Tumor drug resistance is an important problem in the failure of chemotherapy and targeted drug therapy, which is a complex process involving chromatin remodeling. SWI/SNF is one of the most studied ATP-dependent chromatin remodeling complexes in tumorigenesis, which plays an important role in the coordination of chromatin structural stability, gene expression, and post-translation modification. However, its mechanism in tumor drug resistance has not been systematically combed. SWI/SNF can be divided into 3 types according to its subunit composition: BAF, PBAF, and ncBAF. These 3 subtypes all contain two mutually exclusive ATPase catalytic subunits (SMARCA2 or SMARCA4), core subunits (SMARCC1 and SMARCD1), and regulatory subunits (ARID1A, PBRM1, and ACTB, etc.), which can control gene expression by regulating chromatin structure. The change of SWI/SNF complex subunits is one of the important factors of tumor drug resistance and progress. SMARCA4 and ARID1A are the most widely studied subunits in tumor drug resistance. Low expression of SMARCA4 can lead to the deletion of the transcription inhibitor of the BCL2L1 gene in mantle cell lymphoma, which will result in transcription up-regulation and significant resistance to the combination therapy of ibrutinib and venetoclax. Low expression of SMARCA4 and high expression of SMARCA2 can activate the FGFR1-pERK1/2 signaling pathway in ovarian high-grade serous carcinoma cells, which induces the overexpression of anti-apoptosis gene BCL2 and results in carboplatin resistance. SMARCA4 deletion can up-regulate epithelial-mesenchymal transition (EMT) by activating YAP1 gene expression in triple-negative breast cancer. It can also reduce the expression of Ca²⁺ channel IP3R3 in ovarian and lung cancer, resulting in the transfer of Ca²⁺ needed to induce apoptosis from endoplasmic reticulum to mitochondria damage. Thus, these two tumors are resistant to cisplatin. It has been found that verteporfin can overcome the drug resistance induced by SMARCA4 deletion. However, this inhibitor has not been applied in clinical practice. Therefore, it is a promising research direction to develop SWI/SNF ATPase targeted drugs with high oral bioavailability to treat patients with tumor resistance induced by low expression or deletion of SMARCA4. ARID1A deletion can activate the expression of ANXA1 protein in HER2+ breast cancer cells or down-regulate the expression of progesterone receptor B protein in endometrial cancer cells. The drug resistance of these two tumor cells to trastuzumab or progesterone is induced by activating AKT pathway. ARID1A deletion in ovarian cancer can increase the expression of MRP2 protein and make it resistant to carboplatin and paclitaxel. ARID1A deletion also can up-regulate the phosphorylation levels of EGFR, ErbB2, and RAF1 oncogene proteins.The ErbB and VEGF pathway are activated and EMT is increased. As a result, lung adenocarcinoma is resistant to epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs). Although great progress has been made in the research on the mechanism of SWI/SNF complex inducing tumor drug resistance, most of the research is still at the protein level. It is necessary to comprehensively and deeply explore the detailed mechanism of drug resistance from gene, transcription, protein, and metabolite levels by using multi-omics techniques, which can provide sufficient theoretical basis for the diagnosis and treatment of poor tumor prognosis caused by mutation or abnormal expression of SWI/SNF subunits in clinical practice.

During orthodontic treatment, clinical monitoring of patients is a crucial factor in determining treatment success. It aids in timely problem detection and resolution, ensuring adherence to the intended treatment plan. In recent years, digital technology has increasingly permeated orthodontic clinical diagnosis and treatment, facilitating clinical decision-making, treatment planning, and follow-up monitoring. This review summarizes recent advancements in digital technology for monitoring orthodontic tooth movement, related complications, and appliance-wearing compliance. It aims to provide insights for researchers and clinicians to enhance the application of digital technology in orthodontics, improve treatment outcomes, and optimize patient experience. The digitization of diagnostic data and the visualization of dental models make chair-side follow-up monitoring more convenient, accurate, and efficient. At the same time, the emergence of remote monitoring technology allows orthodontists to promptly identify oral health issues in patients and take corresponding measures. Furthermore, the multimodal data fusion method offers valuable insights into the monitoring of the root-alveolar relationship. Artificial intelligence technology has made initial strides in automating the identification of orthodontic tooth movement, associated complications, and patient compliance evaluation. Sensors are effective tools for monitoring patient adherence and providing data-driven support for clinical decision-making. The application of digital technology in orthodontic monitoring holds great promise. However, challenges like technical bottlenecks, ethical considerations, and patient acceptance remain.

