Deep Phenotyping and Prediction of Long-term Cardiovascular Disease: Optimized by Machine Learning

BackgroundPrediction of cardiovascular disease (CVD) is important in clinical practice. Machine learning (ML) may offer an improved alternative to current CVD risk stratification in individual patients. We aim to identify important predictors and compare ML models with traditional models according to their prediction performance in a large long-term follow-up cohort.MethodsThe Atherosclerosis Risk in Communities (ARIC) study was designed to study the progression of subclinical disease to cardiovascular events over a 25-year follow-up period. All phenotypic variables at visit 1 were obtained. All-cause death, CVD, and coronary heart disease were the outcomes for analysis. The ML framework involved variable selection using the random survival forest (RSF) method, model building, and 5-fold cross-validation. Model performance was evaluated by discrimination using the Harrell concordance index (C-index), accuracy using the Brier score (BS), and interpretability using the number of variables in the model.ResultsOf the 14,842 participants in ARIC, the average age was 54.2 years, with 45.2% male and 26.2% Black participants. Thirty-eight unique variables were selected in the RSF top 20 importance ranking of all 6 outcomes. Aging, hypertension, glucose metabolism, renal function, coagulation, adiposity, and sodium retention dominated the predictions of all outcomes. The ML models outperformed the regression models and established risk scores with a higher C-index, lower BS, and varied interpretability.ConclusionsThe ML framework is useful for identifying important predictors of CVD and for developing models with robust performance compared with existing risk models.     

基于FAERS的地舒单抗安全警戒信号挖掘与评价

基于FAERS数据库挖掘安全警戒信号,分析评估地舒单抗潜在不良反应信号,为其临床使用提供一定参考依据。方法 通过Openvigil 2.1访问 FAERS 数据库,将地舒单抗作为主要药物,检索自该药首次上市时间（2010年5月—2021年9月）的数据,获得与地舒单抗相关的不良事件报告记录。使用报告比值比法（ROR）和贝叶斯置信度递进神经网络法（BCPNN）筛选地舒单抗安全警戒信号,挖掘潜在的不良反应,并通过工具BioPortal对不良事件信号挖掘结果进行系统分类,通过判断信号间置信区间的变化,发现与药物不良事件关联性较大的信号。结果 从FAERS数据库中收集到270503份不良反应事件（ADE）报告,根据ROR法和BCPNN法共得到343个不良事件信号,通过信号间同义合并、剔除与药物无关的信号后,得到316个不良事件信号。地舒单抗的不良事件系统分类主要为肌肉骨骼和结缔组织疾病、医学检查、胃肠道疾病。FAERS数据库的信号挖掘结果发现,高风险且说明书中未收录的安全警戒信号包括颞下颌关节综合征、下颌脓肿、雌激素缺乏症、血液甲状旁腺激素增加,计算高风险信号的置信区间显示颞下关节综合征较有可能发展成为新的不良反应;另外,也发现种植体周围炎为具有临床意义的可疑警戒信号,但有待进一步观察研究。结论 基于FAERS数据库的信号挖掘结果提示临床应规范使用地舒单抗,治疗期间需警惕患者是否出现颞下颌关节综合征、下颌脓肿、雌激素缺乏症、血液甲状旁腺激素增加等不良反应事件,以便尽早发现尽早处理,从而有效降低临床用药风险     

Psychological screening for the children with habitual snoring

ObjectivesAdenotonsillar hypertrophy is the major determinants of habitual snoring in pediatric population. Behavioral hyperactivity and schooling problems have been repeatedly reported in these children, and it may underlie more extensive behavioral disturbances, particularly for the obese children. The aim of the present study is to evaluate the incidence and characteristics of emotional and behavioral problems using outpatient-based psychological screening tools in the children with habitual snoring.MethodsTotal 235 patients and 170 controls, who aged 4–9 years were enrolled. Body mass index (BMI) z-score was obtained for age and gender and parental sleep-related breathing disorder (SRBD) questionnaire was used to assess severity of sleep-disordered breathing (SBD). Psychological assessment was performed using standardized questionnaires including Strength and Difficulties Questionnaire (SDQ), Children's Depression Inventory (CDI) and Screen for Child Anxiety Related Emotional Disorders (SCARED).ResultsChildren presenting habitual snoring had significantly higher mean scores on almost all scales of SDQ, and SCARED than community controls. Around 20% of the children with habitual snoring, compared with 10–11% of controls had significant levels of distress that could adversely impact treatment outcomes. There was no interaction between obstructive sleep apnea severity and behavioral ratings. The scores for emotional distress and hyperactivity were more prominent in the obese children. Significant psychological distress or impairment in social interactions was observed in children with higher SRBD scores.ConclusionsOur findings suggest that the presence of habitual snoring in young children is associated wide spectrum of behavioral problems and the level of psychological distress might be evaluated at the time of the diagnosis.     

