calculated. Patients were classified into high/low groups (cutoff: 3.69) and dynamic change groups [increased (Up) vs. decreased (Down)]. Overall survival (OS) and objective response rate (ORR) were assessed according to baseline AFP levels (cutoff: 20 ng/ml). RESULTS:High baseline LMR significantly correlated with longer OS (median not reached vs. 17.3 months, p<0.001). Notably, patients with increased LMR at six weeks (Up group) demonstrated significantly superior OS compared to the Down group (33.6 vs. 18.9 months, p=0.018) and a higher ORR (46.0% vs. 26.0%, p=0.037). The High+Up cohort achieved the most favorable prognosis. In stratified analyses, these prognostic and predictive values of early LMR dynamics were prominently observed in AFP-negative patients (p<0.05), but were attenuated in AFP-positive individuals. CONCLUSION:Early dynamic increase in LMR serves as a powerful, cost-effective on-treatment biomarker for Ate/Bev therapy in advanced HCC. Integrating systemic immune dynamics with tumor biology provides superior risk stratification, particularly for AFP-negative patients.

OBJECTIVES:To identify clinical and MRI features useful for diagnosing placenta accreta spectrum (PAS) in non-previa placenta and to develop diagnostic models integrating these features. METHODS:This retrospective study included 101 pregnant women with non-previa placenta who underwent MRI between January 2022 and June 2024. Nineteen were confirmed as PAS. Clinical variables and 11 MRI findings were evaluated using intraoperative or pathological results as the reference standard. Diagnostic performance was assessed using univariable analysis and repeated cross-validation of a random forest (RF) model, with ROC analysis used to assess discriminative performance. RESULTS:Hormone replacement cycle-frozen embryo transfer (HRC-FET) (sensitivity 0.89, specificity 0.63) and abnormal placental bed vascularization (sensitivity 0.63, specificity 0.90) showed the strongest univariable performance. The RF model using six variables with acceptable interobserver agreement achieved an AUC of 0.88, sensitivity 0.92, specificity 0.79, demonstrating higher discriminative performance than individual predictors. Feature importance analysis highlighted HRC-FET and abnormal placental bed vascularization as the most influential factors. CONCLUSIONS:Integrating clinical and MRI features improves PAS diagnosis in non-previa placenta. The RF model demonstrated a more balanced diagnostic profile than individual predictors in this exploratory cohort and may aid preoperative risk assessment. HRC-FET and abnormal placental bed vascularization were key contributors, supporting their relevance for risk stratification.

Sphincter of Oddi dysfunction (SOD) can cause unexplained biliary pain and idiopathic pancreatitis. Although Rome IV criteria recommend sphincter of Oddi manometry (SOM) for diagnosis, SOM is invasive and carries pancreatitis risk. We hypothesized that cine-dynamic magnetic resonance cholangiopancreatography (MRCP) could non-invasively visualize bile and pancreatic juice flow, enabling functional papillary assessment. In this prospective observational study, 40 participants were enrolled, and 29 were included in the final analysis after excluding 11 participants who did not meet the Rome IV criteria (10 healthy controls, 7 with suspected biliary-type SOD [BSOD], and 12 with suspected pancreatic-type SOD [PSOD]). Cine-dynamic MRCP was performed with 20 sequential frames over 5 min. Two quantitative indices were assessed: flow frequency and secretion grade (distance traveled by bile or pancreatic juice). Bile flow frequency and secretion grade were significantly lower in both BSOD and PSOD than in controls: frequency (median [range], 13.5 [6–19] in controls vs. 2.0 [1–17] in BSOD, p = 0.006; vs. 8.0 [3–14] in PSOD, p = 0.008) and secretion grade (1.6 [0.3–2.05] in controls vs. 0.2 [0.1–1.3] in BSOD, p = 0.001; vs. 0.5 [0.15–1.75] in PSOD, p = 0.03). Pancreatic juice flow showed no significant difference between BSOD and controls but was significantly reduced in PSOD: frequency (16 [14–19] in controls vs. 9.5 [4–17] in PSOD, p < 0.001) and secretion grade (2.15 [0.7–3.25] in controls vs. 0.98 [0.25–2.9] in PSOD, p = 0.003). Cine-dynamic MRCP parameters improved after sphincterotomy in six patients. Cine-dynamic MRCP enables non-invasive visualization and quantification of bile and pancreatic juice flow, providing functional assessment of the sphincter of Oddi.

