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 investigate the effects of subanesthetic dose of esketamine on postoperative anxiety and recovery in patients undergoing laparoscopic cholecystectomy. METHODS A total of 200 patients scheduled for laparoscopic cholecystectomy at Suzhou Ninth Hospital Affiliated to Soochow University from January 2023 to December 2024 were randomly assigned to control group （n＝100） and observation group （n＝100）. One minute before the initiation of anesthesia， patients in the control group received intravenous injections of Propofol emulsion injection， Sufentanil citrate injection， and Succinylcholine chloride injection. On this basis， patients in the observation group received an intravenous injection of Esketamine hydrochloride injection. The anxiety status of patients in both groups was compared， along with their general intraoperative conditions （including sufentanil dosage， duration of pneumoperitoneum， operative time， anesthesia time， and extubation time）， postoperative recovery， incidence of adverse reactions， and the need for dezocine rescue analgesia. Heart rate and mean arterial pressure， entropy index （state entropy and response entropy）， inflammatory marker levels [interleukin-6 （IL-6） and C-reactive protein （CRP）]， numerical rating scale （NRS） for pain intensity were compared between the two groups at different time points. RESULTS No significant differences were found between the two groups in pneumoperitoneum duration， operative time， anesthesia time，extubation time， incidence of postoperative dry mouth， entropy index or length of stay in the post-anesthesia care unit （P＞0.05）. Compared with the control group， the observation group showed significantly lower postoperative STAI-S scores， reduced intraoperative sufentanil consumption， decreased incidence of postoperative nausea， vomiting， and shivering， the need for dezocine rescue analgesia， as well as lower plasma IL-6 and CRP levels at 24 h after surgery， and NRS （P＜0.05）. The heart rate and mean arterial pressure of patients in the observation group at the start of surgery， end of surgery， and during extubation were all significantly higher than those in the control group （P＜0.05）. CONCLUSIONS Subanesthetic dose of esketamine can effectively alleviate postoperative anxiety， reduce intraoperative opioid consumption， suppress postoperative inflammatory response， relieve postoperative pain， and promote recovery in patients undergoing laparoscopic cholecystectomy.

ObjectiveTo evaluate the therapeutic effects of modified Banxia Xiexintang（MBXT） on polycystic ovary syndrome with insulin resistance（PCOS-IR） rats and reveal its potential mechanisms based on the integrated analysis of transcriptomics and metabolomics. MethodsFemale SD rats were selected， and a PCOS-IR model was established by intragastric administration of letrozole combined with a high-fat diet for 21 days. The modeled rats were randomly divided into the model group， MBXT low-， medium-， and high-dose groups（6.62， 13.23， 26.46 g·kg^-1）， and metformin group（0.158 g·kg^-1）， with a normal group set up separately. After 14 days of administration， the estrous cycle was observed， ovarian morphology was examined by hematoxylin-eosin（HE） staining， and the levels of testosterone（T）， estradiol（E₂）， follicle-stimulating hormone（FSH）， and luteinizing hormone（LH） in serum were detected by enzyme-linked immunosorbent assay（ELISA）. Serum metabolites and ovarian tissue gene expression were detected using ultra-performance liquid chromatography-quadrupole-electrostatic orbitrap mass spectrometry（UPLC-Q-Orbitrap-MS） and RNA-Seq technology， respectively， followed by multi-omics integrated analysis. ResultsPharmacodynamic findings revealed that all MBXT dose groups could reversed abnormal estrous cycles in PCOS-IR rats， improve polycystic ovarian lesions， and normalize dysregulated serum hormone levels（T， LH， E₂， FS， P<0.05， P<0.01）. Metabolomic analysis revealed that compared with the model group， MBXT reversed 278 differential metabolites such as estrone and S-formylglutathione， mainly involving pathways such as steroid hormone biosynthesis， glutathione metabolism， and lipid peroxidation regulation. Transcriptomic analysis identified 434 differentially expressed genes， and enrichment analysis revealed that MBXT significantly regulated lipid peroxidation defense systems， including glutathione metabolism， peroxisome function， and fatty acid metabolism， thereby intervening in ferroptosis processes. It also engaged in inflammation-related pathways such as the chemokine signaling pathway. Integrated analysis revealed that both metabolomics and transcriptomics co-enriched metabolic pathways associated with ferroptosis and fatty acid metabolism. And key Hub genes［such as Ras-related C3 botulinum toxin substrate 2 gene（Rac2） and Fas ligand gene（Faslg）］ showed significant correlations with differential metabolites. ConclusionMBXT can effectively ameliorate reproductive dysfunction and metabolic disorders in PCOS-IR rats. Its mechanism may be related to remodeling the immune-metabolism network， particularly by regulating MHC-mediated immune responses， inhibiting local ovarian ferroptosis， and enhancing steroid hormone synthesis pathways.

