ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

EGFR-mutant non-small cell lung cancer (NSCLC) is a common type of lung cancer, with EGFR tyrosine kinase inhibitors (EGFR-TKIs) being the standard first-line treatment. However, most patients with NSCLC eventually develop resistance to EGFR-TKIs. Studies on the mechanism underlying EGFR-TKI resistance have driven the development of personalized and precision medicine. Current strategies to address resistance include targeted therapy, immunotherapy, and novel drug treatments. Selecting the appropriate personalized treatment plan is crucial for improving the survival rate and quality of life of patients with EGFR-mutant NSCLC. Thus, this study provides a brief review of the current status and future perspectives in the treatment of EGFR-mutant NSCLC after progression on EGFR-TKI therapy.

ObjectivePrimary liver cancer, predominantly hepatocellular carcinoma (HCC), is a significant global health issue, ranking as the sixth most diagnosed cancer and the third leading cause of cancer-related mortality. Accurate and early diagnosis of HCC is crucial for effective treatment, as HCC and non-HCC malignancies like intrahepatic cholangiocarcinoma (ICC) exhibit different prognoses and treatment responses. Traditional diagnostic methods, including liver biopsy and contrast-enhanced ultrasound (CEUS), face limitations in applicability and objectivity. The primary objective of this study was to develop an advanced, light-weighted classification network capable of distinguishing HCC from other non-HCC malignancies by leveraging the automatic analysis of brightness changes in CEUS images. The ultimate goal was to create a user-friendly and cost-efficient computer-aided diagnostic tool that could assist radiologists in making more accurate and efficient clinical decisions. MethodsThis retrospective study encompassed a total of 161 patients, comprising 131 diagnosed with HCC and 30 with non-HCC malignancies. To achieve accurate tumor detection, the YOLOX network was employed to identify the region of interest (ROI) on both B-mode ultrasound and CEUS images. A custom-developed algorithm was then utilized to extract brightness change curves from the tumor and adjacent liver parenchyma regions within the CEUS images. These curves provided critical data for the subsequent analysis and classification process. To analyze the extracted brightness change curves and classify the malignancies, we developed and compared several models. These included one-dimensional convolutional neural networks (1D-ResNet, 1D-ConvNeXt, and 1D-CNN), as well as traditional machine-learning methods such as support vector machine (SVM), ensemble learning (EL), k-nearest neighbor (KNN), and decision tree (DT). The diagnostic performance of each method in distinguishing HCC from non-HCC malignancies was rigorously evaluated using four key metrics: area under the receiver operating characteristic (AUC), accuracy (ACC), sensitivity (SE), and specificity (SP). ResultsThe evaluation of the machine-learning methods revealed AUC values of 0.70 for SVM, 0.56 for ensemble learning, 0.63 for KNN, and 0.72 for the decision tree. These results indicated moderate to fair performance in classifying the malignancies based on the brightness change curves. In contrast, the deep learning models demonstrated significantly higher AUCs, with 1D-ResNet achieving an AUC of 0.72, 1D-ConvNeXt reaching 0.82, and 1D-CNN obtaining the highest AUC of 0.84. Moreover, under the five-fold cross-validation scheme, the 1D-CNN model outperformed other models in both accuracy and specificity. Specifically, it achieved accuracy improvements of 3.8% to 10.0% and specificity enhancements of 6.6% to 43.3% over competing approaches. The superior performance of the 1D-CNN model highlighted its potential as a powerful tool for accurate classification. ConclusionThe 1D-CNN model proved to be the most effective in differentiating HCC from non-HCC malignancies, surpassing both traditional machine-learning methods and other deep learning models. This study successfully developed a user-friendly and cost-efficient computer-aided diagnostic solution that would significantly enhances radiologists’ diagnostic capabilities. By improving the accuracy and efficiency of clinical decision-making, this tool has the potential to positively impact patient care and outcomes. Future work may focus on further refining the model and exploring its integration with multimodal ultrasound data to maximize its accuracy and applicability.

Purpose To evaluate the efficacy of ultra-high field 5.0T MRI combined with histogram analysis for diagnosing prostate cancer and benign prostatic hyperplasia.Materials and Methods This retrospective analysis included data from 63 patients with prostatic diseases at the Chongqing University Three Gorges Hospital from January to May 2024,comprising 31 cases of prostate cancer and 32 cases of benign prostatic hyperplasia.MRI sequences included T2WI,T1WI,diffusion-weighted imaging,intravoxel incoherent motion and T2 mapping.Histogram data of apparent diffusion coefficient,true diffusion coefficient,pseudo-diffusion coefficient,perfusion fraction and T2 relaxation time were calculated,and diagnostic efficacy was assessed using the area under the receiver operating characteristic curve.Results In prostate cancer,the 10th percentile,the 90th percentile,mean,median,and minimum values of histogram parameters from apparent diffusion coefficient,true diffusion coefficient,pseudo-diffusion coefficient,perfusion fraction and T2 mapping were significantly lower than those of benign prostatic hyperplasia(Z=-6.036--3.368,all P<0.05).Notably,the combined model of apparent diffusion coefficient,intravoxel incoherent motion and T2 mapping parameters achieved an the area under the curve of 0.987,with sensitivity and specificity of 96.77%and 96.87%,respectively.Conclusion This study confirms that 5.0T MRI histogram analysis technique demonstrates significant diagnostic efficacy in differentiating prostate cancer from benign prostatic hyperplasia.

