Objective To propose a dynamic early warning model based on machine learning methods and validate its predi-ctive efficacy so as to achieve precise assessment and early warning of mortality risk in patients with traumatic hypothermia.Methods Firstly,a total of 480 patients who met inclusion criteria were retrospectively selected from the eICU database and randomly divided into training and test sets at an 8∶2 ratio.Secondly,physiological parameters were extracted from these patients,and five machine learning algorithms including XGBoost,AdaBoost,LightGBM,logistic regression(LR)and random forest(RF)were employed respectively to develop dynamic mortality risk warning models for traumatic hypothermia patients,utilizing a 1-hour observation window.Thirdly,receiver operating characteristic curves(ROC)were plotted using the test set data and the effects of different warning windows on the model performance were analyzed by calculating the AUC.Finally,the interpretability of the models was analyzed using the SHapley Additive exPlanations(SHAP)algorithm to elucidate the contribution of each feature to predictive performance.Results The optimal warning window for the dynamic warning model constructed using the eICU database was 12 hours,and in case of 12-hour warning window the logistic regression model achieved the highest AUC of 0.935 and showed optimal predictive performance.The results of the interpretability analysis by the SHAP algorithm showed that body temperature was the feature that had the greatest impact on the model results,and its reduction was positively correlated with the increased risk of death.Conclusion The machine learning-based dynamic warning model for mortality risk in traumatic hypothermia patients enables real-time dynamic risk assessment,providing robust support for clinicians to identify the patient's condition changes at an early stage and references for the adjustment of clinical treatment programs.[Chinese Medical Equipment Journal,2025,46(3):9-14]

Sacroiliac complex injuries are commonly seen in high-energy pelvic fractures. The injuries make a big difference in treatment patterns due to the diverse injury types, posing considerable challenges in formulating optimal treatment strategies, and hence are persistent clinical difficulties in orthopedic trauma. The clinical management of sacroiliac complex injuries presents several key challenges such as a non-negligible rate of missed diagnoses in associated vascular and visceral injuries, absence of standardized protocols for surgical approaches and reduction-fixation strategies across different injury patterns, and ongoing controversies regarding surgical indications and optimal timing for patients combined with concomitant lumbosacral plexus injuries. Currently, no systematic clinical guidelines are available for the diagnosis and treatment of sacroiliac complex injuries both domestically and internationally. To this end, the Pelvic and Acetabular Surgery Group, Orthopedic Branch, China International Exchange and Promotive Association for Medical and Health Care and Orthopedic Physician Branch, Chinese Medical Doctor Association organized a panel of domestic experts in the field to develop the Clinical guideline for the diagnosis and treatment of sacroiliac complex injuries ( version 2025), based on evidence-based medicine and adhering to the principles of scientific rigor, clinical applicability, and innovation. These guidelines provided 11 recommendations covering diagnosis, therapeutic principles and techniques, management protocols for lumbosacral plexus injuries, outcome evaluation, and postoperative rehabilitation pathways, etc., aiming to standardize the clinical management of sacroiliac complex injuries.