ObjectiveTo investigate the mechanism of modified Banxia Xiexintang（MBXT） in improving ovarian dysfunction in polycystic ovary syndrome with insulin resistance（PCOS-IR） rats by inhibiting ferroptosis through the adenosine monophosphate（AMP）-activated protein kinase（AMPK）/fatty acid synthase（FASN）/glutathione peroxidase 4（GPX4） signaling pathway. MethodsSeventy-six female SD rats were randomly divided into a normal group（n=13） and a modeling group（n=63）. The modeling group established a PCOS-IR model by intragastric administration of letrozole combined with a high-fat diet for 21 days. After successful modeling， these rats were randomly divided into the model group， MBXT low-， medium-， and high-dose groups（6.62， 13.23， 26.46 g·kg^-1）， metformin group（0.158 g·kg^-1）， and high-dose of MBXT combined with ferroptosis inducer Erastin group（15 mg·kg^-1）， with 10 rats in each group. After 14 days of intervention， ovarian pathological morphology was observed by hematoxylin-eosin（HE） staining， the mitochondrial ultrastructure of granulosa cells was observed by transmission electron microscopy（TEM）， ovarian reactive oxygen species（ROS） levels were detected by dihydroethidium（DHE） probe， biochemical methods were used to detect Fe²⁺， malondialdehyde（MDA）， glutathione（GSH） and other indicators in ovarian tissues， serum sex hormone and insulin levels were measured by enzyme-linked immunosorbent assay（ELISA）， and the protein expressions of AMPK， FASN， acyl-CoA synthetase long-chain family member 4（ACSL4）， GPX4， and solute carrier family 7 member 11（SLC7A11） in ovarian tissues were detected by Western blot. ResultsCompared with the normal group， the model group showed polycystic changes in the ovaries， with atrophy of mitochondria in granulosa cells and increased membrane density. Serum levels of testosterone（T）， luteinizing hormone（LH）， and insulin were significantly increased（P<0.01）. The levels of ROS， MDA， 4-hydroxynonenal（4-HNE）， and Fe²⁺ in ovarian tissues were significantly elevated（P<0.01）， while adenosine triphosphate（ATP）， GSH， and reduced nicotinamide adenine dinucleotide phosphate （NADPH） levels were significantly decreased（P<0.01）. The phosphorylation levels of AMPK and acetyl-CoA carboxylase （ACC）， as well as the protein expressions of SLC7A11， GPX4， and ferroptosis suppressor protein 1（FSP1） were significantly downregulated（P<0.01）， whereas the expressions of FASN， ACSL4， and nuclear receptor coactivator 4（NCOA4） were significantly upregulated（P<0.01）. Compared with the model group， MBXT intervention at various doses improved the above pathological changes and biochemical indicators in a dose-dependent manner， with the high-dose group showing the most significant effect（P<0.01）. Compared with the MBXT high-dose group， the high-dose of MBXT combined with ferroptosis inducer Erastin group restored ovarian ferroptosis characteristics in rats， with increased ROS and lipid peroxidation products， and altered expressions of key proteins（P<0.05， P<0.01）. ConclusionMBXT can effectively improve ovarian function and metabolic disorders in PCOS-IR rats. Its mechanism may be related to activating the AMPK/ACC signaling pathway， downregulating FASN and ACSL4 to reduce lipid peroxidation substrates， and restoring glucose-6-phosphate dehydrogenase/phosphoglycerate dehydrogenase（G6PD/PHGDH） metabolic flux to enhance the GPX4/FSP1 antioxidant defense system， thereby inhibiting ferroptosis in ovarian granulosa cells.

We report a case of a female farmer from the Yichang suburb, aged 68 years, who had been suffering from the infection for 3 years and had a history of seeking medical treatment at a community hospital and receiving blood transfusions. Her hemoglobin was as low as 28 g/L during this visit, and she had been ill for a long time with severe anemia. Routine stool examination revealed hookworm eggs. After active treatment, the patient recovered and was discharged from the hospital.

