BACKGROUND: Neoadjuvant chemoradiotherapy followed by radical surgery has been a common practice for patients with locally advanced rectal cancer, but the response rate varies among patients. This study aimed to develop a ResNet-Vision Transformer based magnetic resonance imaging (MRI)-endoscopy fusion model to precisely predict treatment response and provide personalized treatment. METHODS: In this multicenter study, 366 eligible patients who had undergone neoadjuvant chemoradiotherapy followed by radical surgery at eight Chinese tertiary hospitals between January 2017 and June 2024 were recruited, with 2928 pretreatment colonic endoscopic images and 366 pelvic MRI images. An MRI-endoscopy fusion model was constructed based on the ResNet backbone and Transformer network using pretreatment MRI and endoscopic images. Treatment response was defined as good response or non-good response based on the tumor regression grade. The Delong test and the Hanley-McNeil test were utilized to compare prediction performance among different models and different subgroups, respectively. The predictive performance of the MRI-endoscopy fusion model was comprehensively validated in the test sets and was further compared to that of the single-modal MRI model and single-modal endoscopy model. RESULTS: The MRI-endoscopy fusion model demonstrated favorable prediction performance. In the internal validation set, the area under the curve (AUC) and accuracy were 0.852 (95% confidence interval [CI]: 0.744-0.940) and 0.737 (95% CI: 0.712-0.844), respectively. Moreover, the AUC and accuracy reached 0.769 (95% CI: 0.678-0.861) and 0.729 (95% CI: 0.628-0.821), respectively, in the external test set. In addition, the MRI-endoscopy fusion model outperformed the single-modal MRI model (AUC: 0.692 [95% CI: 0.609-0.783], accuracy: 0.659 [95% CI: 0.565-0.775]) and the single-modal endoscopy model (AUC: 0.720 [95% CI: 0.617-0.823], accuracy: 0.713 [95% CI: 0.612-0.809]) in the external test set. CONCLUSION The MRI-endoscopy fusion model based on ResNet-Vision Transformer achieved favorable performance in predicting treatment response to neoadjuvant chemoradiotherapy and holds tremendous potential for enabling personalized treatment regimens for locally advanced rectal cancer patients.
Humans ; Rectal Neoplasms/diagnostic imaging* ; Magnetic Resonance Imaging/methods* ; Male ; Female ; Middle Aged ; Neoadjuvant Therapy/methods* ; Aged ; Adult ; Chemoradiotherapy/methods* ; Endoscopy/methods* ; Treatment Outcome

PURPOSE: This study aims to identify the prevalence and risk factors of military training-related abdominal injuries and help plan and conduct training properly. METHODS: This questionnaire survey study was conducted from October 2021 to May 2022 among military personnel from 6 military units and 8 military medical centers and participants' medical records were consulted to identify the training-related abdominal injuries. All the military personnel who ever participated in military training were included. Those who refused to participate in this study or provided an incomplete questionnaire were excluded. The questionnaire collected demographic information, type of abdominal injury, frequency, training subjects, triggers, treatment, and training disturbance. Chi-square test and t-test were used to compare baseline information. Univariate and multivariate regression analyses were used to explore the risk factors associated with military training-related abdominal injuries. RESULTS: A total of 3058 participants were involved in this study, among which 1797 (58.8%) had suffered training-related abdominal injuries (the mean age was 24.3 years and the service time was 5.6 years), while 1261 (41.2%) had no training-related abdominal injuries (the mean age was 23.1 years and the service time was 4.3 years). There were 546 injured patients (30.4%) suspended the training and 84 (4.6%) needed to be referred to higher-level hospitals. The most common triggers included inadequate warm-up, fatigue, and intense training. The training subjects with the most abdominal injuries were long-distance running (589, 32.8%). Civil servants had the highest rate of abdominal trauma (17.1%). Age ≥ 25 years, military service ≥ 3 years, poor sleep status, and previous abdominal history were independent risk factors for training-related abdominal injury. CONCLUSION More than half of the military personnel have suffered military training-related abdominal injuries. Inadequate warm-up, fatigue, and high training intensity are the most common inducing factors. Scientific and proper training should be conducted according to the factors causing abdominal injuries.
Humans ; Military Personnel ; Risk Factors ; Prevalence ; Male ; Abdominal Injuries/etiology* ; Female ; Adult ; Surveys and Questionnaires ; Young Adult