Objective:To explore the clinical application value of convolutional neural network（CNN）based on deep learning in the automatic measurement of dynamic pelvic floor ultrasound video parameters and the diagnosis and classification of cystocele.Methods:A retrospective analysis was conducted on dynamic pelvic floor ultrasound videos from 398 postpartum women who underwent examinations at the First People's Hospital of Foshan between June 2020 and June 2022. The lowest point of the posterior bladder wall（PWB），urethral rotation angle（URA），and retrovesical angle（RVA）were manually measured by a senior radiologist（R1）and a junior radiologist（R2），and cystocele was classified according to the Green standard. The CNN model was employed to automatically extract the above parameters and to diagnose and classify cystocele. Using R1 measurements as a reference，intraclass correlation coefficient（ICC）was used to evaluate the consistency between the CNN model and R1，as well as between R2 and R1. The Kappa value was used to assess the agreement between the CNN model，R2，and R1 in the diagnosis and classification of cystocele. Additionally，the time consumption of the three measurement methods was compared.Results:The CNN model showed good consistency with R1 in measuring PWB and URA（ICC = 0.983，0.894），while its consistency in measuring RVA was moderate（ICC = 0.614）. The ICC between R2 and R1 in measuring PWB，URA，and RVA was 0.979，0.815，and 0.627，respectively. In the measurement of PWB and URA，the consistency between the CNN model and R1 was superior to that between R2 and R1. For cystocele diagnosis，the Kappa value between the CNN model and R1 was 0.924，which was higher than that between R2 and R1（0.904）. In cystocele classification，the Kappa value between the CNN model and R1 was 0.503，also higher than that between R2 and R1（0.426）. The CNN model processed a single video in 2.5（0.6）s，significantly faster than R1［59.9（16.9）s］and R2［56.8（11.2）s］（all P < 0.001）. Conclusions:The CNN model demonstrates high accuracy and efficiency in the measurement，diagnosis，and classification of cystocele，outperforming a junior radiologist and showing potential for clinical application.

Objective:To explore the clinical features and risk factors for pediatric severe Mycoplasma pneumoniae pneumonia (SMPP) complicated with pulmonary embolism. Methods:SMPP patients admitted to Department of Pediatrics, Jiaxing First Hospital and Pediatric Intensive Care Unit, Shanghai Children′s Hospital from December 2019 to December 2023 were included in this retrospective case-control study.According to whether they were complicated with pulmonary embolism, SMPP patients were divided into a pulmonary embolism group and a non-pulmonary embolism group.Clinical characteristics of the two groups, including general data, laboratory examination and imaging data were compared and analyzed.The t-test and Mann-Whitney rank-sum test were used to compare the measurement data, and the χ2 test was used to compare the count data.The risk factors of SMPP patients developing pulmonary embolism were analyzed by the univariate method. Results:There were 10 out of 62 SMPP children developing pulmonary embolism, showing an incidence of 16.13%.In the pulmonary embolism group, there were 5 boys and 5 girls, with a median age of 7.50 (5.75, 9.25) years.There were 52 children in the non-pulmonary embolism group, including 29 boys and 23 girls, with a median age of 6.50(5.00, 8.00)years.The hospitalization time, body temperature, total white blood cell count, neutrophil count, C-reactive protein levels, lactate dehydrogenase levels, prothrombin time, activated partial thromboplastin time, D-dimer (D-D) levels, fibrinogen degradation product levels, pleural effusion and atelectasis rates in the pulmonary embolism group were significantly higher than those in the non-pulmonary embolism group (all P<0.05). Fibrinogen levels in the pulmonary embolism group were significantly lower than those in the non-pulmonary embolism group ( P<0.05). Univariate Logistic regression analysis showed that the D-D level was a risk factor for SMPP patient developing pulmonary embolism.The receiver operating characteristic curve analysis revealed that the D-D level had the largest area under the curve for predicting pulmonary embolism of 0.990(95% CI: 0.972-1.000, P<0.001), with a sensitivity of 100%, a specificity of 92%, and a cut-off value of 4.63 mg/L. Conclusions:SMPP children complicated with pulmonary embolism are prone to high inflammation and impaired coagulation function.The increase of D-D levels is a risk factor for the development of pulmonary embolism in SMPP.