BACKGROUND/OBJECTIVES:Trace metals, including copper (Cu) and zinc, are associated with the development and prognosis of hepatocellular carcinoma (HCC). However, their interference with magnetic resonance imaging (MRI) limits their use as potential biomarkers. This study investigated the usefulness of Synchrotron Radiation-excited X-ray Fluorescence (SR-XRF) imaging in studying the distribution of trace metals in HCC. METHODS:This case-control study analyzed 33 specimens from 32 patients with HCC who underwent surgical resection (n = 29) or biopsy (n = 3) at Kobe University Hospital between December 1999 and November 2002. The findings of SR-XRF were compared with those of MRI and histopathology. RESULTS:SR-XRF provided two-dimensional mapping of trace metal distribution with high spatial resolution (1.0 µm). The mean tumor-to-liver ratio (TLR) of Cu content was significantly higher in well-differentiated HCCs than in moderately and poorly differentiated HCCs (p < 0.05). Moreover, the mean TLRs of Cu content were significantly higher in high-intensity lesions than in iso- or low-intensity lesions on T1-weighted imaging (p < 0.05). CONCLUSIONS:This study supports previous evidence of the involvement of Cu in HCC development, suggesting its potential as a clinical biomarker for diagnosis and disease progression. Additionally, the results demonstrate that SR-XRF has potential for clinical application due to its ability to map trace metal distribution at high resolution. These findings suggest, rather than demonstrate, the association among Cu accumulation, tumor differentiation, and MRI signal characteristics.

BACKGROUND:Focal nodular hyperplasia (FNH)-like lesions are hyperplastic formations in patients with micronodular cirrhosis and a history of alcohol abuse. Although pathologically similar to hepatocellular carcinoma (HCC) lesions, they are benign. As such, it is important to develop methods to distinguish between FNH-like lesions and HCC. AIM:To evaluate diagnostically differential radiological findings between FNH-like lesions and HCC. METHODS:We studied pathologically confirmed FNH-like lesions in 13 patients with alcoholic cirrhosis [10 men and 3 women; mean age: 54.5 ± 12.5 (33-72) years] who were negative for hepatitis-B surface antigen and hepatitis-C virus antibody and underwent dynamic computed tomography (CT) and magnetic resonance imaging (MRI), including superparamagnetic iron oxide (SPIO) and/or gadoxetic acid-enhanced MRI. Seven patients also underwent angiography-assisted CT. RESULTS:The evaluated lesion features included arterial enhancement pattern, washout appearance (low density compared with that of surrounding liver parenchyma), signal intensity on T1-weighted image (T1WI) and T2-weighted image (T2WI), central scar presence, chemical shift on in- and out-of-phase images, and uptake pattern on gadoxetic acid-enhanced MRI hepatobiliary phase and SPIO-enhanced MRI. Eleven patients had multiple small lesions (< 1.5 cm). Radiological features of FNH-like lesions included hypervascularity despite small lesions, lack of "corona-like" enhancement in the late phase on CT during hepatic angiography (CTHA), high-intensity on T1WI, slightly high- or iso-intensity on T2WI, no signal decrease in out-of-phase images, and complete SPIO uptake or incomplete/partial uptake of gadoxetic acid. Pathologically, similar to HCC, FNH-like lesions showed many unpaired arteries and sinusoidal capillarization. CONCLUSION:Overall, the present study showed that FNH-like lesions have unique radiological findings useful for differential diagnosis. Specifically, SPIO- and/or gadoxetic acid-enhanced MRI and CTHA features might facilitate differential diagnosis of FNH-like lesions and HCC.