PURPOSE: To evaluate the diagnostic accuracy of the observational chart for traumatic limb swelling (OCTLS) for osteofascial compartment syndrome (OCS). METHODS: This was a descriptive-longitudinal study. Data of 316 patients who underwent surgical treatment for tibial fractures in our department from January 2015 to December 2023 were collected. Patients with Gustilo type II or higher open fractures, vascular injury, or bilateral fractures were excluded from the study. Two groups of double-blinded investigators independently assessed patients for the presence of OCS using 2 distinct diagnostic methods. Three senior orthopedic trauma surgeons evaluated patients with post-fracture calf swelling for OCS and the need for fasciotomy based on clinical signs and their extensive clinical experience. Subsequently, fasciotomy was performed according to their judgment, followed by postoperative examination of muscle and soft tissue conditions. Additionally, a follow-up evaluation was conducted to assess for complications such as ischemic muscle contracture. Another 3 trained researchers used OCTLS to grade swelling severity and determine the need for fasciotomy. The final diagnostic gold standard of OCS was determined by referring to whether there was escape of muscles at fasciotomy and/or color change in the muscles or muscle necrosis intraoperatively, and neurological abnormality or contracture at the last follow-up. The results of the 2 diagnostic methods were compared with the final diagnostic result. Kappa consistency test, paired χ² test (McNemar test), and receiver operating characteristic curve were used to evaluate the diagnostic efficacy of the 2 diagnostic methods. RESULTS: Of the 316 patients, 211 were finally included in the study, including 160 males and 51 females, with an average follow-up time of (14.5 ± 2.7) months. Among the 211 patients with tibial fracture-associated swelling, 42 were definitively diagnosed with OCS. Based on clinical symptoms and signs judgment, among the 65 fasciotomy patients, 38 were confirmed as correct, while among the 146 non-fasciotomy patients, 4 developed ischemic muscle contractures. Based on the OCTLS for assessment, fasciotomy was correctly recommended in 36 out of 43 cases, while 6 out of 168 non-fasciotomy patients developed OCS. Compared to the use of the gold standard, clinical signs judgment showed moderate consistency (McNemar's test p < 0.001, Kappa = 0.618, p < 0.001), whereas OCTLS demonstrated strong agreement (McNemar's test p = 1.000, Kappa = 0.808, p < 0.001). Receiver operating characteristic analysis revealed higher diagnostic accuracy for OCTLS (area under curve = 0.908, 95% CI: 0.843 - 0.972) compared to clinical signs judgment (area under curve = 0.872, 95% CI: 0.812 - 0.933). OCTLS achieved superior accuracy (93.8% vs. 85.3%, χ² = 8.221, p < 0.001) and a lower fasciotomy rate (20.4% vs. 30.8%, χ² = 6.023, p = 0.014). CONCLUSION Compared to clinical signs judgment, OCTLS significantly reduces unnecessary fasciotomy, improves diagnostic accuracy for OCS, and enables non-invasive, dynamic, and quantitative assessment, making it a valuable tool for clinical practice.
Humans ; Compartment Syndromes/etiology* ; Male ; Female ; Adult ; Tibial Fractures/surgery* ; Middle Aged ; Fasciotomy ; Edema/etiology* ; Longitudinal Studies ; Aged ; Young Adult