摆位误差对肝癌立体定向放射治疗剂量学的影响

目的:在锥形束CT（CBCT）图像引导下,测量立体定向放射治疗（SBRT）中肝癌的摆位误差,并讨论摆位误差对靶区PTV和危及器官（OAR）剂量的影响。方法:回顾性分析接受SBRT的肝癌患者13例,每日放射治疗前行CBCT扫描,与计划CT图像进行灰度配准,根据肿瘤靶区及OAR位置获取患者移床参数,在计划系统中计算剂量分布,并分析移床参数对靶区PTV、OAR剂量和均匀性指数（HI）、适形度指数（CI）等一系列剂量学参数的影响。结果:校正后X、Y、Z方向上的摆位误差分别为（0.47±2.00）、（1.54±4.16）、（0.10±2.77） mm,Y方向上的摆位误差较大。相对于最小位移,最大位移对靶区、HI和CI影响大;对于OAR,与原始计划相比,左肾、小肠最大位移的剂量分布在Dmean、Dmax上具有统计学意义,右肾最小位移的剂量分布在Dmean上具有统计学意义。结论:≤3 mm的摆位误差对靶区剂量的影响比>5 mm的影响小,但对于最小剂量和覆盖率仍然影响显著。应该尽可能减小摆位误差,以实现精确放疗。     

Evaluation of the SMALL nail: Drive technology and behavior in situ

Although clear advances have been made during the last 5 years, practical difficulties persist for patients and surgeons in procedures for intramedullary lengthening of long bones. In particular, precise adjustment of the desired amount of lengthening and technically reliable checking of the length actually achieved are problematic. An intramedullary nail with a new type of drive that exploits the shape memory effect has been constructed. The drive technology and the behavior of the intramedullary nail in situ were evaluated in a cadaver experiment. Three shape memory alloy limb lengthening (SMALL) nails were implanted in a body donor. The SMALL nail contains a spring coupled to a shape memory element consisting of a nickel–titanium alloy. This shape memory element “remembers” its initial state before the lengthening through the spring and can return to it when it is warmed. A cartridge heater inside the lengthening nail is warmed using transcutaneous induction with high-frequency energy via a subcutaneously implanted coil. For evaluation, two SMALL nails were implanted into the femora (antegrade on the left and retrograde on the right) and one SMALL nail was implanted into the left tibia. Lengthening by 50 mm was attempted using repeated activation of the drive mechanism. At the same time, test parameters for temperature increases and cooling periods were continually monitored and the data were subsequently analyzed. The nail's mechanism worked in principle, but was inadequate in view of success rates (number of lengthening steps attempted versus number of lengthening steps achieved) of 21% for the SMALL nail in the tibia and left femur and 14% for the nail in the right femur. The temperature values measured during the distraction experiments show that high-frequency energy induction in the SMALL nail gives no cause for concern for patients.     

不同疾病状态下新生儿脑组织氧合变化的对照研究

目的采用近红外光谱测定技术（NIRS）检测新生儿脑组织氧饱和度（rSO2）,探讨不同疾病状态下新生儿脑rSO2的变化规律,为临床应用提供依据。方法2007年4月至2008年10月以无特殊疾病的223名足月儿作为正常组足月儿亚组,于生后3d内测定脑rSO2;以196例患有可能影响脑氧合疾病的新生儿作为疾病组,在疾病急性期测定脑rSO2。疾病组分为呼吸系统疾病亚组（97例）,分析脑rSO2与PaO2的关系;循环系统疾病亚组（44例）,分析脑rSO2与心率的关系;脑损伤亚组（55例）,分析脑rSO2与脑血流的关系。结果①疾病组脑rSO2为（56±6）%,显著低于正常组足月儿亚组（P〈0.05）。②轻度与重度呼吸系统疾病亚组脑rSO2分别为（60±3）%和（54±6）%,轻度和重度循环系统疾病亚组脑rSO2分别为（59±3）%和（53±6）%,轻度和重度脑损伤亚组脑rSO2分别为（59±3）%和（54±4）%。3个疾病亚组中轻度与重度间脑rSO2差异均有统计学意义（P〈0.01）。③呼吸系统疾病亚组脑rSO2与PaO2呈三次方程曲线（y=-62.93＋4.75x-0.059x2＋0.00024x3）。PaO2≥60mmHg时,脑rSO2约为62%,脑氧合正常;PaO2〈50mmHg时,脑rSO2〈57%,脑组织缺氧。循环系统疾病亚组脑rSO2与心率呈二次方程曲线（y=1.11＋0.8241x-0.0027x2）。心率在105～200.min-1时,脑rSO2〉58%,脑氧合正常;心率低于105.min-1或高于200.min-1时,脑rSO2〈58%,脑组织缺氧。脑损伤亚组脑rSO2〈58%时,大脑前动脉血流平均速度代偿性增高,阻力指数偏低,脑损伤较重。结论严重疾病状态下可同时伴有脑组织缺氧。脑rSO2的变化与PaO2、心率及脑血流的变化密切相关。NIRS技术为临床提供了一种可靠的、有价值的脑氧合检测方法,有助于临床直观量化地发现脑组织的缺氧。