Objective:To construct a prior based on the inherent properties of PET to accurately segment the lesion areas.Methods:A network framework for PET tumor segmentation driven by geodesic priors was proposed (geodesic network for short). Specifically, partial differential equations were constructed to characterize the geodesic distances between different regions in PET images. Tumor marker points identified by CT labeling were used as the initial conditions for the equations. To enhance the contrast between areas of lung or breast tumors and normal tissues, a smooth Heaviside function was utilized to map the geodesic distances. The network framework adopted a dual-branch architecture, using geodesic priors to assist in PET image segmentation.Results:The proposed method achieved a Dice coefficient of 94.92% in lung cancer segmentation and 90.12% in breast cancer segmentation. With the addition of geodesic priors in the Unet, the Dice coefficient for breast cancer increased by 32.37% (from 42.50% to 74.87%).Conclusion:Geodesic priors can significantly improve segmentation outcomes and enhance the generalization capability of the network.

Objective:To explore the value of chromosomal microarray analysis (CMA) in the genetic diagnosis of different types of fetal growth restriction (FGR).Methods:A retrospective analysis was conducted on 120 cases who were diagnosed with FGR by ultrasound and underwent prenatal diagnosis at the Department of Obstetrics & Gynecology and Reproductive Medicine, Peking University First Hospital, from January 2016 to December 2021. The cases were divided into three groups based on the gestational age at the first diagnosis:<28 weeks (40 cases), 28-31 +6 weeks (65 cases), and ≥32 weeks (15 cases). They were also categorized into isolated and non-isolated FGR based on the presence of other ultrasound abnormalities (69 and 51 cases in each). Chromosomal karyotype analysis and CMA were conducted on all patients. The prenatal diagnosis results were analyzed, as well as the detection of chromosomal abnormalities in different gestational age groups and types of FGR. Statistical analysis was performed using Fisher's exact test. Results:(1) A total of 14 abnormalities were detected by CMA and four cases were detected by chromosomal karyotype analysis. The abnormal detection rate of CMA was higher than that of chromosomal karyotype analysis [11.7% (14/120) vs. 3.3% (4/120), P=0.025]. Among the total 14 cases of chromosomal abnormalities, there were seven pathogenic copy number variations (CNVs) and four variants of unknown significance (VUS), as well as two cases of trisomy-18 and one case of Turner syndrome. Among the 14 cases, eight had associated ultrasound abnormalities. Eleven of the 14 cases opted for induced abortion; three continued pregnancy to delivery, with two neonates showing no abnormalities and one exhibiting slightly delayed physical development. Both methods detected three cases of aneuploidy mnumber abnormalities (2.5%, 3/120) For chromosomal abnormalities <10 Mb, the detection rate of CMA was higher than that of chromosomal karyotype analysis [9.2% (11/120) vs. 0.8% (1/120), Fisher's exact, P=0.005]. Both methods detected one case of <10 Mb CNV, while CMA alone detected ten cases of <10 Mb microdeletions/microduplications (8.3%, 10/120), including six cases of pathogenic CNVs and four cases of VUS. (2) Among the 40 cases in the <28 weeks group, six cases (15.0%) of chromosomal abnormalities were detected, including three cases of aneuploidy, two cases of pathogenic CNVs, and one case of VUS. Among the 65 cases in the 28-31 +6 weeks group, seven cases (10.8%) of chromosomal abnormalities were detected, including five cases of pathogenic CNVs and two cases of VUS. Of the 15 cases in the ≥32 weeks group, one case of chromosomal abnormality was detected, which was VUS. (3) No statistically significant difference was found in the detection rate of chromosomal abnormalities between the isolated FGR and the non-isolated FGR groups [8.7%(6/69) vs. 15.7%(8/51), Fisher's exact, P=0.263]. (4) After excluding the ≥32 weeks non-isolated FGR group (only one case), the <28 weeks non-isolated FGR group had the highest detection rate of chromosomal abnormalities (1/18), while no abnormalities were detected in the ≥32 weeks isolated FGR group. Conclusions:Among FGR fetuses, the highest detection rates of chromosomal abnormalities are found in early-onset and non-isolated FGR. Prenatal diagnosis with CMA testing can significantly improve the detection rate of genetic causes in various types of FGR fetuses.