Objective To analyze the serum of patients with pancreatic cancer by fourier transform infrared spectroscopy(FTIR),explore the characteristic spectral parameters related to pancreatic cancer,and evaluate its potential clinical diagnostic value.Methods Serum samples were collected from 100 patients diagnosed with pancreatic cancer and 92 healthy volunteers at Chongming Hospital Affiliated to Shanghai Health Medical College from August 2022 to July 2023.These samples underwent FTIR and principal component analysis(PCA)to assess spectral differences between the two cohorts.The diagnostic potential of the serum spectra in distinguishing pancreatic cancer patients from healthy individuals was further evaluated using machine learning techniques,specifically support vector machine(SVM),k-nearest neighbor(kNN),and linear discriminant analysis(LDA)as classification methods.The diagnostic efficacy across various thresholds was evaluated using the receiver operating characteristic(ROC)curve.Results The findings indicate that the peak positions within the 1 090～1 070cm-1(1 076.537±15.183cm-1 vs 1 081.061±4.043cm-1),1 420～1 380 cm-1(1 399.958±1.508cm-1 vs 1 400.500±1.782cm-1),2 990～2 950cm-1(2 940.167±15.287cm-1 vs 2 945.124±7.498cm-1)and 3 500～3 000 cm-1(3 293.155±3.096cm-1 vs 3 294.893±2.582cm-1)range in the serum of individuals with pancreatic cancer exhibited a significant blue-shift compared to the healthy group and was statistically significant(t=2.265～4.236,all P<0.05),suggesting alterations in the structures of proteins,lipids and nucleic acids.Furthermore,a statistically significant disparity in peak absorption was observed between the group of patients with pancreatic cancer and the healthy group within the spectral ranges of 1 700～1 600cm-1(0.918±0.012cm-1 vs 0.858±0.021cm-1)and 3 500～3 000 cm-1(0.766±0.096cm-1 vs 0.804±0.090cm-1)(t=-24.031,2.830,all P<0.05),indicating that the protein concentration changes.PCA results showed that the PC2 axis was clearly separated,which could distinguish serum samples from patients with pancreatic cancer.Utilizing a machine learning model to differentiate the serum spectra of patients with pancreatic cancer from those of healthy controls,the sensitivity,the spesitivity,the specificity and accuracy of linear discriminant analysis(LDA)classification method were 93.2%,97.3%and 95.8%,respectively.The area under curve(AUC)as determined by ROC analysis was 0.982.Conclusion Serum spectroscopy using FTIR combined with PCA and machine learning model can be a simple,minimally invasive and reliable diagnostic test for pancreatic cancer detection.

In protein engineering, while computational models are increasingly used to predict mutation effects, their evaluations primarily rely on high-throughput deep mutational scanning (DMS) experiments that use surrogate readouts, which may not adequately capture the complex biochemical properties of interest. Many proteins and their functions cannot be assessed through high-throughput methods due to technical limitations or the nature of the desired properties, and this is particularly true for the real industrial application scenario. Therefore, the desired testing datasets, will be small-size (∼10-100) experimental data for each protein, and involve as many proteins as possible and as many properties as possible, which is, however, lacking. Here, we present VenusMutHub, a comprehensive benchmark study using 905 small-scale experimental datasets curated from published literature and public databases, spanning 527 proteins across diverse functional properties including stability, activity, binding affinity, and selectivity. These datasets feature direct biochemical measurements rather than surrogate readouts, providing a more rigorous assessment of model performance in predicting mutations that affect specific molecular functions. We evaluate 23 computational models across various methodological paradigms, such as sequence-based, structure-informed and evolutionary approaches. This benchmark provides practical guidance for selecting appropriate prediction methods in protein engineering applications where accurate prediction of specific functional properties is crucial.

General anesthesia, pivotal for surgical procedures, requires precise depth monitoring to mitigate risks ranging from intraoperative awareness to postoperative cognitive impairments. Traditional assessment methods, relying on physiological indicators or behavioral responses, fall short of accurately capturing the nuanced states of unconsciousness. This study introduces a machine learning-based approach to decode anesthesia depth, leveraging EEG data across different anesthesia states induced by propofol and esketamine in rats. Our findings demonstrate the model's robust predictive accuracy, underscored by a novel intra-subject dataset partitioning and a 5-fold cross-validation method. The research diverges from conventional monitoring by utilizing anesthetic infusion rates as objective indicators of anesthesia states, highlighting distinct EEG patterns and enhancing prediction accuracy. Moreover, the model's ability to generalize across individuals suggests its potential for broad clinical application, distinguishing between anesthetic agents and their depths. Despite relying on rat EEG data, which poses questions about real-world applicability, our approach marks a significant advance in anesthesia monitoring.
Animals ; Machine Learning ; Electroencephalography/methods* ; Ketamine/administration & dosage* ; Rats ; Male ; Propofol/administration & dosage* ; Rats, Sprague-Dawley ; Anesthesia, General/methods* ; Brain/physiology* ; Intraoperative Neurophysiological Monitoring/methods*

