Assessment of left ventricular ejection fraction in artificial intelligence based on left ventricular opacification.

Background: Left ventricular opacification (LVO) improves the accuracy of left ventricular ejection fraction (LVEF) by enhancing the visualization of the endocardium. Manual delineation of the endocardium by sonographers has observer variability. Artificial intelligence (AI) has the potential to improve the reproducibility of LVO to assess LVEF. Objectives: The aim was to develop an AI model and evaluate the feasibility and reproducibility of LVO in the assessment of LVEF. Methods: This retrospective study included 1305 echocardiography of 797 patients who had LVO at the Department of Ultrasound Medicine, Union Hospital, Huazhong University of Science and Technology from 2013 to 2021. The AI model was developed by 5-fold cross validation. The validation datasets included 50 patients prospectively collected in our center and 42 patients retrospectively collected in the external institution. To evaluate the differences between LV function determined by AI and sonographers, the median absolute error (MAE), spearman correlation coefficient, and intraclass correlation coefficient (ICC) were calculated. Results: In LVO, the MAE of LVEF between AI and manual measurements was 2.6% in the development cohort, 2.5% in the internal validation cohort, and 2.7% in the external validation cohort. Compared with two-dimensional echocardiography (2DE), the left ventricular (LV) volumes and LVEF of LVO measured by AI correlated significantly with manual measurements. AI model provided excellent reliability for the LV parameters of LVO (ICC > 0.95). Conclusions: AI-assisted LVO enables more accurate identification of the LV endocardium and reduces observer variability, providing a more reliable way for assessing LV function.

Obesity and hypertension mediate the effect of education on deep intracerebral hemorrhage: A Mendelian randomization study.

BACKGROUND: Educational attainment (EA) as a stable indicator of socioeconomic status has been confirmed to affect intracerebral hemorrhage (ICH), but the mechanism relating EA and ICH is still unknown. AIM: To explore the causal relationship between EA and ICH through a bidirectional and two-step Mendelian randomization (MR) study. METHODS: Using summary-level Genome-wide Association Study using GWAS data FROM CASES AND CONTROLS of European ancestry, we performed bidirectional and two-step MR analyses to explore the causal relationship between educational attainment and ICH to understand the mediating influence of risk factors in this process. We also carried out subgroup analysis according to the different sites (deep and lobar) of ICH. A set of sensitivity analyses were performed to test valid MR assumptions. RESULTS: Bidirectional MR analysis consistently demonstrated a unidirectional causal effect, revealing that higher EA had a protective influence on ICH. Each additional 1-standard deviation (SD) increase in genetically predicted years of schooling was associated with a reduced risk of all ICH (inverse variance weighted (IVW) OR: 0.381 [95 %CI: 0.264-0.549]), deep ICH (OR: 0.334 [95 %CI: 0.216-0.517]), and lobar ICH (OR: 0.422 [95 %CI: 0.261-0.682]). The mediating effect of EA on all ICH was mediated via systolic blood pressure (SBP) (6.93 % [1.20-13.45 %]) and body mass index (BMI) (17.87 % [3.92-34.64 %]), and the mediating effect of EA on deep ICH was also mediated via SBP (7.85 % [1.55-15.07 %]) and BMI (18.63 % [4.02-36.26 %]). CONCLUSION: This study provides robust genetic evidence for supporting the protective effect of EA on ICH risk, with further evidence that the effect of EA on deep ICH is partially mediated through hypertension and obesity. Further validation is needed to ascertain whether these findings are applicable to other racial or general population groups.

Association between serum creatinine and 30 days all-cause mortality in critically ill patients with non-traumatic subarachnoid hemorrhage: analysis of the MIMIC-IV database.

Background: Serum creatinine is a prognostic marker for various conditions, but its significance of spontaneous subarachnoid hemorrhage is still poorly understood. This study aims to elucidate the correlation between admission serum creatinine (sCr) levels and all-cause mortality within 30 days among individuals affected by non-traumatic subarachnoid hemorrhage (SAH). Methods: This cohort study included 672 non-traumatic SAH adults. It utilized data from the MIMIC-IV database from 2008 to 2019. The patients' first-time serum creatinine was recorded. Subsequently, an examination of the 30-day all-cause mortality was conducted. Employing a multiple logistic regression model, a nomogram was constructed, while the association between sCr and 30-day all-cause mortality was evaluated using Kaplan-Meier survival curves. The calibration curve was employed to assess the model's performance, while subgroup analysis was employed to examine the impact of additional complications and medication therapy on outcomes. Results: A total of 672 patients diagnosed with non-traumatic subarachnoid hemorrhage were included in the study. The mortality rate within this timeframe was found to be 24.7%. Multiple logistic regression analysis revealed that sCr served as an independent prognostic indicator for all-cause mortality within 30 days of admission for SAH patients [OR: 2(1.18-3.41); p = 0.01]. A comprehensive model was constructed, incorporating age, sCr, white blood cell count (WBC), glucose, anion gap, and partial thromboplastin time (PTT), resulting in a prediction model with an AUC value of 0.806 (95% CI: 0.768, 0.843), while the AUC for the test set is 0.821 (95% CI: 0.777-0.865). Conclusion: Creatinine emerges as a significant biomarker, closely associated with heightened in-hospital mortality in individuals suffering from SAH.

