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.

Objective:To analyze the baseline characteristics and prognostic factors of Pneumocystis carinii pneumonia (PCP) in patients with liver diseases (LD), and to develop the nomogram prognosis model. Methods:The patients with liver diseases complicated with PCP (LD-PCP) who admitted to Beijing Youan Hospital, Capital Medical University from January 1st, 2021 to October 31st, 2024 were enrolled in this study. This study was a retrospective case-control study. Data were collected, including baseline characteristics at admission and clinical outcomes. Multivariate logistic regression analysis was employed to identify the independent risk factors of poor prognosis in LD-PCP patients. Subsequently, a nomogram was developed to construct the prognostic model. The performance of this model was evaluated using the receiver operater characteristic (ROC) curve and calibration curve.Results:A total of 61 patients with LD-PCP were enrolled, including 21 in the death group and 40 in the survival group. Multivariate logistic regression analysis revealed that co-infection with Aspergillus and elevated lactate dehydrogenase (LDH) levels were independent risk factors for mortality in patients with LD-PCP (odds ratio ( OR)=12.802, 95% confidence interval ( CI) 1.334 to 122.845, P=0.027; OR=1.009, 95% CI 1.002 to 1.016, P=0.014). The predictive value of prognostic model based on the nomogram was better evidenced by the ROC curve, which yielded an area under the curve (AUC) of 0.835 (95% CI 0.725 to 0.944), with the sensitivity of 71.4% and the specificity of 87.5%. The predictive accuracy of the model was further validated by the calibration curve, and the Brier value was 0.151, the model fitting line had a high degree of agreement with the standard reference line ( P=0.953). Conclusions:Co-infection with Aspergillus and LDH levels are independent risk factors for the poor prognosis of patients with LD-PCP. The prognosis model based on the nomogram has better predictive value and clinical practicality, which could identify high-risk patients in the early stage, and provide theoretical guidance on the patient management.

Body weight abnormalities, including overweight, obesity, and underweight, have become a dual public health challenge in Chinese adults: overweight and obesity lead to a variety of chronic complications, while underweight increases the risks of malnutrition, sarcopenia, and organ dysfunction. To systematically address these issues, multidisciplinary experts in endocrinology, sports science, nutrition, and psychiatry from various regions have held multiple weight management seminars. Based on the latest epidemiological data and clinical evidence, they expanded the guideline to include assessment and intervention strategies for underweight, in addition to the core content of obesity management. This guideline outlines the etiological mechanisms, evaluation methods, and multidimensional management strategies for overweight and obesity, covering key areas such as diagnosis and assessment, medical nutrition therapy, exercise prescription, pharmacological intervention, and psychological support. It is intended to provide a scientific and standardized approach to weight management across the adult population, aiming to curb the rising prevalence of obesity, mitigate complications associated with abnormal body weight, and improve nutritional status and overall quality of life.

OBJECTIVE: To examine the effectiveness of Chinese medicine (CM) Lianhua Qingwen Granule (LHQW) and Jingyin Gubiao Prescription (JYGB) in asymptomatic or mild patients with Omicron infection in the shelter hospital. METHODS: This single-center retrospective cohort study was conducted in the largest shelter hospital in Shanghai, China, from April 10, 2022 to May 30, 2022. A total of 56,244 asymptomatic and mild Omicron cases were included and divided into 4 groups, i.e., non-administration group (23,702 cases), LHQW group (11,576 cases), JYGB group (12,112 cases), and dual combination of LHQW and JYGB group (8,854 cases). The length of stay (LOS) in the hospital was used to assess the effectiveness of LHQW and JYGB treatment on Omicron infection. RESULTS: Patients aged 41-60 years, with nadir threshold cycle (C_T) value of N gene <25, or those fully vaccinated preferred to receive CM therapy. Before or after propensity score matching (PSM), the multiple linear regression showed that LHQW and JYGB treatment were independent influence factors of LOS (both P<0.001). After PSM, there were significant differences in LOS between the LHQW/JYGB combination and the other groups (P<0.01). The results of factorial design ANOVA proved that the LHQW/JYGB combination therapy synergistically shortened LOS (P=0.032). CONCLUSIONS Patients with a nadir C_T value <25 were more likely to accept CM. The LHQW/JYGB combination therapy could shorten the LOS of Omicron-infected individuals in an isolated environment.
Humans ; Drugs, Chinese Herbal/therapeutic use* ; Male ; Female ; Middle Aged ; Adult ; China/epidemiology* ; Hospitalization ; COVID-19 Drug Treatment ; COVID-19/epidemiology* ; SARS-CoV-2 ; Retrospective Studies ; Treatment Outcome ; Length of Stay ; Young Adult ; Aged