Objective:To compare the clinical manifestations,laboratory results,and imaging features between pyogenic spondylitis(PS) and tuberculous spondylitis(TS).Methods:This was a cross-sectional study. A total of 88 patients with infectious diseases of spine(IDS) admitted to Shanghai Sixth People′s Hospital from January 2021 to December 2023 were analyzed,including 61 PS cases(PS group) and 27 TS cases(TS group). The clinical manifestations,laboratory results,and imaging features were compared between two groups. The factors associated with PS were analyzed by multivariate logistic regression. The diagnostic efficacy for pathogen identification was compared between metagenomics next-generation sequencing(mNGS) and bacterial culture methods in PS and TS patients.Results:Compared with the TS group,the PS group had a higher age-adjusted Charlson comorbidity index(aCCI)[3.0(1.5,4.0) points vs. 2.0(1.0,3.0) points, Z=-2.189, P=0.029],shorter onset time of disease[1.0(0.8,3.0) months vs. 6.0(2.0,12.0) months,Z=-4.353, P<0.001],and higher median blood leukocyte counts and serum ferritin(SF) level(7.2×10 9/L vs. 6.3×10 9/L, Z=-2.652, P=0.008; 571.3 ng/ml vs. 266.0 ng/ml, Z=-4.773, P<0.001). The proportions of lumbar spine involvement,non-collapsed involved vertebrae,and bone bridges formed were all higher in the PS group compared to the TS group[68.8%(99/144) vs. 41.4%(29/70), χ2=14.628, P<0.001; 68.9%(42/61) vs. 18.5%(5/27), χ2=19.055, P<0.001; 41.0%(25/61) vs. 7.4%(2/27), χ2=9.921, P=0.002]. The proportions of thoracic spine involvement,severe vertebral collapse,severe narrowing of the involved intervertebral space,sequestrum,and paravertebral soft tissue calcification were all higher in the TS group compared to the PS group[52.9%(37/70) vs. 18.1%(26/144), χ2=27.463, P<0.001; 55.6%(15/27) vs. 13.1%(8/61), χ2=17.462, P<0.001; 74.1%(20/27) vs. 37.7%(23/61), χ2=9.907, P=0.002; 74.1%(20/27) vs. 18.0%(11/61), χ2=25.761, P<0.001; 51.9%(14/27) vs. 6.6%(4/61), χ2=23.599, P<0.001]. Multivariate logistic regression analysis indicated that a symptom duration<5.5 months( OR=30.644,95% CI: 2.022-464.529, P<0.05) and a leukocyte count>7.35×10 9/L( OR=48.653,95% CI: 2.045-1 157.721, P<0.05) indicated a higher likelihood of PS; while the vertebral collapse indicated a higher likelihood of TS( OR=0.025,95% CI: 0.001-0.638, P<0.05). The most common pathogen in the PS group was Staphylococcus aureus(31 cases,50.8%),followed by Streptococcus species(10 cases,16.4%). The positive rates of mNGS testing in the PS and TS groups were 84.1%(37/44) and 12/13,respectively,which were higher than those of conventional bacterial culture[77.8%(42/54)] and Mycobacteriumtuberculosis culture(2/11). Conclusions:Compared with the TS patients,the PS patients have shorter onset time,higher aCCI scores,higher blood leukocyte counts and SF levels,less vertebral collapse and intervertebral space narrowing,and more bone bridge formation. The TS patients have more dead bones and calcifications. The mNGS has a higher diagnostic efficacy than bacterial cultures for PS and TS.

Objective To propose a low-dose CT image denoising method based on an improved Corediff model to recover the detailed features of the image and enhance the image quality.Methods An RS-Corediff model was established by modifying the key component U-Net network of the Corediff model.Firstly,the residual module was introduced in the network input stage for feature extraction;secondly,a new downsampling module was designed in the U-Net network encoder,which learned the semantic information of the feature map by convolution and maintained the learning state during the downsampling process so as to fully extract the image features;thirdly,the feature splicing processing was used to further enhance the learning effect during the upsampling process of the U-Net network decoder;finally,the convolutional kernel size was modified to adjust the sensory field during the convolutional process of the whole U-Net network structure so as to obtain rich features.The RS-Corediff model was compared with the residual encoder-decoder convolutional neural network(RED-CNN)model and the Corediff model on the public dataset AAPM 2016 in order to verify its effectiveness for low-dose CT image denoising.Results The RS-Corediff model gained advantages over the RED-CNN and Corediff models with a peak signal-to-noise ratio(PSNR)of 41.269 8,structural similarity(SSIM)of 0.953 4 and root mean square error(RMSE)of 17.568 7.Conclusion The proposed method effectively preserves the texture and details of low-dose CT images during the denoising process to improve the overall quality of the images.[Chinese Medical Equipment Journal,2025,46(5):9-13]

Objectives:QT interval prolongation during treadmill test exercise is one of the clinical feature of patients with long QT syndrome(LQTS).This study aimed to explore the feasibility and efficacy of treadmill exercise testing as an auxiliary diagnosis tool for LQTS in clinical practice.Methods:We enrolled normal healthy individuals,common cardiovascular disease patients,and clinically diagnosed or suspected LQTS patients,who underwent treadmill exercise test from July 2023 to July 2024 at Fuwai Hospital.A special treadmill exercise testing procedure was designed to record the QT interval correction(QTc)intervals of the twelve lead electrocardiogram at 6 time points when performing the exercise tablet,including supine,sitting,standing,peak exercise,and recovery at 1-minute and 4-minute.The differences in QTc intervals among healthy group,cardiovascular diseases group,and suspected LQTS group were compared.Results:A total of 80 cases were consecutively enrolled,including 37 normal healthy controls,25 patients with common cardiovascular disease,and 18 patients with suspected LQTS.The QTc intervals at 6 points did not differ significantly between normal healthy controls and patients with cardiovascular disease,with QTc intervals less than 480 ms at all measurement.For patients with suspected LQTS,67.7%(12/18)of these patients presented a QTc interval≥480 ms at the 4-minute during recovery period.Among them,5 cases were confirmed to have pathogenic gene mutations of LQTS by genetic testing(including 1 case with a lying electrocardiogram QTc interval of 489 ms diagnosed with LQTS 1 type and a QTc interval of 636 ms during the 4-minute recovery period after exercise);5 clinically diagnosed patients(negative or undetectable in genetic testing)with a Schwartz score≥4,and the remaining 2 patients had a Schwartz score of 3.The remaining 5/18 patients,include 2 patients with clinical Schwartz scores≥4 and 3 patients with clinical suspicion(Schwartz scores 2-3)had a 4 min QTc interval of 445-480 ms during exercise recovery.Another patient with clinical suspicion(Schwartz score 3)had a 4 min QTc interval of<445 ms during exercise recovery and a negative genetic test at a later stage.Receiver operating characteristic curve analysis showed a sensitivity of 83.3%and specificity of 98.4%for QTc interval≥482 ms during the 4-minute recovery period of exercise as the LQTS diagnostic cutoff.Conclusions:This study results suggest that this special treadmill exercise testing protocol is effective in identifying LQTS and has strong feasibility and generalizability for clinical practice.