Sarcopenia is an age-related skeletal muscle degenerative disease.Physiologically aging mice are the most commonly used animal model for studying sarcopenia.As sarcopenia is characterized by decreased skeletal muscle mass and reduced muscle strength,exercise performance as a reflection of muscle function is widely used to evaluate sarcopenia.Methods of evaluating the motor function of sarcopenia mouse models are generally designed based on muscle endurance,muscle strength,coordination,and balance.The method include tests such as treadmill exhaustion,voluntary wheel use,grip strength,horse grid,bars and balance beam tests.By collating recent publications,we have systematically summarized the method used for evaluating motor function,including the tests'principles,procedures,evaluation indexes,advantages,and disadvantages.We then propose an operational program for evaluating the sarcopenia phenotype,which will be of help to researchers wishing to choose evaluation method appropriate to their specific research purposes.Further innovative technology for assessing motor function that could be instructive in the evaluation of skeletal muscle function and diagnosis of sarcopenia is summarized.

Objective To explore the technical process and clinical application of laparoscopic combined upper midline incision minimally invasive liver transplantation. Methods A retrospective analysis was conducted on 30 cases of laparoscopic combined upper midline incision minimally invasive liver transplantation. The cases were divided into cirrhosis group (15 cases) and liver failure group (15 cases) based on the primary disease. The surgical and postoperative conditions of the two groups were compared. Results All patients successfully underwent laparoscopic "clockwise" liver resection, with no cases of passive conversion to open surgery or intolerance to pneumoperitoneum. In 6 cases, the right lobe was relatively large, and the right hepatic ligaments could not be completely mobilized. One case required an additional reverse "L" incision during open surgery. All patients successfully completed the liver transplantation, with no major intraoperative bleeding, cardiovascular events, or other occurrences in the 30 patients. The model for end-stage liver disease (MELD) score in the cirrhosis group was lower than that in the liver failure group (P<0.001). There were no statistically significant differences between the two groups in terms of age, surgical time, blood loss, anhepatic phase, or cold ischemia time (all P>0.05). During the perioperative period, there was 1 case of hepatic artery embolism, 1 case of portal vein anastomotic stenosis, no complications of hepatic vein and inferior vena cava, and 3 cases of biliary anastomotic stenosis, all of which occurred in the liver failure group. Conclusions In strictly selected cases, the minimally invasive liver transplantation technique combining laparoscopic hepatectomy with upper midline incision for graft implantation has the advantages of smaller incisions, less bleeding, relatively easier operation, and faster postoperative recovery, which is worthy of clinical promotion and application.

Objective To explore research hotspots and future development trends in radioactive iodine refractory differentiated thyroid carcinoma (RAIR-DTC) treatment from 2004 to 2024. Methods Literature on RAIR-DTC treatment published from January 2004 to May 2024 was retrieved from the Web of Science (WOS) database. CiteSpace, VOSviewer, and Microsoft Office Excel were used for visual analysis of publication volume, countries, institutions, authors, keywords, and co-citation networks. Results A total of 677 articles were included in the analysis. National and institutional co-occurrence analysis revealed that the United States, along with the MD Anderson Cancer Center at the University of Texas, was the most productive and influential in this field. Author and citation co-occurrence analysis highlighted the substantial contributions of Schlumberger M and Brose MS to the field. The exploration of high-frequency keywords and keyword clustering indicated tyrosine kinase inhibitors and disease prognostic factors were current research hotspots. Keyword burst analysis suggested that future research trends would focus on optimizing clinical benefits through reliable data provided from high-quality clinical trials and achieving personalized, precise treatment management. Conclusion Targeted drugs hold remarkable potential for RAIR-DTC treatment, and emphasizing predictive factors for disease prognosis offers valuable guidance for medical practice.

Early correction of childhood malocclusion is timely managing morphological, structural, and functional abnormalities at different dentomaxillofacial developmental stages. The selection of appropriate imaging examination and comprehensive radiological diagnosis and analysis play an important role in early correction of childhood malocclusion. This expert consensus is a collaborative effort by multidisciplinary experts in dentistry across the nation based on the current clinical evidence, aiming to provide general guidance on appropriate imaging examination selection, comprehensive and accurate imaging assessment for early orthodontic treatment patients.
Humans ; Malocclusion/diagnostic imaging* ; Child ; Consensus