Conceptual-driven classification for coding advise in health insurance reimbursement

Objective

With the non-stop increases in medical treatment fees, the economic survival of a hospital in Taiwan relies on the reimbursements received from the Bureau of National Health Insurance, which in turn depend on the accuracy and completeness of the content of the discharge summaries as well as the correctness of their International Classification of Diseases (ICD) codes. The purpose of this research is to enforce the entire disease classification framework by supporting disease classification specialists in the coding process.

Methodology

This study developed an ICD code advisory system (ICD-AS) that performed knowledge discovery from discharge summaries and suggested ICD codes. Natural language processing and information retrieval techniques based on Zipf's Law were applied to process the content of discharge summaries, and fuzzy formal concept analysis was used to analyze and represent the relationships between the medical terms identified by MeSH. In addition, a certainty factor used as reference during the coding process was calculated to account for uncertainty and strengthen the credibility of the outcome.

Results

Two sets of 360 and 2579 textual discharge summaries of patients suffering from cerebrovascular disease was processed to build up ICD-AS and to evaluate the prediction performance. A number of experiments were conducted to investigate the impact of system parameters on accuracy and compare the proposed model to traditional classification techniques including linear-kernel support vector machines. The comparison results showed that the proposed system achieves the better overall performance in terms of several measures. In addition, some useful implication rules were obtained, which improve comprehension of the field of cerebrovascular disease and give insights to the relationships between relevant medical terms.

Conclusion

Our system contributes valuable guidance to disease classification specialists in the process of coding discharge summaries, which consequently brings benefits in aspects of patient, hospital, and healthcare system.     

基于生物学优化肿瘤放射治疗计划的评估函数研究

目的：从生物学优化肿瘤放射治疗计划的角度来提出可用于评估放射治疗方案可行性的评估函数。方法：在以细胞受到照射后的再修复、再群体化、细胞周期的再分布、肿瘤内乏氧细胞的再氧合,即＂4R理论＂[1]的基础上,分析生物效应剂量函数BED的变种公式,最后得出3个用于评估肿瘤放疗计划好坏的评估函数。再利用评估函数试验性的分析文献报道过的肿瘤放射治疗方案,并与对应的方案结果进行比对,理论上评价评估函数的可行性。结果：综合分析已有治疗方案参数与评估函数理论参数基本一致。结论：基于生物学优化肿瘤放射治疗计划的评估函数在临床上用来预判治疗方案是可行的。     

乳腺专科带教计划在临床实习中的应用

目的：探讨乳腺外科护生临床实习中存在的问题及对策。方法针对乳腺外科护生临床实习中存在的问题进行研究,并进行讨论研究。结果乳腺外科临床护理教学中的问题主要有操作技术的问题、不了解患者的需求、理论与实际联系不足等。制定专科带教计划能够帮助护生提高理论联系之际的能力、学习热情及应对问题的能力,提高临床护理教学质量。结论提高乳腺外科临床护理教学的质量,必须及时发现存在的问题并寻找解决对策,制定乳腺专科带教计划。     

自制口袋书在提高护士对专科应急预案知晓率的应用和效果评价

目的：提高护士对应急预案知晓率,确保护理安全,,加强护理人员对应急情况的处理。方法由妇产科病区护士组成品管圈小组,并根据第八版《妇产科学》及妇产科专科应急预案设计调查问卷,对全科护士进行考核,再通过应急预案口袋书的学习,对活动前后进行对比。结果活动后,护士对专科应急预案知晓率由活动前的62.13%提高到活动后的96.16%,专科应急预案相关技能考核由活动前的84.58%提高到活动后的93.05%。结论活动对护士掌握专科应急预案的相关知识及相关技能考核有明显提高,从而确保护理安全,降低医疗风险,减少护患纠纷,提高满意度[1]。     