The aryl hydrocarbon receptor (AhR) plays a crucial role in regulating many physiological processes. Activating the AhR-CYP1A1 axis has emerged as a novel therapeutic strategy against various inflammatory diseases. Here, a practical in situ cell-based fluorometric assay was constructed to screen AhR-CYP1A1 axis modulators, via functional sensing of CYP1A1 activities in live cells. Firstly, a cell-permeable, isoform-specific enzyme-activable fluorogenic substrate for CYP1A1 was rationally constructed for in-situ visualizing the dynamic changes of CYP1A1 function in living systems, which was subsequently used for discovering the efficacious modulators of the AhR-CYP1A1 axis. Following screening of a compound library, LAC-7 was identified as an efficacious activator of the AhR-CYP1A1 axis, which dose-dependently up-regulated the expression levels of both CYP1A1 and AhR in multiple cell lines. LAC-7 also suppressed macrophage M1 polarization and reduced the levels of inflammatory factors in LPS-induced bone marrow-derived macrophages. Animal tests showed that LAC-7 could significantly mitigate DSS-induced ulcerative colitis and LPS-induced acute lung injury in mice, and markedly reduced the levels of multiple inflammatory factors. Collectively, an optimized fluorometric cell-based assay was devised for in situ functional imaging of CYP1A1 activities in living systems, which strongly facilitated the discovery of efficacious modulators of the AhR-CYP1A1 axis as novel anti-inflammatory agents.

Sensorineural hearing loss (SNHL), the most commonly-occurring form of hearing loss, is caused mainly by injury to or the loss of hair cells and spiral ganglion neurons in the cochlea. Numerous environmental and physiological factors have been shown to cause acquired SNHL, such as ototoxic drugs, noise exposure, aging, infections, and diseases. Several programmed cell death (PCD) pathways have been reported to be involved in SNHL, especially some novel PCD pathways that have only recently been reported, such as ferroptosis, necroptosis, and pyroptosis. Here we summarize these PCD pathways and their roles and mechanisms in SNHL, aiming to provide new insights and potential therapeutic strategies for SNHL by targeting these PCD pathways.
Humans ; Hearing Loss, Sensorineural/metabolism* ; Necroptosis/drug effects* ; Pyroptosis/drug effects* ; Ferroptosis/drug effects* ; Animals

[摘 要] 目的：探讨白蛋白结合型紫杉醇联合PD-1抑制剂用于治疗一线化疗失败的骨与软组织肉瘤的疗效及安全性。方法：回顾性分析北京积水潭医院骨肿瘤科2017年8月至2020年8月收治的一线化疗失败的晚期骨与软组织肉瘤患者。患者接受白蛋白结合型紫杉醇（125~140 mg/m2，第1天和第8天）与PD-1抑制剂（信迪利单抗或特瑞普利单抗，每21 d一次）联合治疗。每2个治疗周期评估1次疗效，按RECIST 1.1标准评估肿瘤疗效，按NCI-CTCAE5.0标准评估不良反应。结果：共20名患者纳入研究，完成1至8个治疗周期，中位治疗周期数为3个。所有患者均可评估疗效，完全缓解4例（20%），部分缓解0例，稳定9例（45%），疾病进展7例（35%）。客观缓解率（ORR）为20%，疾病控制率（DCR）为65%。中位无进展生存期（PFS）为3.0个月。治疗期间主要不良反应包括2级白细胞减少（40%）、1-2级神经毒性反应（20%），以及2级甲状腺功能减退（10%）。结论：白蛋白结合型紫杉醇联合PD-1抑制剂治疗为一线化疗失败的晚期骨与软组织肉瘤患者提供了一种潜在的治疗选择，其不良反应可控，值得开展更大样本的前瞻性研究进一步验证其疗效。

Objective:To construct a prior based on the inherent properties of PET to accurately segment the lesion areas.Methods:A network framework for PET tumor segmentation driven by geodesic priors was proposed (geodesic network for short). Specifically, partial differential equations were constructed to characterize the geodesic distances between different regions in PET images. Tumor marker points identified by CT labeling were used as the initial conditions for the equations. To enhance the contrast between areas of lung or breast tumors and normal tissues, a smooth Heaviside function was utilized to map the geodesic distances. The network framework adopted a dual-branch architecture, using geodesic priors to assist in PET image segmentation.Results:The proposed method achieved a Dice coefficient of 94.92% in lung cancer segmentation and 90.12% in breast cancer segmentation. With the addition of geodesic priors in the Unet, the Dice coefficient for breast cancer increased by 32.37% (from 42.50% to 74.87%).Conclusion:Geodesic priors can significantly improve segmentation outcomes and enhance the generalization capability of the network.