Causal association between depression and intracranial aneurysms: a bidirectional two-sample Mendelian randomization study.

Background: Although observational studies have suggested a bidirectional relation between depression and intracranial aneurysms (IAs), their causal relations remain unclear. Thus we aimed to assess the causal association between depression and IAs. Methods: We conducted a bidirectional two-sample Mendelian randomization (MR) study using summary-level data from publicly available genome-wide association studies of depression (n = 500,199), IAs (n = 79,429), unruptured intracranial aneurysm (uIA) (n = 74,004), and subarachnoid hemorrhage (SAH) (n = 77,074). MR analyses included the inverse-variance weighted (IVW) method as the primary analytic, plus weighted-median, simple mode, weighted mode, MR-Egger, and MR PRESSO. Results: Genetically predicted depression was strongly positively related to IAs (odds ratio [OR] = 1.69, 95% confidence interval [CI] 1.19-2.39, p = 0.003), uIA (OR = 1.96, 95% CI 1.06-3.64, p = 0.032), and SAH (OR = 1.73, 95% CI 1.14-2.61, p = 0.009). Reverse MR analyses showed that while genetically predicted uIA was positively related to depression (OR = 1.02, 95% CI 1.00-1.05, p = 0.044), no causal relations were observed for either IAs or SAH for depression. Conclusion: Our findings provide evidence of a causal effect of depression on IAs, uIA, and SAH. For the reverse MR analyses, we found a causal impact of uIA on depression, but no causal influence of either IAs or SAH for depression.

Development and Validation of a Prediction Model for 30-Day Mortality and Functional Outcome in Patients with Primary Brainstem Hemorrhage.

INTRODUCTION: Primary brainstem hemorrhage (PBSH) is the most fatal subtype of intracerebral hemorrhage and is associated with poor prognosis. We aimed to develop a prediction model for predicting 30-day mortality and functional outcome in patients with PBSH. METHODS: We reviewed records of 642 consecutive patients with first-time PBSH from three hospitals between 2016 and 2021. Multivariate logistic regression was used to establish a nomogram in a training cohort. Cutoff points of the variables were determined by receiver operating characteristic curve analysis, and certain points were assigned to these predictors to produce the PBSH score. The nomogram and PBSH score were compared with other scoring systems for PBSH. RESULTS: Five independent predictors, comprised of temperature, pupillary light reflex, platelet-to-lymphocyte ratio, Glasgow Coma Scale (GCS) score on admission, and hematoma volume, were incorporated to construct the nomogram. The PBSH score consisted of 4 independent factors with individual points assigned as follows: temperature, ≥38°C (=1 point), <38°C (=0 points); pupillary light reflex, absence (=1 point), presence (=0 points); GCS score 3-4 (=2 points), 5-11 (=1 point), and 12-15 (=0 points); PBSH volume >10 mL (=2 points), 5-10 mL (=1 point), and <5 mL (=0 points). Results showed that the nomogram was discriminative in predicting both 30-day mortality (area under the ROC curve [AUC] of 0.924 in the training cohort, and 0.931 in the validation cohort) and 30-day functional outcome (AUC of 0.887). The PBSH score was discriminative in predicting both 30-day mortality (AUC of 0.923 in the training cohort and 0.923 in the validation cohort) and 30-day functional outcome (AUC of 0.887). The prediction performances of the nomogram and the PBSH score were superior to the intracranial hemorrhage (ICH) score, primary pontine hemorrhage (PPH) score, and new PPH score. CONCLUSIONS: We developed and validated two prediction models for 30-day mortality and functional outcome in patients with PBSH. The nomogram and PBSH score could predict 30-day mortality and functional outcome in PBSH patients.

Association between lactate/albumin ratio and 28-day all-cause mortality in ischemic stroke patients without reperfusion therapy: a retrospective analysis of the MIMIC-IV database.