Objective:To evaluate the value of ultrasound radiomics combined with machine learning for early diagnosis of seronegative Hashimoto’s thyroiditis (SN-HT) .Methods:This retrospective study included 164 patients from Liaoning Provincial People’s Hospital , Lixin County People’s Hospital, Linghai Dalinghe Hospital, Fengcheng Phoenix Hospital, who underwent thyroidectomy for solitary nodules with normal thyroid function between Nov. 2016 and Jan. 2024. Postoperative pathology confirmed Hashimoto’s thyroiditis (HT) in some cases, who were further categorized into antibody-positive and antibody-negative groups based on serum antibody status. Patients without Hashimoto’s thyroiditis served as the control group. A total of 298 ultrasound images were analyzed. Radiomics features were extracted from hypoechoic non-nodular areas within 0.5 cm surrounding the tumor. Two senior pathologists and two senior ultrasound physicians independently assessed lymphocytic infiltration, eosinophilic changes of follicular epithelium, and the proportion of hypoechoic areas in pathology and ultrasound images, respectively. A machine learning model, CCH-NET, was developed using linear regression and t-distributed stochastic neighbor embedding (t-SNE) techniques. The dataset was divided into a training set (80%) and a validation set (20%) to compare the diagnostic accuracy of CCH-NET with that of senior ultrasound physicians. Results:In internal validation, CCH-NET achieved a diagnostic accuracy of 88.89% for both antibody-positive and antibody-negative groups, significantly higher than the 66.67% accuracy of senior ultrasound physicians ( P<0.01). In external validation, CCH-NET achieved 75.00% and 66.67% accuracy for the two groups, compared to 50.00% by senior ultrasound physicians. For the control group, both methods achieved 93.33% accuracy. The AUC of CCH-NET was 0.848, outperforming senior ultrasound physicians (0.681) ,demonstrating superior diagnostic performance. Conclusion:The radiomics-based CCH-NET model, using non-nodular hypoechoic areas as a specific indicator, can accurately identify early SN-HT in euthyroid patients. It significantly outperforms senior ultrasound physicians, improving diagnostic accuracy and reducing missed diagnoses.

Objective To investigate the value of MRI of the sternal end of clavicle in bone age assessment in Chinese population,especially its applicability in the determination of criminal responsible age.Methods A total of 431 patients aged from 10.00 to 29.99 years with neck or chest MRI were retrospectively collected.According to the Schmeling grading method,the epiphyseal development of the clavicle MRI was divided into five grades.The consistency of methods was evaluated.The correlation and general descriptive analysis between MRI grades and age was analyzed.The sex difference was analyzed.Curve fitting was used to establish a nonlinear model between age and grades.Results The grades of clavicle MRI showed a significant age-related trend(Figure 2),and the correlation was 0.861(0.887 in males and 0.840 in females).Except for grade 1,there was no significant difference between males and females in other grades.The minimum age of male grade 3 was greater than 14 years old,and the minimum age of female grade 3 was greater than 16 years old.The minimum age in grade 4 and grade 5 was over 18 years old in both sexes.The best curve fitting model was cubic model for both sexes(R2=0.805 for men and 0.722 for women).Conclusion Clavicle MRI can be used for the assessment of bone age in Chinese population.Complete epiphyseal plate closure can be used as a reliable indicator for the determination of age at 18 years old,and it is expected to achieve radiation-free forensic bone age assessment.

