Objective:Glioblastoma(GBM)and brain metastases(BMs)are the two most common malignant brain tumors in adults.Magnetic resonance imaging(MRI)is a commonly used method for screening and evaluating the prognosis of brain tumors,but the specificity and sensitivity of conventional MRI sequences in differential diagnosis of GBM and BMs are limited.In recent years,deep neural network has shown great potential in the realization of diagnostic classification and the establishment of clinical decision support system.This study aims to apply the radiomics features extracted by deep learning techniques to explore the feasibility of accurate preoperative classification for newly diagnosed GBM and solitary brain metastases(SBMs),and to further explore the impact of multimodality data fusion on classification tasks. Methods:Standard protocol cranial MRI sequence data from 135 newly diagnosed GBM patients and 73 patients with SBMs confirmed by histopathologic or clinical diagnosis were retrospectively analyzed.First,structural T1-weight,T1C-weight,and T2-weight were selected as 3 inputs to the entire model,regions of interest(ROIs)were manually delineated on the registered three modal MR images,and multimodality radiomics features were obtained,dimensions were reduced using a random forest(RF)-based feature selection method,and the importance of each feature was further analyzed.Secondly,we used the method of contrast disentangled to find the shared features and complementary features between different modal features.Finally,the response of each sample to GBM and SBMs was predicted by fusing 2 features from different modalities. Results:The radiomics features using machine learning and the multi-modal fusion method had a good discriminatory ability for GBM and SBMs.Furthermore,compared with single-modal data,the multimodal fusion models using machine learning algorithms such as support vector machine(SVM),Logistic regression,RF,adaptive boosting(AdaBoost),and gradient boosting decision tree(GBDT)achieved significant improvements,with area under the curve(AUC)values of 0.974,0.978,0.943,0.938,and 0.947,respectively;our comparative disentangled multi-modal MR fusion method performs well,and the results of AUC,accuracy(ACC),sensitivity(SEN)and specificity(SPE)in the test set were 0.985,0.984,0.900,and 0.990,respectively.Compared with other multi-modal fusion methods,AUC,ACC,and SEN in this study all achieved the best performance.In the ablation experiment to verify the effects of each module component in this study,AUC,ACC,and SEN increased by 1.6%,10.9%and 15.0%,respectively after 3 loss functions were used simultaneously. Conclusion:A deep learning-based contrast disentangled multi-modal MR radiomics feature fusion technique helps to improve GBM and SBMs classification accuracy.

BACKGROUND:Heterotopic ossification is a dynamic growth process.Diverse heterotopic ossification subtypes have diverse etiologies or induction factors,but they exhibit a similar clinical process in the intermediate and later phases of the disease.Acquired heterotopic ossification produced by trauma and other circumstances has a high incidence. OBJECTIVE:To summarize the molecular biological mechanisms linked to the occurrence and progression of acquired heterotopic ossification in recent years. METHODS:The keywords"molecular biology,heterotopic ossification,mechanisms"were searched in CNKI,Wanfang,PubMed,Embase,Web of Science,and Google Scholar databases for articles published from January 2016 to August 2022.Supplementary searches were conducted based on the obtained articles.After the collected literature was screened,131 articles were finally included and summarized. RESULTS AND CONCLUSION:(1)The occurrence and development of acquired heterotopic ossification is a dynamic process with certain concealment,making diagnosis and treatment of the disease difficult.(2)By reviewing relevant literature,it was found that acquired heterotopic ossification involves signaling pathways such as bone morphogenetic protein,transforming growth factor-β,Hedgehog,Wnt,and mTOR,as well as core factors such as Runx-2,vascular endothelial growth factor,hypoxia-inducing factor,fibroblast growth factor,and Sox9.The core mechanism may be the interaction between different signaling pathways,affecting the body's osteoblast precursor cells,osteoblast microenvironment,and related cytokines,thereby affecting the body's bone metabolism and leading to the occurrence of acquired heterotopic ossification.(3)In the future,it is possible to take the heterotopic ossification-related single-cell osteogenic homeostasis as the research direction,take the osteoblast precursor cells-osteogenic microenvironment-signaling pathways and cytokines as the research elements,explore the characteristics of each element under different temporal and spatial conditions,compare the similarities and differences of the osteogenic homeostasis of different types and individuals,observe the regulatory mechanism of the molecular signaling network of heterotopic ossification from a holistic perspective.It is beneficial to the exploration of new methods for the future clinical prevention and treatment of heterotopic ossification.(4)Meanwhile,the treatment methods represented by traditional Chinese medicine and targeted therapy have become research hotspots in recent years.How to link traditional Chinese medicine with the osteogenic homeostasis in the body and combine it with targeted therapy is also one of the future research directions.(5)At present,the research on acquired heterotopic ossification is still limited to basic experimental research and the clinical prevention and treatment methods still have defects such as uncertain efficacy and obvious side effects.The safety and effectiveness of relevant targeted prevention and treatment drugs in clinical application still need to be verified.Future research should focus on clinical prevention and treatment based on basic experimental research combined with the mechanism of occurrence and development.

