It is still challenging to make accurate diagnosis of biliary atresia (BA) with sonographic gallbladder images particularly in rural area without relevant expertise. To help diagnose BA based on sonographic gallbladder images, an ensembled deep learning model is developed. The model yields a patient-level sensitivity 93.1% and specificity 93.9% [with areas under the receiver operating characteristic curve of 0.956 (95% confidence interval: 0.928-0.977)] on the multi-center external validation dataset, superior to that of human experts. With the help of the model, the performances of human experts with various levels are improved. Moreover, the diagnosis based on smartphone photos of sonographic gallbladder images through a smartphone app and based on video sequences by the model still yields expert-level performances. The ensembled deep learning model in this study provides a solution to help radiologists improve the diagnosis of BA in various clinical application scenarios, particularly in rural and undeveloped regions with limited expertise.

No abstract

No abstract

A new tetranuclear Zn(II) complex, [{Zn(L)(μ‐OAc)Zn(H2O)}2], based on an asymmetrical salamo‐type bisoxime chelating ligand H3L (6‐hydroxy‐4′‐chloro‐2,2′‐[ethylenediyldioxybis(nitrilomethylidyne)]diphenol) was synthesized and characterized by elemental analyses, differential thermal methods, single‐crystal X‐ray crystallography, and IR, UV–vis, and fluorescence spectra. The Zn(II) complex crystallizes in the triclinic system, space group P‐1 with cell parameters a = 9.0742(6) Å, b = 11.8225(5) Å, c = 12.4182(8) Å, Z = 2, V = 1212.56(12) Å3, R1 = 0.0572, and wR2 = 0.1734. The environment of the tetranuclear Zn(II) complex is penta‐coordinated having a slightly distorted trigonal bipyramidal geometry. Moreover, a 1D chain supramolecular structure is formed along the c‐axis by the intermolecular C1–H1B⋯O14 hydrogen bonds; in the same manner, C2–H2C⋯Cg2 functions in the formation of supramolecular structures along the a‐axis of the 1D chain. A 2D supramolecular structure along the ac plane extends infinitely under the force of intermolecular hydrogen bonds. Differential scanning calorimetry‐thermogravimetry thermal analysis provides evidence of the coordination of Zn(II) atoms to the ligand H3L. The Zn(II) complex shows intense photoluminescence with a maximum emission at ~453 nm upon excitation at 360 nm.

Background. The prognosis of patients with biliary atresia (BA) after Kasai portoenterostomy (KPE) varies, and precisely predicting the outcomes of KPE before surgery is still challenging. Methods. A total of 158 patients who underwent KPE in our hospital were included in this study. The patients in the training cohort were recruited from January 2012 to October 2017 (n = 118), and then, those in the validation cohort were recruited from November 2017 to April 2019 (n = 40). Combined nomogram models were developed based on twodimensional shear wave elastography (2D SWE) values and other biomarkers. The utility of the proposed models was evaluated by C-index. Results. 2D SWE played a potentially important role in predicting native liver survival (NLS) of BA patients with a C-index of 0.69 (0.63 to 0.75) in the training cohort and 0.76 (0.67 to 0.85) in the validation cohort. The nomogram A based on 2D SWE values, age, gamma-glutamyl transferase (GGT) and aspartate aminotransferase-to-platelet ratio (APRI) had a better C-index in the training cohort [0.74 (0.68-0.80) vs. 0.66 (0.60-0.73), P = 0.017] and in the validation cohort [0.78 (0.70-0.86) vs. 0.60 (0.49-0.71), P = 0.002] than the nomogram B (without 2D SWE). Using risk score developed from nomogram A, we successfully predicted 88.0% (22/25) of patients in the training cohort and 75.0% (9/12) in the validation cohort to have survival time of less than 12 months after KPE. Conclusion. The combined nomogram model based on 2D SWE values, age, GGT and APRI prior to KPE can effectively predict NLS in BA infants.