肘关节三联征治疗方案的深入讨论

文题释义： 肘关节三联征：桡骨头骨折、冠突骨折伴肘关节后或后外侧脱位于1996年被Hotchkiss命名为“terrible triad of the elbow”,国内习惯称之为肘关节恐怖三联征,常合并外侧韧带复合体、内侧韧带复合体起止部位的撕脱损伤及冠突骨折周围前关节囊破裂。由于近年来在恐怖三联征损伤的诊断和治疗方面取得了迅速进展,预后也相应地比过去好很多,国内很多专家建议直接将其命名为肘关节三联征。 肘关节后外侧旋转不稳定：指由于外侧尺侧副韧带(外侧尺骨副韧带)损伤导致后侧半脱位或肱桡关节脱位,但不破坏近端尺桡关节,在1991由O’Driscoll等首次描述。目前对其的经典解释为：因摔倒致肘关节在轴向载荷下开始屈曲时,在身体在接近地面时肘关节受到外翻、旋后和轴向的应力综合作用,引起肘关节后脱位;同时导致的软组织损伤用“Horii环”解释：软组织破坏开始于外侧,并随着半脱位程度的增加向前后向内侧推进。 背景：近年来肘关节三联征损伤的诊断和治疗方面取得了迅速进展。 目的：简单介绍肘关节三联征的定义、相关基本解剖及稳定结构、损伤机制、分型,深入探讨其治疗方案。 方法：第一作者检索PubMed、CNKI及万方数据库建库至2019年9月的文献,以“terrible triad of the elbow,elbow joint,Fracture dislocation”为英文检索词,以“肘关节恐怖三联征,肘关节骨折脱位”为中文检索词,查阅大量近年肘关节三联征的相关文献,全面分析并总结其治疗策略。 结果与结论：对于肘关节三联征损伤患者,术前必须仔细检查,认真分析影像学检查结果及打印3D模型,准确判断肘部骨性结构和韧带损伤程度;术中通常先通过单独外侧入路或内外联合入路重建桡骨头的稳定性、修复冠突骨折及外侧韧带复合体损伤后,然后行肘关节稳定性检测,以决定是否需要进一步修复内侧韧带复合体或应用铰链式外固定器来恢复肘关节的稳定性;术后积极行早期康复功能锻。通过上述系统治疗可相对减少术后并发症,加速骨折愈合,改善肘关节功能和患者生活质量。 ORCID: 0000-0002-8244-4769(李宁鑫) 中国组织工程研究杂志出版内容重点：人工关节;骨植入物;脊柱;骨折;内固定;数字化骨科;组织工程     

医学影像信息系统辅助全膝置换前计划中的作用与意义

BACKGROUND: In clinical application, Picture Archiving and Communication System gradually replaces the traditional preoperative planning mode of acetate template measurement on film, which can accurately obtain the parameters needed in the operation of total knee arthroplasty.     

最小子野宽度对肺癌容积调强弧形治疗计划质量的影响

目的:探讨肺癌容积调强弧形治疗（VMAT）计划中,最小子野宽度对放疗计划质量的影响。方法:选择病理证实为肺癌的患者12例,计划靶区处方剂量为60 Gy/30次。使用Monaco治疗计划系统设计全弧VMAT计划,计算网格为0.3 cm,采用6 MV X射线,每个弧的最大控制点数为200。计算不确定度为1%。最小子野宽度依次选取0.5、0.7、1.0、1.2和1.5 cm。以最小子野宽度为1.0 cm的计划为参考,与其他子野宽度下的计划相比较,分析各剂量学指标随最小子野宽度变化的情况。结果:随着最小子野宽度增大为1.2、1.5 cm时,靶区Dmean、D2变化明显（P?0.05）。相比参考最小子野宽度1.0 cm,最小子野宽度为1.5 cm时,适形度指数变化较大（Z=-2.694, P?0.05）;最小子野宽度为0.5、0.7、1.2和1.5 cm时,靶区覆盖率相差较明显（Z=-2.904、-3.061、-2.747、-2.941, P?0.05）。最小子野宽度1.0 cm时,VMAT计划平均机器跳数为684.4 MU;最小子野宽度为0.5、0.7、1.2、1.5 cm时,计划的平均机器跳数分别为为972.8、763.8、660.1、665.3 MU。控制点数随着最小子野宽度的增大呈降低趋势。结论:利用Monaco系统设计肺癌VMAT计划,最小子野宽度为1.0 cm时VMAT计划的质量更好。通过调整VMAT计划的最小子野宽度可大大降低机器跳数,提高射束的有效使用率,同时提高患者的治疗舒适度。     