Objective To explore the automated identification and diameter measurement methods for inferior vena cava (IVC) based on clinical ultrasound images of IVC. Methods An automated identification and localization method based on topology and automatic tracking algorithm was proposed. Tracking algorithm was used for identifying and continuously locating to improve the efficiency and accuracy of measurement. Tests were conducted on 18 sets of ultrasound data collected from 18 patients in intensive care unit (ICU),with clinicians' measurements as the gold standard. Results The recognition accuracy of the automated method was 94.44％ (17/18),and the measurement error of IVC diameter was within the range of±1.96s (s was the standard deviation). The automated method could replace the manual method. Conclusion The proposed IVC automated identification and localization algorithm based on topology and automatic tracking algorithm has high recognition success rate and IVC diameter measurement accuracy. It can assist clinicians in identifying and locating IVC,so as to improve the accuracy of IVC measurement.

Objective: To investigate the correlation between magnetic resonance imaging-based vertebral bone quality (VBQ) score and screw loosening after dynamic pedicle screw fixation with polyetheretherketone (PEEK) rods, and evaluate its predictive value. Methods: A retrospective analysis was conducted on the patients who underwent dynamic pedicle screw fixation with PEEK rods from March 2017 to June 2022. Data on age, sex, body mass index, hypertension, diabetes, hyperlipidemia history, long-term smoking, alcohol consumption, VBQ score, L1–4 average Hounsfield unit (HU) value, surgical fixation length, and the lowest instrumented vertebra were collected. Logistic regression analysis was employed to assess the relationship between VBQ score and pedicle screw loosening (PSL). Results: A total of 24 patients experienced PSL after surgery (20.5%). PSL group and non-PSL group showed statistical differences in age, number of fixed segments, fixation to the sacrum, L1–4 average HU value, and VBQ score (p < 0.05). The VBQ score in the PSL group was higher than that in the non-PSL group (3.56 ± 0.45 vs. 2.77 ± 0.31, p < 0.001). In logistic regression analysis, VBQ score (odds ratio, 3.425; 95% confidence interval, 1.552–8.279) were identified as independent risk factors for screw loosening. The area under the receiver operating characteristic curve for VBQ score predicting PSL was 0.819 (p < 0.05), with the optimal threshold of 3.15 (sensitivity, 83.1%; specificity, 80.5%). Conclusion The VBQ score can independently predict postoperative screw loosening in patients undergoing lumbar dynamic pedicle screw fixation with PEEK rods, and its predictive value is comparable to HU value.