Objective:To construct a risk classification model based on logistic regression for medical equipment,so as to improve the application efficiency of medical equipment in the department of cardiovascular medicine. Methods:The logistic regression algorithm was used to construct the risk grade of adverse event of medical equipment of the department of cardiovascular medicine,and data collation and analysis were used to realize monitoring and management control for medical equipment. The 31 sets used medical equipment in the Department of Cardiovascular equipment of Xi'an No.3 Hospital from October 2021 to October 2022 were selected. Equipment management was conducted using conventional methods for risk management from October 2021 to October 2022,while equipment management from November 2022 to November 2023 adopted a risk classification model based on logistic regression was used to conduct risk management. A total of 204 logs of equipment application of the two kinds of management methods were selected,and each method selected 102 logs. The error rate of clinical operation,the occurrence of equipment failure,the timeliness score of risk management of equipment and the rate of hidden danger of safety risk of equipment were compared between the two kinds of management methods. Results:The number of error use of equipment,operational error and man-made mistake were respectively 3,2 and 2 in 102 logs that were managed by using risk classification model,and the incidence rates of them were respectively 2.94％,1.97％ and 1.97％,which were lower than those by using conventional management method,and the differences were statistically significant (x2=11.613,13.058,14.191,P<0.05). The average failure rate,self-maintenance rate of failure and average scrap rate of the medical equipment of the department of cardiovascular medicine of the management with risk classification model were respectively (0.56±0.22)％,(0.79±0.19)％ and (0.90±0.22)％,all of which were lower than those of conventional management method,and the differences were statistically significant (x2=16.971,15.531,15.809,P<0.05). The risk early warning,risk identification,and the average timeliness scores of risk prevention and control of using the management with risk classification model were respectively (90.29±8.69),(89.69±7.69),and (94.58±6.69),all of which were higher than those of using the management with conventional management method,and the differences were statistically significant (t=13.325,11.003,11.676,P<0.05). The number of mechanical injury,associated infection,and abnormal operation of equipment were respectively 1,1 and 2 in 31 medical equipment that were managed by risk classification model,and the incidence rates of them were respectively 3.23％,3.23％ and 6.45％,all of which were lower than those of conventional management methods,and the differences were statistical significances (x2=5.167,7.631,5.413,P<0.05),respectively. Conclusion:The application of a risk classification model based on logistic regression for medical equipment of the department of cardiovascular medicine can improve the utilization rate of equipment,and reduce the potential risk hidden danger of safety,and enhance the operation quality of equipment.

Objective:To construct a risk classification model based on logistic regression for medical equipment,so as to improve the application efficiency of medical equipment in the department of cardiovascular medicine. Methods:The logistic regression algorithm was used to construct the risk grade of adverse event of medical equipment of the department of cardiovascular medicine,and data collation and analysis were used to realize monitoring and management control for medical equipment. The 31 sets used medical equipment in the Department of Cardiovascular equipment of Xi'an No.3 Hospital from October 2021 to October 2022 were selected. Equipment management was conducted using conventional methods for risk management from October 2021 to October 2022,while equipment management from November 2022 to November 2023 adopted a risk classification model based on logistic regression was used to conduct risk management. A total of 204 logs of equipment application of the two kinds of management methods were selected,and each method selected 102 logs. The error rate of clinical operation,the occurrence of equipment failure,the timeliness score of risk management of equipment and the rate of hidden danger of safety risk of equipment were compared between the two kinds of management methods. Results:The number of error use of equipment,operational error and man-made mistake were respectively 3,2 and 2 in 102 logs that were managed by using risk classification model,and the incidence rates of them were respectively 2.94％,1.97％ and 1.97％,which were lower than those by using conventional management method,and the differences were statistically significant (x2=11.613,13.058,14.191,P<0.05). The average failure rate,self-maintenance rate of failure and average scrap rate of the medical equipment of the department of cardiovascular medicine of the management with risk classification model were respectively (0.56±0.22)％,(0.79±0.19)％ and (0.90±0.22)％,all of which were lower than those of conventional management method,and the differences were statistically significant (x2=16.971,15.531,15.809,P<0.05). The risk early warning,risk identification,and the average timeliness scores of risk prevention and control of using the management with risk classification model were respectively (90.29±8.69),(89.69±7.69),and (94.58±6.69),all of which were higher than those of using the management with conventional management method,and the differences were statistically significant (t=13.325,11.003,11.676,P<0.05). The number of mechanical injury,associated infection,and abnormal operation of equipment were respectively 1,1 and 2 in 31 medical equipment that were managed by risk classification model,and the incidence rates of them were respectively 3.23％,3.23％ and 6.45％,all of which were lower than those of conventional management methods,and the differences were statistical significances (x2=5.167,7.631,5.413,P<0.05),respectively. Conclusion:The application of a risk classification model based on logistic regression for medical equipment of the department of cardiovascular medicine can improve the utilization rate of equipment,and reduce the potential risk hidden danger of safety,and enhance the operation quality of equipment.