Objectives:QT interval prolongation during treadmill test exercise is one of the clinical feature of patients with long QT syndrome(LQTS).This study aimed to explore the feasibility and efficacy of treadmill exercise testing as an auxiliary diagnosis tool for LQTS in clinical practice.Methods:We enrolled normal healthy individuals,common cardiovascular disease patients,and clinically diagnosed or suspected LQTS patients,who underwent treadmill exercise test from July 2023 to July 2024 at Fuwai Hospital.A special treadmill exercise testing procedure was designed to record the QT interval correction(QTc)intervals of the twelve lead electrocardiogram at 6 time points when performing the exercise tablet,including supine,sitting,standing,peak exercise,and recovery at 1-minute and 4-minute.The differences in QTc intervals among healthy group,cardiovascular diseases group,and suspected LQTS group were compared.Results:A total of 80 cases were consecutively enrolled,including 37 normal healthy controls,25 patients with common cardiovascular disease,and 18 patients with suspected LQTS.The QTc intervals at 6 points did not differ significantly between normal healthy controls and patients with cardiovascular disease,with QTc intervals less than 480 ms at all measurement.For patients with suspected LQTS,67.7%(12/18)of these patients presented a QTc interval≥480 ms at the 4-minute during recovery period.Among them,5 cases were confirmed to have pathogenic gene mutations of LQTS by genetic testing(including 1 case with a lying electrocardiogram QTc interval of 489 ms diagnosed with LQTS 1 type and a QTc interval of 636 ms during the 4-minute recovery period after exercise);5 clinically diagnosed patients(negative or undetectable in genetic testing)with a Schwartz score≥4,and the remaining 2 patients had a Schwartz score of 3.The remaining 5/18 patients,include 2 patients with clinical Schwartz scores≥4 and 3 patients with clinical suspicion(Schwartz scores 2-3)had a 4 min QTc interval of 445-480 ms during exercise recovery.Another patient with clinical suspicion(Schwartz score 3)had a 4 min QTc interval of<445 ms during exercise recovery and a negative genetic test at a later stage.Receiver operating characteristic curve analysis showed a sensitivity of 83.3%and specificity of 98.4%for QTc interval≥482 ms during the 4-minute recovery period of exercise as the LQTS diagnostic cutoff.Conclusions:This study results suggest that this special treadmill exercise testing protocol is effective in identifying LQTS and has strong feasibility and generalizability for clinical practice.

Objective To propose a dynamic early warning model based on machine learning methods and validate its predi-ctive efficacy so as to achieve precise assessment and early warning of mortality risk in patients with traumatic hypothermia.Methods Firstly,a total of 480 patients who met inclusion criteria were retrospectively selected from the eICU database and randomly divided into training and test sets at an 8∶2 ratio.Secondly,physiological parameters were extracted from these patients,and five machine learning algorithms including XGBoost,AdaBoost,LightGBM,logistic regression(LR)and random forest(RF)were employed respectively to develop dynamic mortality risk warning models for traumatic hypothermia patients,utilizing a 1-hour observation window.Thirdly,receiver operating characteristic curves(ROC)were plotted using the test set data and the effects of different warning windows on the model performance were analyzed by calculating the AUC.Finally,the interpretability of the models was analyzed using the SHapley Additive exPlanations(SHAP)algorithm to elucidate the contribution of each feature to predictive performance.Results The optimal warning window for the dynamic warning model constructed using the eICU database was 12 hours,and in case of 12-hour warning window the logistic regression model achieved the highest AUC of 0.935 and showed optimal predictive performance.The results of the interpretability analysis by the SHAP algorithm showed that body temperature was the feature that had the greatest impact on the model results,and its reduction was positively correlated with the increased risk of death.Conclusion The machine learning-based dynamic warning model for mortality risk in traumatic hypothermia patients enables real-time dynamic risk assessment,providing robust support for clinicians to identify the patient's condition changes at an early stage and references for the adjustment of clinical treatment programs.[Chinese Medical Equipment Journal,2025,46(3):9-14]