Objective: The lactate/albumin ratio (LAR) has been used as a novel prognostic indicator for aneurysmal subarachnoid hemorrhage, traumatic brain injury, sepsis, heart failure, and acute respiratory failure. However, its potential in predicting all-cause mortality in patients with ischemic stroke (IS) has not been evaluated. Therefore, this study aimed to elucidate the correlation between LAR and 28-day all-cause mortality in IS patients without reperfusion therapy. Methods: This retrospective cohort study used data from the Medical Information Mart for Intensive Care (MIMIC-IV) (v2.0) database. It included 568 IS adult patients admitted to the intensive care unit (ICU). The correlation between LAR and ICU 28-day all-cause mortality rate was analyzed using multiple COX regression analysis and Kaplan-Meier survival analysis. Restricted cubic spline (RCS) curves were used to assess the relationship between LAR and 28-day mortality. In addition, a subgroup analysis was performed to investigate the impact of other influencing factors on outcomes. The primary outcome was the ability of LAR to predict 28-day mortality in IS patients. Results: Among the 568 patients with IS, 370 survived (survival group) and 198 died (non-survival group) within 28 days of admission (mortality rate: 34.9%). A multivariate COX regression analysis indicated that LAR was an independent predictor of all-cause mortality within 28 days after admission for patients with IS (hazard ratio: 1.32; 95% confidence interval: 1.03-1.68; P = 0.025). We constructed a model that included LAR, age, race, sex, white blood cell count, Sequential Organ Failure Assessment (SOFA) score, and anion gap (AG) and established a prediction model with an area under the curve (AUC) value of 71.5% (95% confidence interval: 67.1%-75.8%). The optimal cutoff value of LAR that separated the survival group and the non-survival group based on the Youden index was 0.55. The Kaplan-Meier survival curves plotted using this critical value showed that patients with LAR ≥ 0.55 had a significantly higher 28-day all-cause mortality rate than patients with LAR < 0.55 (P = 0.0083). Conclusion: LAR can serve as an independent predictor of all-cause mortality within 28 days after admission for patients with IS.

Co-learning of appearance and shape for precise ejection fraction estimation from echocardiographic sequences.

Accurate estimation of ejection fraction (EF) from echocardiography is of great importance for evaluation of cardiac function. It is usually obtained by the Simpson's bi-plane method based on the segmentation of the left ventricle (LV) in two keyframes. However, obtaining accurate EF estimation from echocardiography is challenging due to (1) noisy appearance in ultrasound images, (2) temporal dynamic movement of myocardium, (3) sparse annotation of the full sequence, and (4) potential quality degradation during scanning. In this paper, we propose a multi-task semi-supervised framework, which is denoted as MCLAS, for precise EF estimation from echocardiographic sequences of two cardiac views. Specifically, we first propose a co-learning mechanism to explore the mutual benefits of cardiac segmentation and myocardium tracking iteratively on appearance level and shape level, therefore alleviating the noisy appearance and enforcing the temporal consistency of the segmentation results. This temporal consistency, as shown in our work, is critical for precise EF estimation. Then we propose two auxiliary tasks for the encoder, (1) view classification to help extract the discriminative features of each view, and automatize the whole pipeline of EF estimation in clinical practice, and (2) EF regression to help regularize the spatiotemporal embedding of the echocardiographic sequence. Both two auxiliary tasks can improve the segmentation-based EF prediction, especially for sequences of poor quality. Our method is capable of automating the whole pipeline of EF estimation, from view identification, cardiac structures segmentation to EF calculation. The effectiveness of our method is validated in aspects of segmentation, tracking, consistency analysis, and clinical parameters estimation. When compared with existing methods, our method shows obvious superiority for LV volumes on ED and ES phases, and EF estimation, with Pearson correlation of 0.975, 0.983 and 0.946, respectively. This is a significant improvement for echocardiography-based EF estimation and improves the potential of automated EF estimation in clinical practice. Besides, our method can obtain accurate and temporal-consistent segmentation for the in-between frames, which enables it for cardiac dynamic function evaluation.

Improved Segmentation of Echocardiography With Orientation-Congruency of Optical Flow and Motion-Enhanced Segmentation.

Quantification of left ventricular (LV) ejection fraction (EF) from echocardiography depends upon the identification of endocardium boundaries as well as the calculation of end-diastolic (ED) and end-systolic (ES) LV volumes. It's critical to segment the LV cavity for precise calculation of EF from echocardiography. Most of the existing echocardiography segmentation approaches either only segment ES and ED frames without leveraging the motion information, or the motion information is only utilized as an auxiliary task. To address the above drawbacks, in this work, we propose a novel echocardiography segmentation method which can effectively utilize the underlying motion information by accurately predicting optical flow (OF) fields. First, we devised a feature extractor shared by the segmentation and the optical flow sub-tasks for efficient information exchange. Then, we proposed a new orientation congruency constraint for the OF estimation sub-task by promoting the congruency of optical flow orientation between successive frames. Finally, we design a motion-enhanced segmentation module for the final segmentation. Experimental results show that the proposed method achieved state-of-the-art performance for EF estimation, with a Pearson correlation coefficient of 0.893 and a Mean Absolute Error of 5.20% when validated with echo sequences of 450 patients.

