OBJECTIVE To construct a predictive model for evaluating the efficacy of a triple antiemetic regimen （neurokinin- 1 receptor antagonist+5-hydroxytryptamine 3 receptor antagonist+dexamethasone） for preventing nausea and vomiting induced by highly emetogenic chemotherapy （HEC） based on interpretable deep learning algorithms. METHODS Clinical data of cancer patients who received HEC and were treated with the standard triple antiemetic regimen in the oncology department of Tianjin Medical University General Hospital from January 2018 to December 2022 were collected retrospectively. Demographic， clinical and metabolism-related variables were integrated. After data pre-processing， two deep learning algorithms （deep random forest and dense neural network） and four machine learning algorithms （support vector machine， categorical boosting， random forest and decision tree） were used to build predictive models. Subsequently， model performance evaluation and model interpretability analysis were conducted. RESULTS Among the six candidate models， the deep random forest model demonstrated the best predictive performance on the test set， with an area under the receiver operating characteristic curve of 0.850， an accuracy of 0.911， a precision of 0.805， a recall of 0.783， an F1 score of 0.793， and a Brier score of 0.075. Interpretability analysis revealed that creatinine clearance rate （Ccr） was the key predictive factor， and low Ccr levels， female gender， younger age， highly emetogenic drugs （particularly cisplatin-containing chemotherapy regimens）， and anticipatory nausea and vomiting were positively correlated with the risk of HEC-related nausea and vomiting. CONCLUSIONS The deep random forest model exhibits the best performance in predicting the efficacy of triple antiemetic regimen for preventing HEC-related nausea and vomiting. The key predictors in this model primarily include Ccr，anticipatory nausea and vomiting， gender， age， and highly emetogenic drugs.

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 effect of manual acupuncture manipulations (MAMs) on subcutaneous muscle tissue, by developing quantitative models of "lifting and thrusting" and "twisting and rotating", based on machine learning techniques. METHODS: A depth camera was used to capture the acupuncture operator's hand movements during "lifting and thrusting" and "twisting and rotating" of needle. Simultaneously, the ultrasound imaging was employed to record the muscle tissue responses of the participants. Amplitude and angular features were extracted from the movement data of operators, and muscle fascicle slope features were derived from the data of ultrasound images. The dynamic time warping barycenter averaging algorithm was adopted to align the dual-source data. Various machine learning techniques were applied to build quantitative models, and the performance of each model was compared. The most optimal model was further analyzed for its interpretability. RESULTS: Among the quantitative models built for the two types of MAMs, the random forest model demonstrated the best performance. For the quantitative model of the "lifting and thrusting" technique, the coefficient of determination (R²) was 0.825. For the "twisting and rotating" technique, R² reached 0.872. CONCLUSION Machine learning can be used to effectively develop the models and quantify the effects of MAMs on subcutaneous muscle tissue. It provides a new perspective to understand the mechanism of acupuncture therapy and lays a foundation for optimizing acupuncture technology and designing personalized treatment regimen in the future.
Humans ; Acupuncture Therapy/methods* ; Machine Learning ; Male ; Adult ; Female ; Subcutaneous Tissue/diagnostic imaging* ; Young Adult

Objective To explore the applicative value of spectral CT in increasing positive rates of lung cancer puncture and reducing complications during CT guided percutaneous transthoracic needle biopsy(PTNB).Methods The pathological results and complica-tion incidences of 260 PTNB patients were analyzed retrospectively.All patients were divided into three groups:group A(conventional CT group,103 cases)used a scheme based on conventional enhanced CT;group B(PET/CT group,84 cases)used a scheme combining the maximum standardized uptake value(SUVmax)with conventional enhanced CT;group C(spectral CT group,73 cases)used a scheme of quantitative spectral CT parameters and images.Results Group A included 103 cases in total,of which 87 were positive(84.47％),41 pneumothorax(39.81％),and 31 hemorrhage(30.10％).Group B totaled 84 cases,including 82 positive cases(97.62％),19 cases of pneumothorax(22.62％),and 11 cases of hemorrhage(13.10％).Group C was of 73 cases,including 70 positive cases(95.89％),16 cases of pneumothorax(21.92％),and 10 cases of hemorrhage(13.70％).There were statistically significant differ-ences in biopsy positive rates,pneumothorax incidences,and hemorrhage incidences among groups A,B,and C(P＜0.05).There were also statistically significant differences in biopsy positive rates,pneumothorax incidences,and hemorrhage incidences between groups A and B or groups A and C(P＜0.016 7),respectively.However,no statistically significant differences were found between groups B and C in biopsy positive rates,pneumothorax incidences,and hemorrhage incidences(P＞0.016 7).Conclusion Spectral CT can improve the positive rate of lung cancer and reduce the risk of pneumothorax and hemorrhage with PTNB.