In this work,with tris(4-aminophenyl)amine(TAPA)and 1,3,5-tris(4-formylphenyl)benzene(TFPB)as monomers,an imine-type porous organic polymer,TAPA-TFPB,was synthesized using a simple method under the catalysis of acetic acid.The material TAPA-TFPB was used as solid-phase extraction adsorbent and combined with ultra-performance liquid chromatography/quadrupole time-of-flight-tandem mass spectrometry(UHPLC-QTOF-MS)to establish a detection method for four kinds of nitrofuran metabolites(NFMs)residues in meat samples.The parameters of the adsorbent dosage,the pH value and volume of sample,and the type and volume of washing and eluent solvents were optimized,respectively.Under the optimal extraction conditions,low detection limits(0.11-1.60 μg/kg)were achieved for four kinds of NFMs.At three different spiked levels,the intra-day and inter-day precisions(Relative standard deviations)were 2.8% -10.9% and 4.3% -16.2%,respectively,and the spiked recoveries were 72.0% -107.2%.The results showed that the method chould be used for efficient extraction and analysis of trace NFMs residues in meat samples,indicating that TAPA-TFPB was a kind of promising SPE adsorbent.

Objective To initially explore the possibility of applying the fluorescence micro-optical sectioning tomography(fMOST)high-resolution 3D reconstruction system to the morphological study of the intestinal nervous system and to preliminarily establish a method for studying the morphology of the intestinal nervous system using this system.Methods fMOST high-resolution 3D reconstruction system was used to study the intestinal nervous system of C57BL/6 mice in detail.Based on this method,a new morphological method of the visceral nervous system of small animal models was explored at the single-cell level.Results Compared with the large intestine,the small intestine lacked the typical myenteric plexus(Auerbach),deep mucosal plexus(Henley),and submucosal superficial plexus(Meissner).Conclusion The result of this paper provide a clearer and systematic display of the anatomical structure of the enteric nervous system in C57BL/6 mice,and further clarify the similarities and differences between the enteric nervous system of mice and human,and provide a theoretical basis for its rational application in the study of digestive system diseases.The morphological study of fMOST high-resolution 3D reconstruction system is not limited to the central nervous system,but can be extended to the morphological study of multiple visceral nervous systems.

OBJECTIVE: To explore short-term clinical efficacy of minimally invasive diagnosis and treatment system for hallux valgus in guiding the treatment of hallux valgus. METHODS: From March 2021 to November 2023, 68 patients (136 feet) with hallux valgus were treated under guidance of minimally invasive diagnosis and treatment system, including 12 males and 56 females;aged from 25 to 68 years old with an average of (42.5±8.5) years old, the course of disease ranged from 3.2 to 15.6 years with an average of (10.3±2.6) years. The changes of hallux valgus angle (HVA) and intermetatarsal angle (IMA), visual analog scale (VAS) and American Orthopaedic Foot Ankle Society (AOFAS) forefoot score were recorded and compared before operation and 12 months after operation. RESULTS: Sixty-five patients (130 feet) were followed up for 12 to 15 months with an average of (13.8±0.5) months, 3 patients (6 feet) were not followed up as required. HVA and IMA improved from (35.5±3.5) ° and (12.5±2.0) ° before operation to (10.5±2.5) ° and (8.5±1.5) °12 months after operation, respectively, with statistically significant differences (P<0.05);VAS decreased from (5.5±1.2) before operation to (1.2±0.5) at 12 months after operation, and the difference was statistically significant (P<0.05);AOFAS forefoot score increased from (50.6±5.1) before operation to (93.8±5.6) at 12 months after operation, with a statistically significant difference (P<0.05). Among them, 102 feet were got excellent result, 24 feet good, and 4 feet fair. Two patients were developed calf intermuscular vein thrombosis, and were cured after 3 months of symptomatic treatment. CONCLUSION Under the guidance of minimally invasive diagnosis and treatment system for hallux valgus, the treatment of HV could obviously improve HVA and IMA, and significantly alleviate pain symptoms, and accelerate functional recovery.
Humans ; Hallux Valgus/diagnosis* ; Male ; Female ; Middle Aged ; Adult ; Aged ; Minimally Invasive Surgical Procedures/methods* ; Treatment Outcome