医院制剂生产管理系统的设计与实现

目的设计开发完整的医院制剂生产管理系统，使其满足物资的入出存管理和生产管理的需求。方法在细致分析医院制剂生产的管理流程的基础上，结合企业资源规划系统，采用客户端，服务器模式，应用SQL Server数据库管理系统和PowerBuilder开发工具设计、开发了该系统，主要由库存管理模块和生产管理模块组成，同时具有综合查询、报表输出和系统维护等功能。结果该系统能够与HIS连接，根据HIS销售情况制定生产计划，通过生产计划和现有库存快速制定采购计划．理顺了医院制剂室的工作流程，能够准确、快捷地输出相应的报表。结论该系统设计方案可行，系统操作简单．提高了工作效率，加强了医院的管理工作。     

蒙特卡罗N粒子运输法计算与实际测量的60 Co治疗机剂量分布比较

目的 探讨蒙特卡罗N粒子运输法(MCNP)模拟计算的可行性.方法 用百分深度剂量(PDD)分布及标准峰值散射因子(NPSP),比较水模体计算值和实际测量及报告值之间的差异.结果 在10 cm×10 cm射野时,测量值和计算值之间差异无统计学意义(t=-0.41,P>0.05),而在5 cm×5 cm及12 cm×12 cm时,测量值与计算值之间差异有统计学意义(t=7.2、-4.6,P<0.05).计算值和报告值之间符合良好,差异无统计学意义(t=-1.906,P>0.05).同一射野最大剂量点下百分深度剂量随深度增大而减少,同一深度处百分深度剂量随射野增大而增大;同一深度处射野中心轴上的剂量最高,向射野边缘剂量逐渐减少.结论 利用蒙特卡罗MCNP可以建立一组准确和全面的百分深度剂量及标准峰值散射因子参数,为放疗质量保证和质量控制提供依据.

Abstract：

Objective To discuss the feasibility of Monte Carlo N-particle transport code(MCNP)simulated calculation.Methods The calculation in water phantom was contrasted with the practical measurements and the reported values using the percent depth dose(PDD)curve and normal peak scatter factor.Results There Was no significant difference between calculated and measured results in the 10 cm×10 cm field(t＝-0.41,P>0.05),however,there were significant differences in the 5 cm×5 cm field(t=7.2,P<0.05)and in the 12 cm×12 cm field(t=-4.6,P<0.05).There was no significant difierence between the calculated results and the reported values(t=-1.91,P>0.05).In the same radiation field,the PDD decreased as the depth increased,but increased as the size of the radiation field increased at the same depth.PDD and normal peak scatter factor were both important parameters for calculation of prescribed dose.Conclusions It is possible to establish a set of accurate and comprehensive percent depth doses and normal peak scatter factor parameters so as to provide the basis of clinical use, quality assurance and quality control for radiotherapy.     

Noise Correlations for Faster and More Robust Learning

Distributed population codes are ubiquitous in the brain and pose a challenge to downstream neurons that must learn an appropriate readout. Here we explore the possibility that this learning problem is simplified through inductive biases implemented by stimulus-independent noise correlations that constrain learning to task-relevant dimensions. We test this idea in a set of neural networks that learn to perform a perceptual discrimination task. Correlations among similarly tuned units were manipulated independently of an overall population signal-to-noise ratio to test how the format of stored information affects learning. Higher noise correlations among similarly tuned units led to faster and more robust learning, favoring homogenous weights assigned to neurons within a functionally similar pool, and could emerge through Hebbian learning. When multiple discriminations were learned simultaneously, noise correlations across relevant feature dimensions sped learning, whereas those across irrelevant feature dimensions slowed it. Our results complement the existing theory on noise correlations by demonstrating that when such correlations are produced without significant degradation of the signal-to-noise ratio, they can improve the speed of readout learning by constraining it to appropriate dimensions.SIGNIFICANCE STATEMENT Positive noise correlations between similarly tuned neurons theoretically reduce the representational capacity of the brain, yet they are commonly observed, emerge dynamically in complex tasks, and persist even in well-trained animals. Here we show that such correlations, when embedded in a neural population with a fixed signal-to-noise ratio, can improve the speed and robustness with which an appropriate readout is learned. In a simple discrimination task such correlations can emerge naturally through Hebbian learning. In more complex tasks that require multiple discriminations, correlations between neurons that similarly encode the task-relevant feature improve learning by constraining it to the appropriate task dimension.