Objective: To investigate the correlation between magnetic resonance imaging-based vertebral bone quality (VBQ) score and screw loosening after dynamic pedicle screw fixation with polyetheretherketone (PEEK) rods, and evaluate its predictive value. Methods: A retrospective analysis was conducted on the patients who underwent dynamic pedicle screw fixation with PEEK rods from March 2017 to June 2022. Data on age, sex, body mass index, hypertension, diabetes, hyperlipidemia history, long-term smoking, alcohol consumption, VBQ score, L1–4 average Hounsfield unit (HU) value, surgical fixation length, and the lowest instrumented vertebra were collected. Logistic regression analysis was employed to assess the relationship between VBQ score and pedicle screw loosening (PSL). Results: A total of 24 patients experienced PSL after surgery (20.5%). PSL group and non-PSL group showed statistical differences in age, number of fixed segments, fixation to the sacrum, L1–4 average HU value, and VBQ score (p < 0.05). The VBQ score in the PSL group was higher than that in the non-PSL group (3.56 ± 0.45 vs. 2.77 ± 0.31, p < 0.001). In logistic regression analysis, VBQ score (odds ratio, 3.425; 95% confidence interval, 1.552–8.279) were identified as independent risk factors for screw loosening. The area under the receiver operating characteristic curve for VBQ score predicting PSL was 0.819 (p < 0.05), with the optimal threshold of 3.15 (sensitivity, 83.1%; specificity, 80.5%). Conclusion The VBQ score can independently predict postoperative screw loosening in patients undergoing lumbar dynamic pedicle screw fixation with PEEK rods, and its predictive value is comparable to HU value.

Objective: To investigate the correlation between magnetic resonance imaging-based vertebral bone quality (VBQ) score and screw loosening after dynamic pedicle screw fixation with polyetheretherketone (PEEK) rods, and evaluate its predictive value. Methods: A retrospective analysis was conducted on the patients who underwent dynamic pedicle screw fixation with PEEK rods from March 2017 to June 2022. Data on age, sex, body mass index, hypertension, diabetes, hyperlipidemia history, long-term smoking, alcohol consumption, VBQ score, L1–4 average Hounsfield unit (HU) value, surgical fixation length, and the lowest instrumented vertebra were collected. Logistic regression analysis was employed to assess the relationship between VBQ score and pedicle screw loosening (PSL). Results: A total of 24 patients experienced PSL after surgery (20.5%). PSL group and non-PSL group showed statistical differences in age, number of fixed segments, fixation to the sacrum, L1–4 average HU value, and VBQ score (p < 0.05). The VBQ score in the PSL group was higher than that in the non-PSL group (3.56 ± 0.45 vs. 2.77 ± 0.31, p < 0.001). In logistic regression analysis, VBQ score (odds ratio, 3.425; 95% confidence interval, 1.552–8.279) were identified as independent risk factors for screw loosening. The area under the receiver operating characteristic curve for VBQ score predicting PSL was 0.819 (p < 0.05), with the optimal threshold of 3.15 (sensitivity, 83.1%; specificity, 80.5%). Conclusion The VBQ score can independently predict postoperative screw loosening in patients undergoing lumbar dynamic pedicle screw fixation with PEEK rods, and its predictive value is comparable to HU value.