Liver cancer remains a significant global health concern, ranking as the sixth most common malignancy and the third leading cause of cancer-related deaths worldwide. Medical imaging plays a vital role in managing liver tumors, particularly hepatocellular carcinoma (HCC) and metastatic lesions. However, the large volume and complexity of imaging data can make accurate and efficient interpretation challenging. Artificial intelligence (AI) is recognized as a promising tool to address these challenges. Therefore, this review aims to explore the recent advances in AI applications in liver tumor imaging, focusing on key areas such as image reconstruction, image quality enhancement, lesion detection, tumor characterization, segmentation, and radiomics. Among these, AI-based image reconstruction has already been widely integrated into clinical workflows, helping to enhance image quality while reducing radiation exposure. While the adoption of AI-assisted diagnostic tools in liver imaging has lagged behind other fields, such as chest imaging, recent developments are driving their increasing integration into clinical practice. In the future, AI is expected to play a central role in various aspects of liver cancer care, including comprehensive image analysis, treatment planning, response evaluation, and prognosis prediction. This review offers a comprehensive overview of the status and prospects of AI applications in liver tumor imaging.

The objective of this article is to provide a comprehensive overview of the imaging characteristics of various renal cell tumors using 18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET), based on the latest WHO-2022 classification. Due to the physiological accumulation of FDG in the kidneys, the clinical utility of FDG-PET in the evaluation of renal tumors has traditionally been considered limited. However, recent studies have re-evaluated its potential value. FDG-PET has demonstrated particular utility in detecting metastases and postoperative recurrence of renal cell carcinoma (RCC), as well as in identifying RCC in patients with chronic kidney failure, where FDG excretion into the urinary tract is reduced. Renal tumors are occasionally detected incidentally on FDG-PET, and FDG uptake varies depending on the tumor subtype. Therefore, a comprehensive understanding of these imaging characteristics is clinically important, as it may serve as a valuable guide for subsequent diagnostic evaluations. Furthermore, recent advancements in the development of novel PET tracers hold promise for future applications in the imaging of renal tumors. We believe that the insights gained from this study will contribute to routine diagnostic practice and the planning of future research.

Background/Objectives: Although immunotherapy is the primary treatment option for intermediate-stage hepatocellular carcinoma (HCC), its efficacy varies. This study aimed to identify non-invasive imaging biomarkers predictive of the immunoscore linked to dynamic contrast-enhanced computed tomography (CECT). Methods: We performed immunohistochemical staining with CD3+ and CD8+ antibodies and counted the positive cells in the invasive margin (IM) and central tumor (CT), converting them to an immunoscore of 0 to 4 points. We assessed the dynamic CECT findings obtained from 96 patients who underwent hepatectomy for HCC and evaluated the relationship between dynamic CECT findings and immunoscores. For validation, we assessed the treatment effects on 81 nodules using the Response Evaluation Criteria in Solid Tumors in another cohort of 41 patients who received combined immunotherapy with atezolizumab and bevacizumab (n = 27) and durvalumab and tremelizumab (n = 14). Results: HCCs with peritumoral enhancement in the arterial phase (p < 0.001) and rim APHE (p = 0.009) were associated with the immunoscore in univariate linear regression analysis and peritumoral enhancement in the arterial phase (p = 0.004) in multivariate linear regression analysis. The time to nodular progression in HCCs with peritumoral enhancement in the arterial phase was significantly longer than that in HCCs without this feature (p < 0.001). Conclusions: We identified HCCs with peritumoral enhancement in the arterial phase as a noninvasive imaging biomarker to predict immune-inflamed HCC with a high immunoscore tendency. These HCCs were most likely to respond to combined immunotherapy.