[Recognition of abnormal changes in echocardiographic videos by an artificial intelligence assisted diagnosis model based on 3D CNN].

Objective: To investigate the diagnostic efficiency and clinical application value of an artificial intelligence-assisted diagnosis model based on a three-dimensional convolutional neural network (3D CNN) on echocardiographic videos of patients with hypertensive heart disease, chronic renal failure (CRF) and hypothyroidism with cardiac involvement. Methods: This study is a retrospective study. The patients with hypertensive heart disease, CRF and hypothyroidism with cardiac involvement, who admitted in Henan Provincial People's Hospital from April 2019 to October 2021, were enrolled. Patients were divided into hypertension group, CRF group, and hypothyroidism group. Additionally, a simple random sampling method was used to select control healthy individuals, who underwent physical examination at the same period. The echocardiographic video data of enrolled participants were analyzed. The video data in each group was divided into a training set and an independent testing set in a ratio of 5 to 1. The temporal and spatial characteristics of videos were extracted using an inflated 3D convolutional network (I3D). The artificial intelligence assisted diagnosis model was trained and tested. There was no case overlapped between the training and validation sets. A model was established according to cases or videos based on video data from 3 different views (single apical four chamber (A4C) view, single parasternal left ventricular long-axis (PLAX) view and all views). The statistical analysis of diagnostic performance was completed to calculate sensitivity, specificity and area under the ROC curve (AUC). The time required for the artificial intelligence and ultrasound physicians to process cases was compared. Results: A total of 730 subjects aged (41.9±12.7) years were enrolled, including 362 males (49.6%), and 17 703 videos were collected. There were 212 cases in the hypertensive group, 210 cases in the CRF group, 105 cases in the hypothyroidism group, and 203 cases in the normal control group. The diagnostic performance of the model predicted by cases based on single PLAX view and all views data was excellent: (1) in the hypertensive group, the sensitivity, specificity and AUC of models based on all views data were 97%, 89% and 0.93, respectively, while those of models based on a single PLAX view were 94%, 95%, and 0.94, respectively; (2) in the CRF group, the sensitivity, specificity and AUC of models based on all views data were 97%, 95% and 0.96, respectively, while those of models based on a single PLAX view were 97%, 89%, and 0.93, respectively; (3) in the hypothyroidism group, the sensitivity, specificity and AUC of models based on all views data were 64%, 100% and 0.82, respectively, while those of models based on a single PLAX view were 82%, 89%, and 0.86, respectively. The time required for the 3D CNN model to measure and analyze the echocardiographic videos of each subject was significantly shorter than that for the ultrasound physicians ((23.96±6.65)s vs. (958.25±266.17)s, P<0.001). Conclusions: The artificial intelligence assisted diagnosis model based on 3D CNN can extract the dynamic temporal and spatial characteristics of echocardiographic videos jointly, and quickly and efficiently identify hypertensive heart disease and cardiac changes caused by CRF and hypothyroidism.

[The value of myocardial work in detecting the reduction of left ventricular global systolic function in acute myocardial infarction patients with preserved ejection fraction].