Objective: To investigate the correlation between magnetic resonance imaging-based vertebral bone quality (VBQ) score and screw loosening after dynamic pedicle screw fixation with polyetheretherketone (PEEK) rods, and evaluate its predictive value. Methods: A retrospective analysis was conducted on the patients who underwent dynamic pedicle screw fixation with PEEK rods from March 2017 to June 2022. Data on age, sex, body mass index, hypertension, diabetes, hyperlipidemia history, long-term smoking, alcohol consumption, VBQ score, L1–4 average Hounsfield unit (HU) value, surgical fixation length, and the lowest instrumented vertebra were collected. Logistic regression analysis was employed to assess the relationship between VBQ score and pedicle screw loosening (PSL). Results: A total of 24 patients experienced PSL after surgery (20.5%). PSL group and non-PSL group showed statistical differences in age, number of fixed segments, fixation to the sacrum, L1–4 average HU value, and VBQ score (p < 0.05). The VBQ score in the PSL group was higher than that in the non-PSL group (3.56 ± 0.45 vs. 2.77 ± 0.31, p < 0.001). In logistic regression analysis, VBQ score (odds ratio, 3.425; 95% confidence interval, 1.552–8.279) were identified as independent risk factors for screw loosening. The area under the receiver operating characteristic curve for VBQ score predicting PSL was 0.819 (p < 0.05), with the optimal threshold of 3.15 (sensitivity, 83.1%; specificity, 80.5%). Conclusion The VBQ score can independently predict postoperative screw loosening in patients undergoing lumbar dynamic pedicle screw fixation with PEEK rods, and its predictive value is comparable to HU value.

Objective: To investigate the correlation between magnetic resonance imaging-based vertebral bone quality (VBQ) score and screw loosening after dynamic pedicle screw fixation with polyetheretherketone (PEEK) rods, and evaluate its predictive value. Methods: A retrospective analysis was conducted on the patients who underwent dynamic pedicle screw fixation with PEEK rods from March 2017 to June 2022. Data on age, sex, body mass index, hypertension, diabetes, hyperlipidemia history, long-term smoking, alcohol consumption, VBQ score, L1–4 average Hounsfield unit (HU) value, surgical fixation length, and the lowest instrumented vertebra were collected. Logistic regression analysis was employed to assess the relationship between VBQ score and pedicle screw loosening (PSL). Results: A total of 24 patients experienced PSL after surgery (20.5%). PSL group and non-PSL group showed statistical differences in age, number of fixed segments, fixation to the sacrum, L1–4 average HU value, and VBQ score (p < 0.05). The VBQ score in the PSL group was higher than that in the non-PSL group (3.56 ± 0.45 vs. 2.77 ± 0.31, p < 0.001). In logistic regression analysis, VBQ score (odds ratio, 3.425; 95% confidence interval, 1.552–8.279) were identified as independent risk factors for screw loosening. The area under the receiver operating characteristic curve for VBQ score predicting PSL was 0.819 (p < 0.05), with the optimal threshold of 3.15 (sensitivity, 83.1%; specificity, 80.5%). Conclusion The VBQ score can independently predict postoperative screw loosening in patients undergoing lumbar dynamic pedicle screw fixation with PEEK rods, and its predictive value is comparable to HU value.