Objective To investigate the effect of closed-chain exercise training on hemiplegic shoulder pain and shoulder joint sta-bility in stroke patients,and to explore the relationship between hemiplegic shoulder pain and shoulder joint sta-bility.Methods A total of 52 stroke patients with hemiplegic shoulder pain in Jiangsu Province Hospital from October,2020 to January,2023 were selected,and were randomly divided into control group(n=26)and experimental group(n=26).Both groups received conventional rehabilitation therapy,while the experimental group additionally under-went closed-chain exercise training for shoulder joint.Pain severity and motor function were assessed using Visu-al Analog Scale(VAS)and Fugl-Meyer Assessment-Upper Extremities(FMA-UE)before and after intervention.Musculoskeletal ultrasound was used to measure acromion-greater tuberosity distance(AGT),acromion-lesser tu-berosity distance(ALT),acromiohumeral distance(AHD)and supraspinatus thickness(SST)to evaluate shoulder joint stability.Correlation analysis was conducted on the improvements in shoulder pain and shoulder joint stabil-ity in the experimental group.Results Two cases in the control group and two in the experimental group dropped down.Both groups showed signifi-cant improvements in VAS and FMA-UE scores after intervention(|t|>5.214,P<0.001),and the scores im-proved more in the experimental group than in the control group(|t|>2.087,P<0.05).The experimental group also showed significant improvements in AGT,ALT,AHD and SST(|t|>4.187,P<0.001),with AGT,ALT and AHD being superior to those in the control group(|t|>2.155,P<0.05).The difference of VAS score in the exper-imental group was correlated with the difference of FMA-UE,AGT and ALT after intervention(r>0.434,P<0.05).Conclusion Closed-chain shoulder exercise training can significantly improve shoulder joint stability while alleviating shoulder pain,and effectively enhancing upper limb function in stroke patients with hemiplegic shoulder pain.Hemiplegic shoulder pain is correlated with shoulder joint stability.

Aim To study the mechanism of adipose mesenchymal stem cell exosomes(ASC-exo)inhibition of fluorescein isothiocyanate(FITC)-induced atopic dermatitis(AD).Methods The mouse age,extrac-tion method,and the concentration of a solution of typeⅠ collagen enzyme and other conditions were compared to study the effects on the morphology and quantity of adipose mesenchymal stem cells(ASCs)after extrac-ted.FITC-induced mouse model in vivo was estab-lished and different doses of ASC-exo were given to measure ear thickness,ear weight and ear scratching times of mice.HE staining was used to observe the pathological changes of ear tissue of mice.The non-toxicity of ASC-exo was detected.IgE,IL-5,IL-13 and other cytokines were detected by ELISA.The gene ex-pressions of TSLP,IL-33,occludin,Claudin-1(CLDN-1)and E-cadherin were detected by RT-qPCR.The protein expression was detected by immunohistochemis-try.Results An efficient method for extracting ASCs was established.Compared with the blank group,mice in the model group showed obvious AD symptoms.Compared with the model group,ASC-exo administra-tion group significantly reduced the number of ear scratches,epidermal thickening,inflammatory cell infil-tration and the secretion of Th2 cytokines IL-5 and IL-13.Meanwhile,ASC-exo administration group signifi-cantly increased the expression of structural proteins CLDN-1 and occludin in epithelial cells and decreased the expression of TSLP and IL-33.Conclusions ASC-exo can significantly improve Th2 skin inflamma-tion in AD mice,and its mechanism may be through in-creasing the expression of tight junction proteins and adhesion link protein in epithelial cells,repairing the skin barrier,and inhibiting the key promoters of allergy TSLP and IL-33.

