Background Many clinical concepts are standardized under a categorical and hierarchical taxonomy such as ICD-10, ATC, etc. These taxonomic clinical concepts provide insight into semantic meaning and similarity among clinical concepts and have been applied to patient similarity measures. However, the effects of diverse set sizes of taxonomic clinical concepts contributing to similarity at the patient level have not been well studied. Methods In this paper the most widely used taxonomic clinical concepts system, ICD-10, was studied as a representative taxonomy. The distance between ICD-10-coded diagnosis sets is an integrated estimation of the information content of each concept, the similarity between each pairwise concepts and the similarity between the sets of concepts. We proposed a novel method at the set-level similarity to calculate the distance between sets of hierarchical taxonomic clinical concepts to measure patient similarity. A real-world clinical dataset with ICD-10 coded diagnoses and hospital length of stay (HLOS) information was used to evaluate the performance of various algorithms and their combinations in predicting whether a patient need long-term hospitalization or not. Four subpopulation prototypes that were defined based on age and HLOS with different diagnoses set sizes were used as the target for similarity analysis. The F-score was used to evaluate the performance of different algorithms by controlling other factors. We also evaluated the effect of prototype set size on prediction precision. Results The results identified the strengths and weaknesses of different algorithms to compute information content, code-level similarity and set-level similarity under different contexts, such as set size and concept set background. The minimum weighted bipartite matching approach, which has not been fully recognized previously showed unique advantages in measuring the concepts-based patient similarity. Conclusions This study provides a systematic benchmark evaluation of previous algorithms and novel algorithms used in taxonomic concepts-based patient similarity, and it provides the basis for selecting appropriate methods under different clinical scenarios. Electronic supplementary material The online version of this article (10.1186/s12911-019-0807-y) contains supplementary material, which is available to authorized users.

Background: To investigate the optimal monoenergetic level of spectral reconstructions in coronary computed tomography angiography (coronary CTA) on a dual-layer spectral detector computed tomography (SDCT) with half-dose contrast media.Methods: Two hundred patients with suspected coronary artery disease (CAD) were enrolled in this prospective coronary CTA study and randomly divided into a routine-dose contrast media group and a half-dose contrast media group (each n=100). Coronary CTA was performed using SDCT with prospective electrocardiogram (ECG)-gated mode. A tube voltage of 120 kVp was used, along with an automated tube current modulation. A dose of iodixanol 270 mgI/mL of 0.8 and 0.4 mL/kg was administered to the routine and half-dose groups, respectively. For the routine-dose group, 120 kVp polychromatic images with a model-based iterative reconstruction (IMR) (Group A) were reconstructed. For the half-dose group, three monoenergetic levels of images were reconstructed (Group B, 45 keV; Group C, 50 keV; and Group D, 55 keV).Objective indicators [mean CT values; noise; signal-to-noise ratio (SNR); and contrast-to-noise ratio (CNR)] and subjective indicators (contrast, sharpness, subjective noise, and acceptability) in each group were compared.Results: There were no significant differences in demographics or radiation dose (1.83±0.51 vs. 1.80± 0.53 mSv, P=0.78) between the routine-and half-dose groups. The average iodine loads were 15.33±2.26 and 7.48±1.14 g, respectively. Mean CT values, SNR, CNR, and subjective contrast in Group C were higher than those in Group A (P<0.05), and there were no significant differences in other indicators between Group C and Group A (P>0.05). The objective and subjective noise in Group B were worse than those in Group A (P<0.05). The contrast, sharpness, and acceptability of Group D were all worse than those of Group A (P<0.05).Conclusions: Compared to routine polychromatic images, 50 keV monoenergetic images can provide equivalent or improved coronary image quality in coronary CTA performed on SDCT with half the amount of contrast media.

No abstract

Epidemiological knowledge of pediatric diseases may improve professionals’ understanding of the pathophysiology of and risk factors for diseases and is also crucial for decision making related to workforce and resource planning in pediatric departments. In this study, a pediatric disease epidemiology knowledgebase called PedMap (http://pedmap.nbscn.org) was constructed from the clinical data from 5 447 202 outpatient visits of 2 189 868 unique patients at a children’s hospital (Hangzhou, China) from 2013 to 2016. The top 100 most-reported pediatric diseases were identified and visualized. These common pediatric diseases were clustered into 4 age groups and 4 seasons. The prevalence, age distribution and co-occurrence diseases for each disease were also visualized. Furthermore, an online prediction tool based on Gaussian regression models was developed to predict pediatric disease incidence based on weather information. PedMap is the first comprehensive epidemiological resource to show the full view of age-related, seasonal, climate-related variations in and co-occurrence patterns of pediatric diseases.

No abstract