Poincaré plot can help predict the curative effect of metoprolol for pediatric postural orthostatic tachycardia syndrome.

Purpose: To study whether a Poincaré plot can help predict the curative effect of metoprolol for postural orthostatic tachycardia syndrome (POTS) in children. Methods: Pediatric patients with POTS who were administered metoprolol were retrospectively included. The collected data included general data (sex, age, height, weight, and body mass index), the manifestations and treatment (baseline orthostatic intolerance symptom score and course of metoprolol treatment), vital signs (supine heart rate [HR], supine blood pressure, and increased HR during the standing test), HR variability indexes (standard deviation of normal-to-normal intervals [SDNN]; standard deviation of the averages of normal-to-normal intervals [SDANN]; mean standard deviation of the NN intervals for each 5-min segment [SDNNI]; root mean square of the successive differences [rMSSD]; percentage of adjacent NN intervals that differ by >50 ms [pNN50]; triangular index; ultra-low [ULF], very low [VLF], low [LF], and high frequency [HF]; total power [TP]; and LF/HF ratio), and graphical parameters of the Poincaré plot (longitudinal axis [L], transverse axis [T], and L/T). Receiver operator characteristic curves were used to calculate the predictive function of the indexes with significant differences between patients who responded and those who did not. The index combination with the highest predictive value was obtained through series-parallel analysis. Results: Overall, 40 responders and 23 non-responders were included. The L and T in the Poincaré plots and rMSSD, pNN50, HF, and TP of the HR variability data were significantly lower in participants who responded to metoprolol than in participants who did not (p < 0.001). The L/T of participants who responded to metoprolol was greater than that of non-responders (p < 0.001). Moreover, we noted a strong correlation between every two indexes among L, T, rMSSD, pNN50, HF, TP, and L/T (p < 0.05). T < 573.9 ms combined with L/T > 2.9 had the best performance for predicting the effectiveness of metoprolol, with a sensitivity of 85.0%, specificity of 82.6%, and accuracy of 84.1%. Conclusion: In the Poincaré plot, a T < 573.9 ms combined with an L/T > 2.9 helps predict good outcomes of using metoprolol to treat pediatric POTS.

Controllable Image Synthesis With Attribute-Decomposed GAN.

This paper proposes Attribute-Decomposed GAN (ADGAN) and its enhanced version (ADGAN++) for controllable image synthesis, which can produce realistic images with desired attributes provided in various source inputs. The core ideas of the proposed ADGAN and ADGAN++ are both to embed component attributes into the latent space as independent codes and thus achieve flexible and continuous control of attributes via mixing and interpolation operations in explicit style representations. The major difference between them is that ADGAN processes all component attributes simultaneously while ADGAN++ utilizes a serial encoding strategy. More specifically, ADGAN consists of two encoding pathways with style block connections and is capable of decomposing the original hard mapping into multiple more accessible subtasks. In the source pathway, component layouts are extracted via a semantic parser and the segmented components are fed into a shared global texture encoder to obtain decomposed latent codes. This strategy allows for the synthesis of more realistic output images and the automatic separation of un-annotated component attributes. Although the original ADGAN works in a delicate and efficient manner, intrinsically it fails to handle the semantic image synthesizing task when the number of attribute categories is huge. To address this problem, ADGAN++ employs the serial encoding of different component attributes to synthesize each part of the target real-world image, and adopts several residual blocks with segmentation guided instance normalization to assemble the synthesized component images and refine the original synthesis result. The two-stage ADGAN++ is designed to alleviate the massive computational costs required when synthesizing real-world images with numerous attributes while maintaining the disentanglement of different attributes to enable flexible control of arbitrary component attributes of the synthesized images. Experimental results demonstrate the proposed methods' superiority over the state of the art in pose transfer, face style transfer, and semantic image synthesis, as well as their effectiveness in the task of component attribute transfer. Our code and data are publicly available at https://github.com/menyifang/ADGAN.

Poincaré Plot Is Useful for Distinguishing Vasovagal Syncope From Postural Tachycardia Syndrome in Children.

Objectives: To explore the role of the Poincaré plot derived from a 24-hour Holter recording in distinguishing vasovagal syncope (VVS) from postural tachycardia syndrome (POTS) in pediatric patients. Materials and Methods: Pediatric patients with VVS or POTS, hospitalized in Peking University First Hospital between January 2012 and December 2018, were included in a derivation study. The transverse axis (T), longitudinal axis (L), T/L ratio, product T × L, distance between the origin and the proximal end of the longitudinal axis (pro-D), and distance between the origin and distal end of the longitudinal axis (dis-D) of the Poincaré plot were compared between the VVS and POTS groups, and the differential diagnostic performance of the above-mentioned graphic parameters was evaluated using receiver operating characteristic curve analysis. A validation study was conducted in pediatric patients hospitalized between January 2019 and December 2020. Results: In school-aged children, the T, L, T/L, T × L, and dis-D values of patients with VVS were greater than those of patients with POTS; in adolescents, the T, T/L, T × L, and pro-D values of patients with VVS were greater than those of patients with POTS. Using a T/L cut-off value of 0.3 to distinguish between the two diseases, the sensitivity and specificity were 91.0 and 90.5%, respectively, for the total participants; 91.6 and 88.9%, respectively, for the school-aged children; and 82.1 and 95.7%, respectively, for the adolescents. In the validation study, a T/L cut-off value of 0.3 yielded an accuracy, sensitivity, and specificity of 81.8, 87.2, and 77.6%, respectively, in the total participants; 76.5, 82.6, and 71.4%, respectively, in the school-aged children; and 89.2, 93.8, and 85.7%, respectively, in the adolescents, in distinguishing VVS from POTS validated by clinical diagnosis. Conclusions: The graphic parameters of the Poincaré plot are significantly different between VVS and POTS in pediatric patients, and the T/L of the Poincaré plot may be a useful measure to help differentiate VVS from POTS in children and adolescents.