Objective To characterize the features of patent foramen ovale(PFO)-related cryptogenic stroke using diffusion-weighted imaging(DWI)and to investigate the correlation between infarct size and cardiac CT characteristics of PFO.Methods A retrospective,consecutive cohort study was conducted on acute ischemic stroke patients admitted to Neurology Department of Xuanwu Hospital,Capital Medical University from January 2022 to September 2024.Patients were categorized into PFO group,arterio-arterial embolism(AAE)group,and atrial fibrillation(AF)group based on etiological diagnosis.Baseline clinical data,including age,height,body mass index,admission National Institutes of Health stroke scale(NIHSS)score,history of old cerebral infarction,hypertension,diabetes mellitus,coronary heart disease,dyslipidemia,and smoking history were collected and compared.All patients underwent head MR within 24 h of admission.DWI was used to analyze and compare infarct characteristics across the three groups,including lesion number(single or multiple),location(cortical+subcortical,deep white matter,cortical+subcortical+deep white matter,cerebellum+thalamus+brainstem),size(≥15 mm or＜15 mm,based on maximum transverse diameter;for multiple lesions,if any lesion had a maximum diameter≥15 mm,it was categorized as≥15 mm),infarcted vascular territory(anterior,posterior,or both circulations),and specific arterial supply(anterior cerebral artery,middle cerebral artery,posterior cerebral artery,basilar artery,posterior inferior cerebellar artery,superior cerebellar artery,anterior choroidal artery,or multiple arteries).Patients in the PFO group additionally underwent cardiac CT to measure PFO-related parameters:tunnel length,width,height,septum secundum thickness,and fossa ovalis length.Spearman correlation analysis was performed to evaluate the relationship between infarct size and PFO cardiac CT features.Results A total of 232 acute ischemic stroke patients were included(mean age[57±17]years,ranged 19-86 years;141 males,91 females),comprising 116 in the PFO group,36 in the AAE group,and 80 in the AF group.(1)The proportion of males in the PFO group was higher than that in the AF group,it was lower than that in the AAE group.The age,body mass index and proportions of patients with hypertension,diabetes,hyperlipidemia,coronary heart disease were all lower than those in the other two groups(both P＜0.016 7),while other baseline characteristics showed no significant differences(all P＞0.05).(2)The PFO group exhibited a higher proportion of multiple infarcts compared to the AAE group(83.62％[97/116]vs.61.11％[22/36],P＜0.016 7),but a lower proportion than the AF group(83.62％[97/116]vs.98.75％[79/80],P＜0.016 7).The PFO group also showed a significantly higher proportion of cortical+subcortical infarcts(47.41％[55/116]vs.11.11％[4/36]and 6.25％[5/80],respectively,both P＜0.016 7)and infarcts with a maximum diameter＜15 mm compared to both AAE and AF groups(66.38％[77/116]vs.36.11％[13/36]and 11.25％[9/80],respectively,both P＜0.016 7).Furthermore,the PFO group had a lower proportion of anterior circulation infarcts(27.59％[32/116]vs.69.44％[25/36]in AAE group and 67.50％[54/80]in AF group,both P＜0.016 7),but a higher proportion of posterior circulation infarcts(62.07％[72/116]vs.16.67％[6/36]in AAE group and 8.75％[7/80]in AF group,both P＜0.016 7).Specifically,middle cerebral artery infarcts were less common in the PFO group(18.97％[22/116]vs.66.67％[24/36]in AAE group and 52.50％[42/80]in AF group,both P＜0.016 7),while posterior cerebral artery infarcts were more common(48.28％[56/116]vs.8.33％[3/36]in AAE group and 8.75％[7/80]in AF group,both P＜0.016 7).(3)Spearman correlation analysis revealed that infarct size was negatively correlated with PFO tunnel length(rs=-0.429,P=0.029),fossa ovalis length(rs=-0.408,P=0.038),and septum secundum thickness(rs=-0.525,P=0.006),but not correlated with PFO width or height(both P＞0.05).Conclusions PFO-related cryptogenic stroke is predominantly characterized by multiple small infarcts,primarily located in the cortical+subcortical regions and posterior circulation.Infarct size was found to be negatively correlated with PFO tunnel length,fossa ovalis length,and septum secundum thickness.Comprehensive assessment integrating DWI and cardiac CT features may facilitate the identification of PFO-related stroke.These findings warrant further validation through larger,prospective studies.