ObjectiveTo investigate the protective effect and mechanism of Euphorbia helioscopia aqueous extract （EHE） on mice with chronic obstructive pulmonary disease （COPD） and its influence on precancerous lesion-associated proteins in lung tissues induced by cigarette smoke （CS）. MethodThe COPD model was induced by CS in 60 mice and the model mice were randomly divided into control group， model group， positive drug group （dexamethasone， 2 mg·kg^-1）， and low-， medium-， and high-dose EHE groups （1.875， 3.75， 7.5 g·kg^-1）. The high-performance liquid chromatography （HPLC） method was used to determine the related components in EHE. The changes in end-expiratory pause （EEP）， airway resistance （Penh）， expiratory flow at 50% vital capacity （EF50）， and other pulmonary function indexes were detected by the spirometer. The levels of inflammatory factors， such as interleukin （IL）-2， IL-5， IL-18， IL-17A， and IL-27 in bronchoalveolar lavage fluid （BALF） of mice were detected by high-throughput liquid protein chip technology. Hematoxylin-eosin （HE） staining was used to detect the pathological changes in lung tissues in mice. The content of malondialdehyde （MDA）， myeloperoxidase （MPO）， and glutathione peroxidase （GSH-Px） in lung tissues was determined by the colorimetric method. The mRNA relative expression of tumor necrosis factor-α （TNF-α）， transforming growth factor-β （TGF-β）， matrix metalloproteinase-2 （MMP-2）， matrix metalloproteinase-9 （MMP-9）， and matrix metalloproteinase-12 （MMP-12） was detected by Real-time fluorescence quantitative polymerase chain reaction （Real-time PCR）. Immunohistochemistry （IHC） was used to detect the expression of tumor protein （P53） and cell proliferation-associated antigen （Ki67） in lung tissues， and Western blot was used to detect the relative expression of tumor suppressor protein （P16）， DNA （cytosine-5）-methyltransferase 1 （DNMT1）， and fragile histidine triad （FHIT） in lung tissues. ResultThe results showed that the main compounds in EHE included phenols （gallic acid and protocatechuic acid） and flavonoids （such as hyperoside， rutin， myricetin， naringenin， quercetin， luteolin， kaempferol， and licorice chalcone A）， among which gallic acid and rutin were the highest in content. Compared with normal group， model group showed increased levels of EEP， EF50， and Penh （P<0.05）， and showed increased MDA and MPO levels （P<0.01） and decreased GSH-Px （P<0.01）， and the model group displayed increased levels of IL-2， IL-5， IL-18， IL-17A， IL-27， TNF-α， TGF-β， MMP-2， MMP-9， and MMP-12 （P<0.05）. And the model group exhibited up-regulated expression of P53， Ki67， and FHIT in lung tissues （P<0.01） and down-regulated expression of DNMT1 and P16 （P<0.01）. Compared with model group， the EHE groups showed decreased EEP and EF50 levels （P<0.05）. The pathological injury of lung tissues in mice of the model group was observed under HE staining， and the pathological injury of basal cell hyperplasia of lung tissues was gradually improved after treatment with EHE. The EHE groups showed reduced levels of MDA and MPO （P<0.01） and increased GSH-Px （P<0.01）. The EHE groups displayed decreased levels of IL-2， IL-5， IL-18， IL-17A， IL-27， TNF-α， TGF-β， MMP-2， MMP-9， and MMP-12 （P<0.05）. And the EHE groups showed down-regulated Ki67 and FHIT in lung tissues （P<0.05） and up-regulated expression of P53 and DNMT1 （P<0.05）. ConclusionEHE can protect mice from COPD and inhibit precancerous lesions， and the mechanism may be related to the inhibition of inflammation and oxidative stress response， regulation of protease and antiprotease imbalance， and regulation of epithelial cell growth.

Objective:To explore the feasibility of deep learning-based restoration of obscured thyroid ultrasound images.Methods:A total of 358 images of thyroid nodules were retropectively collected from January 2020 to October 2021 at Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, and the images were randomly masked and restored using DeepFillv2. The difference in grey values between the images before and after restoration was compared, and 6 sonographers (2 chief physicians, 2 attending physicians, 2 residents) were invited to compare the rate of correctness of judgement and detection of image discrepancies. The ultrasound features of thyroid nodules (solid composition, microcalcifications, markedly hypoechoic, ill-defined or irregular margins, or extrathyroidal extensions, vertical orientation and comet-tail artifact) were extracted according to the Chinese Thyroid Imaging Reporting and Data System (C-TIRADS). The consistency of ultrasound features of thyroid nodules before and after restoration were compared.Results:The mean squared error of the images before and after restoration ranged from 0.274 to 0.522, and there were significant differences in the rate of correctness of judgement and detection of image discrepancies between physicians of different groups(all P<0.001). The overall accuracy rate was 51.95%, the overall detection rate was 1.79%, there were significant differences also within the chief physicians and resident groups (all P<0.001). The agreement rate of all ultrasound features of the nodules before and after image restoration was higher than 70%, over 90% agreement rate for features such as solid composition and comet-tail artifact. Conclusions:The algorithm can effectively repair obscured thyroid ultrasound images while preserving image features, which is expected to expand the deep learning image database, and promote the development of deep learning in the field of ultrasound images.