舌诊是中医望诊的重要手段,同时,温度与人体的健康息息相关。为了研究舌面的脏腑功能定位及舌象温度关系的反映,论文提出了一种红外技术的感兴趣区域(region of interest, ROI)模型研究方法。首先,利用葛立恒扫描法和Bezier曲线对多边形ROI模型进行改进;然后,借助U-Net分割网络将提取出的温度信息进行训练与学习,从而做到批量处理舌体温度信息;最后,利用HSV色彩模型进行3D可视化,达成舌象温度分区的可视化研究。此外,为了验证该方法的准确性,实验还对模型截取出的舌体进行了评价指标验证,准确度可以达到0.991 1,分割效果极佳。研究表明:改进后的红外信息提取技术既能直观地观察到舌体的分区状况,也可以完整保留舌体的信息变化,为中医的数据化提供了完整可行性方案。实现了舌体红外信息数据的提取与中医诊断技术的有机结合。解决了中医一体化望诊的舌体信息完整性及准确性问题。
Introduction: Infant and child morbidity and mortality linked to infections remain a major concern in both developed and developing countries. The general objective was to determine the characteristics of hospital morbidity and mortality linked to infectious pathologies in the pediatric department of the King Baudouin hospital center in Dakar. Methods: This was a retrospective, descriptive and analytical study focusing on children hospitalized from January 1, 2018 to December 31, 2021 in the pediatric department of the King Baudouin hospital center in Dakar. Results: During the study period, 1474 children were hospitalized for an infectious pathology in the pediatric department. The sex ratio was 1.48. Children aged between one and twelve months represented 34.12%. Morbidity due to respiratory and digestive infections was 42.33% and 33.45%, respectively. Pediatric hospital mortality was 0.75% with a predominance noted in infants between 1 and 12 months (55.86%). Respiratory, digestive infections, sepsis and neurological infections were the main causes of death. Factors associated with mortality were age (1 to 12 months) with p Conclusion: Reducing infant and child mortality linked to infectious diseases remains a crucial challenge to address for improving children’s health. Early detection of related signs, prevention of infections, and adequate care and monitoring of infants and children can contribute to this reduction.
目的:颈前路减压融合术是治疗退行性颈椎病的经典手术方式,钉板的使用增加了融合率及稳定性的同时,间接导致了邻近椎体退变和术后吞咽困难的发生。文章通过Meta分析方法比较ROI-C^(TM)自锁系统和传统融合器联合钉板内固定治疗退行性颈椎病患者的临床结果和并发症情况,为颈前路减压融合术中内固定方式的选择提供循证学支持。方法:检索中国知网、万方、维普、PubMed、Cochrane Library、Web of Science和Embase数据库,检索关于颈前路减压融合术中应用ROI-C^(TM)自锁系统与融合器联合钉板内固定治疗退行性颈椎病的中英文文献。检索时间范围为各数据库建库至2023年7月。由2名研究者严格按照纳入与排除标准选择文献,采用Cochrane偏倚风险工具对随机对照试验进行质量评价,NOS量表对队列研究进行质量评价。采用RevMan 5.4软件进行Meta分析。结局指标包括手术时间、术中出血量、日本骨科协会(Japanese Orthopaedic Association Scores,JOA)评分、颈椎功能障碍指数、C_(2)-C_(7)Cobb角、融合率、邻近椎体退变发生率、融合器沉降率和吞咽困难发生率。结果:共纳入13项研究,其中回顾性队列研究11项,随机对照试验2项,共1136例患者,ROI-C组569例,融合器联合钉板组567例。Meta分析结果显示:ROI-C组与融合器联合钉板组在手术时间(MD=-15.52,95%CI:-18.62至-12.42,P<0.00001),术中出血量(MD=-24.53,95%CI:-32.46至-16.61,P<0.00001),术后邻近节段退变率(RR=0.40,95%CI:0.27-0.60,P<0.00001)和术后总吞咽困难发生率(RR=0.18,95%CI:0.13-0.26,P<0.00001)均具有显著性差异。两者在术后JOA评分、颈椎功能障碍指数、C_(2)-C_(7)Cobb角、融合率和融合器沉降率方面无显著性差异(P≥0.05)。结论:在颈椎前路减压融合术中应用ROI-C^(TM)自锁系统与传统融合器联合钉板内固定治疗退行性颈椎病均可达到满意的临床效果,ROI-C^(TM)自锁系统操作更加�
嵌入式设备中部署深度学习检测模型往往面临算力不足的问题,而感兴趣区域(ROI)提取可作为一种高效的性能优化手段。文章提出一种基于HSV(Hue,Saturation,Value)色彩空间模型的ROI提取的方法,将检测目标的像素信息转化到HSV色彩空间,在色相-饱和度(H-S)平面引入DBSCAN(Density-Based Spatial Clustering of Applications with Noise)聚类算法,精确定位目标的主色彩像素在H-S平面上的分布位置,同时过滤杂乱色彩,然后通过Quickhull(快壳)凸包算法,从散点数据中拟合出主色彩的精确分布范围。根据获取的主色彩范围对像素进行遍历,可以根据色彩信息有效地提取ROI。实验结果表明,经过该方法优化后的Faster R-CNN(Faster Regions with Convolutional Neural Networks)算法,较原模型减少了57.08%的平均推理耗时,同时精确率提升了0.9百分点。这对于嵌入式设备中进行实时目标检测具有重要的现实意义。
