A method to screen left ventricular dysfunction through ECG based on convolutional neural network.

OBJECTIVE: This study aims to develop an artificial intelligence-based method to screen patients with left ventricular ejection fraction (LVEF) of 50% or lesser using electrocardiogram (ECG) data alone. METHODS: Convolutional neural network (CNN) is a class of deep neural networks, which has been widely used in medical image recognition. We collected standard 12-lead ECG and transthoracic echocardiogram (TTE) data including the LVEF value. Then, we paired the ECG and TTE data from the same individual. For multiple ECG-TTE pairs from a single individual, only the earliest data pair was included. All the ECG-TTE pairs were randomly divided into the training, validation, or testing data set in a ratio of 9:1:1 to create or evaluate the CNN model. Finally, we assessed the screening performance by overall accuracy, sensitivity, specificity, positive predictive value, and negative predictive value. RESULTS: We retrospectively enrolled a total of 26 786 ECG-TTE pairs and randomly divided them into training (n = 21 732), validation (n = 2 530), and testing data set (n = 2 530). In the testing set, the CNN algorithm showed an overall accuracy of 73.9%, sensitivity of 69.2%, specificity of 70.5%, positive predictive value of 70.1%, and negative predictive value of 69.9%. CONCLUSION: Our results demonstrate that a well-trained CNN algorithm may be used as a low-cost and noninvasive method to identify patients with left ventricular dysfunction.

Distortion of terahertz signals due to imperfect synchronization with chirped probe pulses.

Terahertz (THz) signals measured by means of the spectral-encoding technique with different temporal discrepancies between probe pulses and THz signals are investigated. It is found that imperfect synchronization between the chirped probe and THz pulses induce a distortion and this distortion affects significantly the retrieved THz spectrum if the temporal discrepancy is large. The distortion becomes more prominent if the probe pulse length is less than the optimal chirped probe pulse duration. A simple approach is proposed to realize the synchronization and minimize the distortion. THz signals from a high-voltage-biased air plasma filament are measured with this approach and distortion similar to the simulation results is observed.

AChE inhibitory alkaloids from Coptis chinensis.

Coptis chinensis is a widely used traditional Chinese herbal medicine. In this work, 6 new alkaloids (coptisine A-F, 1-6) and 26 known alkaloids (7-32) were isolated from the chloroform extract of the rhizomes of C. chinensis. Compounds 1-3 are α-carbonylated benzylisoquinolines, and 4-6 are berberidic acid type alkaloids. Their structures were elucidated on the basis of extensive NMR and MS analyses. Seven compounds (7, 20, 22, 23, 25, 26, 29) exhibited significant AChE inhibitory activities at 10 µM (inhibition rates >80%).

Characterization of chemical constituents and rats metabolites of an alkaloidal extract of Alstonia scholaris leaves by liquid chromatography coupled with mass spectrometry.

Alstonia scholaris has been used in "Dai" ethnic medicine to treat chronic respiratory diseases for a long history, and the major bioactive constituents are alkaloids. An alkaloidal extract of A. scholaris leaves (AAS) has been developed into an investigational new drug, and has been approved for phase I/II clinical trials by China Food and Drug Administration. However, little is known on the chemical composition and in vivo metabolism of AAS, thus far. In this study, an ultra-high performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (UHPLC/qTOF-MS) method was established to characterize the chemical constituents of AAS. Samples were separated on an ACQUITY UPLC CSH column (2.1×100mm, 1.7µm) with acetonitrile and water containing 0.3% formic acid as the mobile phase. On the basis of high-accuracy mass spectral analysis, a total of 35 alkaloids were characterized from AAS, including 11 scholaricine-type, 9 vallesamine-type, 12 picrinine-type, and 3 tubotaiwine-type alkaloids. Furthermore, the metabolic pathways of 4 representative alkaloids in rats were studied. They mainly undertook hydroxylation and glucuronidation reactions. Based on the above metabolic pathways, the metabolism of AAS (10mg/kg) in rats after oral administration was studied by LC/MS. A total of 33 compounds in plasma, 40 compounds in urine, and 38 compounds in feces were characterized. The results indicated that scholaricine-type alkaloids could get into circulation more readily than the other types. This is the first systematic study on chemical profiling and metabolites identification of AAS.

New triterpene saponins from the roots of Glycyrrhiza yunnanensis and their rapid screening by LC/MS/MS.

The roots of Glycyrrhiza yunnanensis Cheng f. et L. K. Dai ex P. C. Li are used as licorice in traditional medicine of Southwest China. Triterpene saponins are the major chemical constituents. In this study, one new oleanane-type triterpenoid, glyyunnansapogenin I (I), seven new triterpene saponins, yunganosides E3 (II), L (III), M (IV), N1 (V), O (VI), P (VII) and N2 (VIII), together with four known saponins (IX-XII) were isolated from the roots of G. yunnanensis by preparative chromatography. Their structures were identified by spectroscopic analysis including NMR and HR-MS. Based on (-)-ESI-MS/MS fragmentation behaviors of these reference standards, an LC/MS/MS method using neutral loss scan and precursor ion scan on a triple quadrupole mass spectrometer was established to rapidly and comprehensively analyze triterpene saponins in G. yunnanensis. Combined with high-accuracy qTOF mass spectrometry analysis, a total of 50 saponins were detected, and their structures were identified or tentatively characterized. This is the first systematic study on triterpene saponins in G. yunnanensis.