Objective To analyze the serum of patients with pancreatic cancer by fourier transform infrared spectroscopy(FTIR),explore the characteristic spectral parameters related to pancreatic cancer,and evaluate its potential clinical diagnostic value.Methods Serum samples were collected from 100 patients diagnosed with pancreatic cancer and 92 healthy volunteers at Chongming Hospital Affiliated to Shanghai Health Medical College from August 2022 to July 2023.These samples underwent FTIR and principal component analysis(PCA)to assess spectral differences between the two cohorts.The diagnostic potential of the serum spectra in distinguishing pancreatic cancer patients from healthy individuals was further evaluated using machine learning techniques,specifically support vector machine(SVM),k-nearest neighbor(kNN),and linear discriminant analysis(LDA)as classification methods.The diagnostic efficacy across various thresholds was evaluated using the receiver operating characteristic(ROC)curve.Results The findings indicate that the peak positions within the 1 090～1 070cm-1(1 076.537±15.183cm-1 vs 1 081.061±4.043cm-1),1 420～1 380 cm-1(1 399.958±1.508cm-1 vs 1 400.500±1.782cm-1),2 990～2 950cm-1(2 940.167±15.287cm-1 vs 2 945.124±7.498cm-1)and 3 500～3 000 cm-1(3 293.155±3.096cm-1 vs 3 294.893±2.582cm-1)range in the serum of individuals with pancreatic cancer exhibited a significant blue-shift compared to the healthy group and was statistically significant(t=2.265～4.236,all P<0.05),suggesting alterations in the structures of proteins,lipids and nucleic acids.Furthermore,a statistically significant disparity in peak absorption was observed between the group of patients with pancreatic cancer and the healthy group within the spectral ranges of 1 700～1 600cm-1(0.918±0.012cm-1 vs 0.858±0.021cm-1)and 3 500～3 000 cm-1(0.766±0.096cm-1 vs 0.804±0.090cm-1)(t=-24.031,2.830,all P<0.05),indicating that the protein concentration changes.PCA results showed that the PC2 axis was clearly separated,which could distinguish serum samples from patients with pancreatic cancer.Utilizing a machine learning model to differentiate the serum spectra of patients with pancreatic cancer from those of healthy controls,the sensitivity,the spesitivity,the specificity and accuracy of linear discriminant analysis(LDA)classification method were 93.2%,97.3%and 95.8%,respectively.The area under curve(AUC)as determined by ROC analysis was 0.982.Conclusion Serum spectroscopy using FTIR combined with PCA and machine learning model can be a simple,minimally invasive and reliable diagnostic test for pancreatic cancer detection.

Objective::This study aimed to assess the impact of proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor treatment immediately after percutaneous coronary intervention (PCI) on the myocardial salvage index (MSI) in patients with anterior ST-segment elevation myocardial infarction (STEMI) 5-10 d after the procedure.Methods::The early PCSK9 inhibitor thERapy Following pErcutaneous Coronary inTervention (PERFECT) trial is a prospective randomized controlled trial. From January 2021 to December 2023, 32 patients with anterior STEMI from Tongren Hospital, Shanghai Jiao Tong University School of Medicine, Songjiang Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, and Shanghai Tenth People’s Hospital were enrolled in the PERFECT trial. Patients were randomly assigned in a 1∶1 ratio to the PCSK9 inhibitor group ( n = 16) or the control group ( n = 16), and their baseline data were collected. Patients in the PCSK9 inhibitor group (ie, alirocumab group) received a subcutaneous injection of PCSK9 inhibitor (alirocumab, 75 mg) immediately after PCI based on conventional treatment. In the control group, patients received only conventional treatment. The primary endpoint was the MSI measured by cardiovascular magnetic resonance 5-10 d after PCI. The secondary endpoints included the left ventricular ejection fraction measured by cardiovascular magnetic resonance 5-10 d after PCI and the time to peak of creatine kinase isoenzyme-MB and high-sensitivity cardiac troponin T. Safety endpoints included any clinical adverse events that occurred during the 6-month follow-up period. Results::Baseline data during admission showed no intergroup significance. No significant difference in MSI (55.54% ± 14.80% vs. 44.72% ± 15.42%, P = 0.056) and left ventricular ejection fraction (51.24% ± 8.91% vs. 44.99% ± 8.84%, P = 0.060) was observed. Additional, there was no significant difference in the time to peak of creatine kinase isoenzyme-MB ((12.97 ± 5.67) h vs. (14.31 ± 7.04) h, P = 0.557) and high-sensitivity cardiac troponin T ((21.03 ± 12.46) h vs. (21.44 ± 9.99) h, P = 0.920) between the 2 groups. During the 6-month follow-up period, only 1 patient in the PCSK9 inhibitor group developed cerebral hemorrhage 6 months after PCI. Conclusions::Early treatment with alirocumab did not exhibit a significant increase in MSI at 5-10 d in patients with anterior STEMI. Larger trials are necessary to evaluate the impact of early administration of PCSK9 inhibitors after myocardial infarction.