Objective:To explore the value of chromosomal microarray analysis (CMA) in the genetic diagnosis of different types of fetal growth restriction (FGR).Methods:A retrospective analysis was conducted on 120 cases who were diagnosed with FGR by ultrasound and underwent prenatal diagnosis at the Department of Obstetrics & Gynecology and Reproductive Medicine, Peking University First Hospital, from January 2016 to December 2021. The cases were divided into three groups based on the gestational age at the first diagnosis:<28 weeks (40 cases), 28-31 +6 weeks (65 cases), and ≥32 weeks (15 cases). They were also categorized into isolated and non-isolated FGR based on the presence of other ultrasound abnormalities (69 and 51 cases in each). Chromosomal karyotype analysis and CMA were conducted on all patients. The prenatal diagnosis results were analyzed, as well as the detection of chromosomal abnormalities in different gestational age groups and types of FGR. Statistical analysis was performed using Fisher's exact test. Results:(1) A total of 14 abnormalities were detected by CMA and four cases were detected by chromosomal karyotype analysis. The abnormal detection rate of CMA was higher than that of chromosomal karyotype analysis [11.7% (14/120) vs. 3.3% (4/120), P=0.025]. Among the total 14 cases of chromosomal abnormalities, there were seven pathogenic copy number variations (CNVs) and four variants of unknown significance (VUS), as well as two cases of trisomy-18 and one case of Turner syndrome. Among the 14 cases, eight had associated ultrasound abnormalities. Eleven of the 14 cases opted for induced abortion; three continued pregnancy to delivery, with two neonates showing no abnormalities and one exhibiting slightly delayed physical development. Both methods detected three cases of aneuploidy mnumber abnormalities (2.5%, 3/120) For chromosomal abnormalities <10 Mb, the detection rate of CMA was higher than that of chromosomal karyotype analysis [9.2% (11/120) vs. 0.8% (1/120), Fisher's exact, P=0.005]. Both methods detected one case of <10 Mb CNV, while CMA alone detected ten cases of <10 Mb microdeletions/microduplications (8.3%, 10/120), including six cases of pathogenic CNVs and four cases of VUS. (2) Among the 40 cases in the <28 weeks group, six cases (15.0%) of chromosomal abnormalities were detected, including three cases of aneuploidy, two cases of pathogenic CNVs, and one case of VUS. Among the 65 cases in the 28-31 +6 weeks group, seven cases (10.8%) of chromosomal abnormalities were detected, including five cases of pathogenic CNVs and two cases of VUS. Of the 15 cases in the ≥32 weeks group, one case of chromosomal abnormality was detected, which was VUS. (3) No statistically significant difference was found in the detection rate of chromosomal abnormalities between the isolated FGR and the non-isolated FGR groups [8.7%(6/69) vs. 15.7%(8/51), Fisher's exact, P=0.263]. (4) After excluding the ≥32 weeks non-isolated FGR group (only one case), the <28 weeks non-isolated FGR group had the highest detection rate of chromosomal abnormalities (1/18), while no abnormalities were detected in the ≥32 weeks isolated FGR group. Conclusions:Among FGR fetuses, the highest detection rates of chromosomal abnormalities are found in early-onset and non-isolated FGR. Prenatal diagnosis with CMA testing can significantly improve the detection rate of genetic causes in various types of FGR fetuses.

Objective To investigate the value of histone deacetylase 3(HDAC3)in evaluating the risk of slow/no reflow in AMI patients after PCI.Methods A total of 280 AMI patients undergo-ing PCI in our hospital from June 2020 to June 2022 were recruited,and according to TIMI blood flow grading,they were divided into slow/no reflow group(TIMI≤grade 11,n=54)and normal flow group(TIMI＞grade Ⅱ,n=226).The demographic characteristics,underlying diseases,baseline data at admission,and preoperative results of coronary angiography and laboratory tests were compared between the two groups.Multivariate logistic regression analysis was used to determine the influencing factors for slow/no reflow in AMI patients after PCI,and ROC curve was drawn to analyze the predictive value of related indicators for slow/no reflow.Results Obvi-ously larger proportions of smoking history and Killip grade Ⅱ,higher heart rate,longer reperfu-sion time,and higher serum levels of LDL-C,NLR,D-D and HDAC3 were observed in the slow/no reflow group than the normal flow group(P＜0.05,P＜0.01).Multivariate logistic regression analysis showed that reperfusion time,NLR and HDAC3 were influencing factors for slow/no reflow in AMI patients after PCI(P＜0.05,P＜0.01).The AUC value of reperfusion time+NLR in predicting the slow/no reflow after PCI in AMI patients was 0.798(95％CI:0.664-0.922,P=0.002),with a sensitivity and specificity of 0.87 and 0.73,respectively.And when serum HDAC3 level was combined in the prediction,the AUC value was 0.903(95％CI:0.790-0.922,P＜0.01),with a sensitivity and specificity of 0.93 and 0.84,respectively.Conclusion Preoperative HDAC3 level is an influencing factor for slow/no reflow in AMI patients after PCI.Based on reperfusion time and NLR,combination of the 3 indicators can provide additional predictive value for slow/no reflow in these patients.