Objective: To investigate the diagnosis value of myocardial work (MW) in evaluating left ventricular global systolic function among acute myocardial infarction (AMI) patients with preserved ejection fraction (LVEF). Methods: This study was a diagnostic trial in a prospective case-control design. AMI patients with preserved LVEF were enrolled as AMI(LVEF>50%) group and age and sex-matched healthy individuals undergoing healthy checkup in our hospital were collected as control group. Two-dimensional dynamic images, including standard apical two-chamber, long-axis and four-chamber views in 3 consecutive cardiac cycles were acquired. General clinical data, routine echocardiography and myocardial work parameters were obtained from all subjects. The indices were compared between the two groups. Intra-observer and inter-observer repeatability of myocardial work parameters were evaluated by intra-group correlation coefficient (ICC). Receiver operator characteristic (ROC) curve was used to determine the diagnostic value of global constructive work (GCW), global wasted work (GWW), global work efficiency (GWE) and global work index (GWI) on the reduction of left ventricular global systolic function in AMI(LVEF>50%) group. Results: There were 30 patients in AMI(LVEF>50%) group, the age was (67.3±9.7)years, and 14 cases were female(46.7%). Thirty participants were included in the control group, the age was (68.1±8.6)years, and 12 cases were female (40.0%). Compared with the control group, left ventricular end-diastolic diameter (LVEDD), left ventricular end-systolic diameter (LVESD), interventricular septum thickness (IVSD), left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV), peak early diastolic velocity of mitral orifice/tissue Doppler velocity of posterior mitral annulus (E/e), left ventricular mass (LVM), left ventricular mass index (LVMI) were significantly higher, while E and e values were significantly lower in AMI (LVEF>50%) group (all P<0.05). Compared with the control group, GCW ((1 145.9±440.1)mmHg% (1 mmHg=0.133 kPa) vs. (1 425.7±355.4)mmHg%), GWE ((80.9±9.5)% vs. (87.3±5.5)%), GWI ((1 001.3±416.2)mmHg% vs. (1 247.6±341.7)mmHg%) and the absolute value of long axis integral strain (GLS) ((8.5±3.4)% vs. (11.4±3.7)%) were significantly lower, while peak strain dispersion(PSD)((101.3±66.4)ms vs. (74.7±31.9)ms) was significantly higher in AMI(LVEF>50%) group (P all<0 05). There was no significant difference in GWW((177.2±71.1)mmHg% vs. (155.7±64.6)mmHg%) between the two groups (P>0.05). The reproducibility of GCW, GWW, GWE and GWI within and between observers were satisfactory (all ICC>0.75). ROC curve analysis showed that all four parameters, GCW, GWW, GWE, and GWI, could be used for the diagnosis of reduced left ventricular systolic function in patients with AMI (LVEF>50%), and their areas under the ROC curve were 0.896,0.929,0.808,0.862. Conclusion: Myocardial work assessment is valuable on diagnosing left ventricular global systolic function reduction in AMI patients with preserved LVEF.

Network over-connectivity differentiates autism spectrum disorder from other developmental disorders in toddlers: A diffusion MRI study.

Advanced connectivity studies in toddlers with Autism Spectrum Disorder (ASD) are increasing and consistently reporting a disruption of brain connectivity. However, most of these studies compare ASD and typically developing subjects, thus providing little information on the specificity of the abnormalities detected in comparison with other developmental disorders (other-DD). We recruited subjects aged below 36 months who received a clinical diagnosis of Neurodevelopmental Disorder (32 ASD and 16 other-DD including intellectual disability and language disorder) according to DSM-IV TR. Structural and diffusion MRI were acquired to perform whole brain probabilistic and anatomically constrained tractography. Network connectivity matrices were built encoding the number of streamlines (DNUM ) and the tract-averaged fractional anisotropy (DFA ) values connecting each pair of cortical and subcortical regions. Network Based Statistics (NBS) was finally applied on the connectivity matrices to evaluate the network differences between the ASD and other-DD groups. The network differences resulted in an over-connectivity pattern (i.e., higher DNUM and DFA values) in the ASD group with a significance of P < 0.05. No contra-comparison results were found. The over-connectivity pattern in ASD occurred in networks primarily involving the fronto-temporal nodes, known to be crucial for social-skill development and basal ganglia, related to restricted and repetitive behaviours in ASD. To our knowledge, this is the first network-based diffusion study comparing toddlers with ASD and those with other-DD. Results indicate the detection of different connectivity patterns in ASD and other-DD at an age when clinical differential diagnosis is often challenging. Hum Brain Mapp 38:2333-2344, 2017. © 2017 Wiley Periodicals, Inc.

Influence of glutathione levels and activity of glutathione-related enzymes in the brains of tumor-bearing mice.

Oxidative stress takes place due to an imbalance between the production of reactive oxygen species (ROS) and the protection provided by cellular antioxidants. High levels of ROS are caused by tumor cells during tumor progression and may affect the functions of other important organs. The present study sought to investigate whether non-primary brain tumors affect reduced glutathione (GSH) levels and the activity of related enzymes in the brain. GSH contents, the activity of glutathione peroxidase (GPx), glutathione-s-transferase (GST), glutathione reductase (GR) as well as glutamate cysteine ligase (GCL) were determined in the brains of normal and tumor-bearing mice treated with the chemotherapy drug 5-Fluorouracil (5-Fu) or not. The results in S180 and H22 tumor-bearing mice showed that GSH levels and the activity of GPx, GST, and GCL all decreased while GR activity markedly increased in the brains of tumor-bearing mice compared to those of normal mice. Further investigation found that 5-Fu, a typical chemotherapy drug, significantly inhibited tumor growth but did not improve the loss of redox homeostasis in the brain caused by non-primary brain tumors. Overall, these results suggest that non-primary brain tumors can induce an ROS burden in the brain that cannot be reversed by the chemotherapy drug 5-Fu.