Objective To evaluate the diagnostic value of transrectal contrast-enhanced ultrasound (CEUS) for rectal cancer with intestinal stenosis caused by tumors. Methods Forty-nine patients with rectal cancer underwent transrectal CEUS and magnetic resonance imaging (MRI) before surgery.Intraoperative tumor localization and postoperative pathological results were taken as the gold standard for diagnosis.The differences in T stage,localization,and tumor length of rectal cancer were compared between the two methods. Results The total accuracy rates of transrectal CEUS and MRI in diagnosing T stage were 75.5% (36/49) and 67.3% (33/49),which had no significant difference (χ²=0.8,P=0.371).The total accuracy rates of transrectal CEUS and MRI in judging tumor localization were 79.5% (39/49) and 77.5% (38/49),which had no significant difference (χ²=0.061,P=0.806).The measurement results of tumor length in pathological examination had no significant difference from the transrectal CEUS results (t=1.42,P=0.162) but a significant difference from the MRI results (t=3.38,P=0.001).Furthermore,transrectal CEUS detected 8 (16.3%) cases of colonic polyps among the 49 patients,while MRI did not detect colon lesions. Conclusions Transrectal CEUS has good consistency with MRI in T staging and localization judgement of rectal cancer with intestinal stenosis,and this method can more accurately evaluate the tumor length and simultaneously evaluate whether there is a lesion in the entire colon at the proximal end of stenosis.It can be used as a supplementary examination before rectal cancer treatment in clinical practice.
Humans ; Rectal Neoplasms/complications* ; Male ; Middle Aged ; Female ; Aged ; Contrast Media ; Ultrasonography ; Adult ; Magnetic Resonance Imaging ; Constriction, Pathologic/diagnostic imaging* ; Aged, 80 and over ; Intestinal Obstruction/etiology*

Objective To evaluate the diagnostic value of intestinal ultrasound (IUS) for organic lesions in the intestines of patients with chronic abdominal pain or diarrhea. Methods The IUS signs in 263 patients with chronic abdominal pain or diarrhea were retrospectively analyzed.With the endoscopic examination results as the gold standard,comparison was performed for the IUS signs between the groups with positive and negative endoscopic results,as well as between the inflammatory bowel disease group and the non-specific intestinal inflammation group of positive cases.Furthermore,the detection rates of IUS in different intestinal segments were analyzed to evaluate the accuracy of IUS in the diagnosis and localization of intestinal lesions. Results Among the 263 patients,194 (73.8%) and 69 (26.2%) patients were in the groups with positive and negative endoscopic results,respectively.The diagnosis sensitivity,specificity,and accuracy of IUS were 82.0%,71.0%,and 79.1%,respectively.The proportions of positive IUS signs in the group with positive endoscopic results were higher than that in the group with negative endoscopic results (all P<0.001).The proportions of positive IUS signs in the inflammatory bowel disease group were higher than those in the non-specific bowel inflammation group (all P<0.001).When the lesion was located in the ileum,ileocecal region,and colon,IUS demonstrated good consistency with endoscopic results in locating the lesion (kappa=0.642,0.686,and 0.601,respectively),with sensitivity and specificity of 82.7% (95%CI=75.4%-88.6%) and 81.5% (95%CI=73.5%-87.9%),73.7% (95%CI=62.3%-83.1%) and 93.0% (95%CI=88.4%-96.2%),and 68.9% (95%CI=58.3%-78.2%) and 89.6% (95%CI=84.1%-93.7%),respectively. Conclusions IUS can be used for screening the patients with chronic abdominal pain or diarrhea to detect organic lesions in the intestines. Moreover,it can effectively locate the affected intestinal segment,which is helpful for the monitoring and follow-up of intestinal diseases.
Humans ; Diarrhea/diagnostic imaging* ; Female ; Male ; Abdominal Pain/diagnostic imaging* ; Middle Aged ; Adult ; Ultrasonography ; Retrospective Studies ; Aged ; Young Adult ; Intestines/diagnostic imaging* ; Adolescent ; Chronic Disease ; Sensitivity and Specificity ; Aged, 80 and over

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.