Objective:To evaluate the value of ultrasound radiomics combined with machine learning for early diagnosis of seronegative Hashimoto’s thyroiditis (SN-HT) .Methods:This retrospective study included 164 patients from Liaoning Provincial People’s Hospital , Lixin County People’s Hospital, Linghai Dalinghe Hospital, Fengcheng Phoenix Hospital, who underwent thyroidectomy for solitary nodules with normal thyroid function between Nov. 2016 and Jan. 2024. Postoperative pathology confirmed Hashimoto’s thyroiditis (HT) in some cases, who were further categorized into antibody-positive and antibody-negative groups based on serum antibody status. Patients without Hashimoto’s thyroiditis served as the control group. A total of 298 ultrasound images were analyzed. Radiomics features were extracted from hypoechoic non-nodular areas within 0.5 cm surrounding the tumor. Two senior pathologists and two senior ultrasound physicians independently assessed lymphocytic infiltration, eosinophilic changes of follicular epithelium, and the proportion of hypoechoic areas in pathology and ultrasound images, respectively. A machine learning model, CCH-NET, was developed using linear regression and t-distributed stochastic neighbor embedding (t-SNE) techniques. The dataset was divided into a training set (80%) and a validation set (20%) to compare the diagnostic accuracy of CCH-NET with that of senior ultrasound physicians. Results:In internal validation, CCH-NET achieved a diagnostic accuracy of 88.89% for both antibody-positive and antibody-negative groups, significantly higher than the 66.67% accuracy of senior ultrasound physicians ( P<0.01). In external validation, CCH-NET achieved 75.00% and 66.67% accuracy for the two groups, compared to 50.00% by senior ultrasound physicians. For the control group, both methods achieved 93.33% accuracy. The AUC of CCH-NET was 0.848, outperforming senior ultrasound physicians (0.681) ,demonstrating superior diagnostic performance. Conclusion:The radiomics-based CCH-NET model, using non-nodular hypoechoic areas as a specific indicator, can accurately identify early SN-HT in euthyroid patients. It significantly outperforms senior ultrasound physicians, improving diagnostic accuracy and reducing missed diagnoses.

Objective:To explore the clinical features and risk factors for pediatric severe Mycoplasma pneumoniae pneumonia (SMPP) complicated with pulmonary embolism. Methods:SMPP patients admitted to Department of Pediatrics, Jiaxing First Hospital and Pediatric Intensive Care Unit, Shanghai Children′s Hospital from December 2019 to December 2023 were included in this retrospective case-control study.According to whether they were complicated with pulmonary embolism, SMPP patients were divided into a pulmonary embolism group and a non-pulmonary embolism group.Clinical characteristics of the two groups, including general data, laboratory examination and imaging data were compared and analyzed.The t-test and Mann-Whitney rank-sum test were used to compare the measurement data, and the χ2 test was used to compare the count data.The risk factors of SMPP patients developing pulmonary embolism were analyzed by the univariate method. Results:There were 10 out of 62 SMPP children developing pulmonary embolism, showing an incidence of 16.13%.In the pulmonary embolism group, there were 5 boys and 5 girls, with a median age of 7.50 (5.75, 9.25) years.There were 52 children in the non-pulmonary embolism group, including 29 boys and 23 girls, with a median age of 6.50(5.00, 8.00)years.The hospitalization time, body temperature, total white blood cell count, neutrophil count, C-reactive protein levels, lactate dehydrogenase levels, prothrombin time, activated partial thromboplastin time, D-dimer (D-D) levels, fibrinogen degradation product levels, pleural effusion and atelectasis rates in the pulmonary embolism group were significantly higher than those in the non-pulmonary embolism group (all P<0.05). Fibrinogen levels in the pulmonary embolism group were significantly lower than those in the non-pulmonary embolism group ( P<0.05). Univariate Logistic regression analysis showed that the D-D level was a risk factor for SMPP patient developing pulmonary embolism.The receiver operating characteristic curve analysis revealed that the D-D level had the largest area under the curve for predicting pulmonary embolism of 0.990(95% CI: 0.972-1.000, P<0.001), with a sensitivity of 100%, a specificity of 92%, and a cut-off value of 4.63 mg/L. Conclusions:SMPP children complicated with pulmonary embolism are prone to high inflammation and impaired coagulation function.The increase of D-D levels is a risk factor for the development of pulmonary embolism in SMPP.