To formulate an expert consensus on the principles of preoperative hair removal and provide scientific guidance for standardized removal of hair before surgical procedures so as to reduce the incidence of surgical site infections.METHODS Led by the Hospital Management Institute of National Health Commission of the People's Republic of China,this consensus was reached with the joint efforts from the expects of relevant fields such as surgeries,interventional therapies,nursing,and infection prevention and control.The consensus facilitates the classification and evaluation of literatures by following the evidence grade formulated by Oxford Evidence-based Medicine Center and focuses on the association of preoperative hair removal with surgical site infection,it reaches the evidence grade of expert consensus and recommendation intensity by integrating with discussions on meetings and clinical experience of the expects from relevant fields.RESULTS A total of 6 items of consensus were reached by summarizing the latest evidence on the aspects including the indications for preoperative hair removal,tools,range,timing and places.CONCLUSION The consensus,to some extent,make supplements to and complete the exiting regulations and standards.It provides guidance for the medical institutions to carry out the preoperative hair removal.

Objective:Currently, there is a lack of targeted training programs for resident physicians in cardiopulmonary resuscitation (CPR) in China. This study aims to establish a stratified training indicator system for resident physicians in CPR that is in line with the medical realities in China.Methods:Expert consultation questionnaires were designed through literature review, group discussions, and surveys. Two rounds of expert consultation were conducted using the modified Delphi method to develop the stratified training indicator system for resident physicians in CPR. Statistical analysis was performed using SPSS 26.0 software and Kendall's coefficient of concordance was used to assess the consistency of expert opinions.Results:A total of 15 experts participated in two rounds of questionnaire consultation, with a positive coefficient of 100.00% and an authority coefficient of 0.91, indicating high expert engagement and authority, thus ensuring the credibility of the results. The coefficient of variation for expert ratings in the first round of questionnaire consultation ranged from 0 to 0.40, with a concordance coefficient of 0.22 ( P<0.001). In the second round of questionnaire consultation, the coefficient of variation for expert ratings ranged from 0 to 0.24, with a concordance coefficient of 0.25 ( P<0.001). The Kendall's coefficient of concordance for expert opinions showed statistically significant differences in both rounds of questionnaire consultation, indicating consensus among experts and reliable results. This system consisted of 6 primary indicators and 32 secondary indicators. Conclusions:Based on the modified Delphi method, this study established a stratified training indicator system for resident physicians in CPR. In the future, based on the research findings, CPR training programs adapted to the medical realities in China will be developed and promoted to validate their scientific and practical value.

On the occasion of the 110th Anniversary of the establishment of the Chinese Medical Association, the third conference of the Fourth Editorial Board of Chinese Journal of Digestive Surgery, and the Second Elite Group of Chinese Journal of Digestive Surgery, was successfully held in Kunming on July 4, 2025. This conference systematically summarizes the development experience of the journal over the past 20 years from three aspects: the role of ecological construction of thought in the discipline construction of digestive surgery, the display of the latest academic achievements in the field of digestive surgery, the development difficulties, and breakthrough paths of the discipline, and strategically plans the path of discipline construction in the new era.

Objective:To establish a TaqMan-MGB probe-based method for detecting the polymorphic loci C677T and A1298C of the MTHFR gene.Methods:Specific primers and TaqMan-MGB probes targeting the C677T and A1298C polymorphic loci of the MTHFR gene were designed and optimized based on the gene sequence information.A real-time quantitative PCR detection system was established.Gradient dilution experiments were conducted to determine the limit of detection,and reproducibility experiments were performed to evaluate detection consistency.Specificity was validated using wild-type and mutant plasmid templates.The method was applied to detect 56 clinical samples,and its accuracy and practicality were assessed through comparison with traditional Sanger sequencing.Results:The TaqMan-MGB probe method demonstrated high specificity for detecting the C677T and A1298C loci,with no cross-reactivity between wild-type and mutant probes,enabling accurate genotype differentiation.Sensitivity experiments revealed detection limits of 1.13 × 103 copies/μL for C677T and 8.39 × 101 copies/μL for A1298C.Reproducibility experiments showed coefficients of variation below 1％,indicating stable and reliable results.Among the 56 clinical samples,the overall detection rate for the C677T locus was 86.99％,and for the A1298C locus,it was 97.92％.The TaqMan-MGB method exhibited good concordance with Sanger sequencing results.Conclusion:The TaqMan-MGB method exhibits high specificity,sensitivity,and excellent reproducibility in detecting the polymorphic loci C677T and A1298C of the MTHFR gene,making it suitable for rapid detection in large-scale clinical samples.This method provides an effective molecular diagnostic tool for the early diagnosis and prevention of folate-related diseases.