Objective:To study the clinical efficacy and impact on quality of life of the first administration of different doses of 131iodine ( 131I) therapy in patients with differentiated thyroid cancer (DTC) after surgical resection. Methods:A total of 97 DTC patients who underwent elective surgical resection at Hongqi Hospital Affiliated to Mudanjiang Medical University from June 2022 to June 2024 were selected and divided into a high-dose group ( n = 49) and a low-dose group ( n = 48) based on their first postoperative treatment with different doses of 131I. All patients underwent DTC endoscopic thyroidectomy. The high-dose group received > 150 - 200 mCi 131I treatment after surgery, while the low-dose group received 100 - 150 mCi 131I treatment after surgery. Both groups of DTC patients were followed up for 12 months. A comparison was conducted between the two groups regarding the success rate of clearing residual thyroid tissue (referred to as thyroid clearance), disease-free survival rate, recurrence rate, incidence of adverse reactions, and changes in quality of life before and after treatment. Results:There was no statistically significant difference in the success rate of thyroid clearance between the two groups of DTC patients (χ 2 = 1.43, P = 0.232). The disease-free survival rate of DTC patients in the low-dose group was higher than that in the high-dose group (χ 2 = 4.86, P = 0.027), and the recurrence rate was lower than that in the high-dose group (χ 2 = 4.86, P = 0.027). The incidence of adverse reactions in DTC patients in the low-dose group was lower than that in the high-dose group (χ 2 = 4.59, P = 0.032). The scores of the short form 36 health survey (SF-36) of both groups of DTC patients after treatment were higher than those before treatment ( P < 0.05), and the SF-36 scores of DTC patients in the low-dose group were higher than those in the high-dose group after treatment ( P < 0.001). The scores of the thyroid cancer specific quality of life (THYCA-QoL) of both groups of DTC patients after treatment were lower than those before treatment ( P < 0.05), and the THYCA-QoL scores of DTC patients in the low-dose group were lower than those in the high-dose group after treatment ( t = 4.48, P < 0.001). Conclusions:The initial treatment effect of different doses of 131I after DTC surgical resection is comparable, but the low-dose 131I can improve the disease-free survival rate, reduce adverse reactions, and improve the quality of life of patients, which is worthy of clinical practice.

The existing relevant standards and registration of disposable sterile surgical drapes were introduced,and four times of national supervision and sample inspection of disposable sterile surgical drapes from 2017 to 2023 were described.The unqualified items found in the supervision and sample inspection were analyzed,the exploratory research projects carried out were summarized and other problems encountered during the sample inspection were pointed out.Relevant enterprises were suggested to strengthen the implementation of industry standards,take seriously the product material identification information,clearly limit the raw materials used and standardize the product registration name.References were provided for the industrial supervision and development of disposable sterile surgical drapes.[Chinese Medical Equipment Journal,2025,46(5):66-72]