Objective To investigate the value of MRI of the sternal end of clavicle in bone age assessment in Chinese population,especially its applicability in the determination of criminal responsible age.Methods A total of 431 patients aged from 10.00 to 29.99 years with neck or chest MRI were retrospectively collected.According to the Schmeling grading method,the epiphyseal development of the clavicle MRI was divided into five grades.The consistency of methods was evaluated.The correlation and general descriptive analysis between MRI grades and age was analyzed.The sex difference was analyzed.Curve fitting was used to establish a nonlinear model between age and grades.Results The grades of clavicle MRI showed a significant age-related trend(Figure 2),and the correlation was 0.861(0.887 in males and 0.840 in females).Except for grade 1,there was no significant difference between males and females in other grades.The minimum age of male grade 3 was greater than 14 years old,and the minimum age of female grade 3 was greater than 16 years old.The minimum age in grade 4 and grade 5 was over 18 years old in both sexes.The best curve fitting model was cubic model for both sexes(R2=0.805 for men and 0.722 for women).Conclusion Clavicle MRI can be used for the assessment of bone age in Chinese population.Complete epiphyseal plate closure can be used as a reliable indicator for the determination of age at 18 years old,and it is expected to achieve radiation-free forensic bone age assessment.

Objective:To explore the clinical application value of convolutional neural network（CNN）based on deep learning in the automatic measurement of dynamic pelvic floor ultrasound video parameters and the diagnosis and classification of cystocele.Methods:A retrospective analysis was conducted on dynamic pelvic floor ultrasound videos from 398 postpartum women who underwent examinations at the First People's Hospital of Foshan between June 2020 and June 2022. The lowest point of the posterior bladder wall（PWB），urethral rotation angle（URA），and retrovesical angle（RVA）were manually measured by a senior radiologist（R1）and a junior radiologist（R2），and cystocele was classified according to the Green standard. The CNN model was employed to automatically extract the above parameters and to diagnose and classify cystocele. Using R1 measurements as a reference，intraclass correlation coefficient（ICC）was used to evaluate the consistency between the CNN model and R1，as well as between R2 and R1. The Kappa value was used to assess the agreement between the CNN model，R2，and R1 in the diagnosis and classification of cystocele. Additionally，the time consumption of the three measurement methods was compared.Results:The CNN model showed good consistency with R1 in measuring PWB and URA（ICC = 0.983，0.894），while its consistency in measuring RVA was moderate（ICC = 0.614）. The ICC between R2 and R1 in measuring PWB，URA，and RVA was 0.979，0.815，and 0.627，respectively. In the measurement of PWB and URA，the consistency between the CNN model and R1 was superior to that between R2 and R1. For cystocele diagnosis，the Kappa value between the CNN model and R1 was 0.924，which was higher than that between R2 and R1（0.904）. In cystocele classification，the Kappa value between the CNN model and R1 was 0.503，also higher than that between R2 and R1（0.426）. The CNN model processed a single video in 2.5（0.6）s，significantly faster than R1［59.9（16.9）s］and R2［56.8（11.2）s］（all P < 0.001）. Conclusions:The CNN model demonstrates high accuracy and efficiency in the measurement，diagnosis，and classification of cystocele，outperforming a junior radiologist and showing potential for clinical application.