Objective To explore whether butorphanol tartrate as an adjuvant of ropivacaine for brachial plexus block can reduce the incidence of rebound pain after brachial plexus block.Methods Based on sample size calculation,174 patients undergoing upper limb bone surgery were included in this study and randomized into three groups using statistical software:butorphanol tartrate compound local anesthetic(group B1),brachial plexus block with 0.25%ropivacaine 20 mL(including adjuvant butorphanol 1mg);intravenous butorphanol group(group B2),brachial plexus block with 0.25%ropivacaine 20 mL,in addition,1mg of butorphanol was administered i.v;control(group C),only 0.25%ropivacaine 20ml for brachial plexus block.The patients were visited the day before operation,and the basic information of the patients was obtained.At the same time,the Douleur Neuropathique 4 questions(DN4)was used to evaluate whether there were neuropathic components(DN4≥4)in the site to be operated on,and the Numerical rating scale(NRS)was introduced to the patients,and the preoperative NRS value was obtained.30 minutes before the operation,the same anesthesiologist with rich experience in nerve block completed the ultrasound-guided brachial plexus block(interscalene approach),and tested whether the block effect was perfect.After entering the operating room,the patients were given general anesthesia,and the duration of operation,vital signs during operation,dosage of analgesics and whether or not using tourniquet were recorded.After the operation,the patients were sent to the postanesthesia care unit,and then sent to the ward when the patients reached the standard of leaving the room.Distribute pain diaries to patients and their families and instruct them to fill in relevant matters.The patients were followed up at 0 h,6 h,12 h,18 h,24 h and 36 h after operation to obtain the NRS value at each time point after operation,the time when the block disappeared and the highest NRS value within 12 hours,the first use of rescue analgesics,the use of postoperative analgesics,postoperative adverse events and the quality of patient recovery.Results The incidence of rebound pain was 31.6%in B1,48.2%in B2,and 54.4%in C.The pairwise comparison showed statistical difference between B1 and C(P<0.05).Rebound pain score in the three groups was B1 group0.05).There were no statistical differences in the duration of motor block,sensory block duration,duration of analgesia,time to first request for rescue analgesia,and QoR-15 recovery quality score(P>0.05).There was no statistical difference in the incidence of postoperative adverse effects among the three groups(P>0.05).Conclusion This study found that brachial plexus block with 1mg butorphanol tartrate as an adjuvant for ropivacaine reduced the incidence of rebound pain and rebound pain scores after upper limb orthopedic surgery and reduced postoperative opioid consumption.

[Objective] To explore the strategy of blood management in patients with massive transfusion in the frigid plateau region. [Methods] The treatment process of a patient with liver rupture in the frigid plateau region was analyzed, and the blood management strategy of the frigid plateau region was discussed in combination with the difficulties of blood transfusion and literature review. [Results] The preoperative complete blood count (CBC) test results of the patient were as follows: RBC 3.14×10¹²/L, Hb 10⁶ g/L, HCT 30.40%, PLT 115.00×10⁹/L; coagulation function: PT 18.9 s, FiB 1.31 g/L, DD > 6 μg/mL, FDP 25.86 μg/mL; ultrasound examination and imaging manifestations suggested liver contusion and laceration / intraparenchymal hematoma, splenic contusion and laceration, and massive blood accumulation in the abdominal cavity; it was estimated that the patient's blood loss was ≥ 2 000 mL, and massive blood transfusion was required during the operation; red blood cell components were timely transfused during the operation, and the blood component transfusion was guided according to the patient's CBC and coagulation function test results, providing strong support and guarantee for the successful treatment of the patient. The patient recovered well after the operation, and the CBC test results were as follows: RBC 4.32×10¹²/L, Hb 144 g/L, HCT 39.50%, PLT 329.00×10⁹/L; coagulation function: APTT 29.3 s, PT 12.1 s, FiB 2.728 g/L, DD>6 μg/mL, FDP 25.86 μg/mL. The patient was discharged after 20 days, and regular follow-up reexamination showed no abnormal results. [Conclusion] Individualized blood management strategy should comprehensively consider the patient’s clinical symptoms, the degree of hemoglobin decline, dynamic coagulation test results and existing treatment conditions. Efficient and reasonable patient blood management strategies can effectively improve the clinical outcomes of massive transfusion patients in the frigid plateau region.