Automatic view classification of contrast and non-contrast echocardiography.

Background: Contrast and non-contrast echocardiography are crucial for cardiovascular diagnoses and treatments. Correct view classification is a foundational step for the analysis of cardiac structure and function. View classification from all sequences of a patient is laborious and depends heavily on the sonographer's experience. In addition, the intra-view variability and the inter-view similarity increase the difficulty in identifying critical views in contrast and non-contrast echocardiography. This study aims to develop a deep residual convolutional neural network (CNN) to automatically identify multiple views of contrast and non-contrast echocardiography, including parasternal left ventricular short axis, apical two, three, and four-chamber views. Methods: The study retrospectively analyzed a cohort of 855 patients who had undergone left ventricular opacification at the Department of Ultrasound Medicine, Wuhan Union Medical College Hospital from 2013 to 2021, including 70.3% men and 29.7% women aged from 41 to 62 (median age, 53). All datasets were preprocessed to remove sensitive information and 10 frames with equivalent intervals were sampled from each of the original videos. The number of frames in the training, validation, and test datasets were, respectively, 19,370, 2,370, and 2,620 from 9 views, corresponding to 688, 84, and 83 patients. We presented the CNN model to classify echocardiographic views with an initial learning rate of 0.001, and a batch size of 4 for 30 epochs. The learning rate was decayed by a factor of 0.9 per epoch. Results: On the test dataset, the overall classification accuracy is 99.1 and 99.5% for contrast and non-contrast echocardiographic views. The average precision, recall, specificity, and F1 score are 96.9, 96.9, 100, and 96.9% for the 9 echocardiographic views. Conclusions: This study highlights the potential of CNN in the view classification of echocardiograms with and without contrast. It shows promise in improving the workflow of clinical analysis of echocardiography.

Higher cystatin C level increases the risk of delayed cerebral ischemia after endovascular treatment of aneurysmal subarachnoid hemorrhage: a case-control study.

Cystatin C (CysC) has been found to be associated with hemorrhagic and ischemic stroke in many studies. However, the association between CysC level and the risk of delayed cerebral ischemia after endovascular treatment of aneurysmal subarachnoid hemorrhage has been reported rarely. Our study was proposed to explore this association. Consecutive patients from June 2015 to February 2021 in this single-center retrospective study were selected. Univariate and multivariate analyses were used to identify potential prognostic risk factors for delayed cerebral ischemia, and the stability of the association was demonstrated by several statistical methods, such as subgroup analysis, interaction testing, generalized linear models, and propensity score matching. A total of 424 patients were included in the analysis. Cystatin C was independently associated with delayed cerebral ischemia. The independent effects of CysC on delayed cerebral ischemia were shown in generalized linear models with a logit link, and the results were relatively stable in crude, partial, and full models with ORs (95% CIs) for delayed cerebral ischemia. Subgroup analysis showed no significant subgroup differences in the effect of CysC on delayed cerebral ischemia. There was also no interaction effect between CysC and other confounders. Patients in the high CysC group had a higher risk of delayed cerebral ischemia than those in the low CysC group before and after propensity score matching. CysC level could be an independent predictor for the risk of delayed cerebral ischemia after endovascular treatment of aneurysmal subarachnoid hemorrhage.

Improved collateral flow and reduced damage after remote ischemic perconditioning during distal middle cerebral artery occlusion in aged rats.

Circulation through cerebral collaterals can maintain tissue viability until reperfusion is achieved. However, collateral circulation is time limited, and failure of collaterals is accelerated in the aged. Remote ischemic perconditioning (RIPerC), which involves inducing a series of repetitive, transient peripheral cycles of ischemia and reperfusion at a site remote to the brain during cerebral ischemia, may be neuroprotective and can prevent collateral failure in young adult rats. Here, we demonstrate the efficacy of RIPerC to improve blood flow through collaterals in aged (16-18 months of age) Sprague Dawley rats during a distal middle cerebral artery occlusion. Laser speckle contrast imaging and two-photon laser scanning microscopy were used to directly measure flow through collateral connections to ischemic tissue. Consistent with studies in young adult rats, RIPerC enhanced collateral flow by preventing the stroke-induced narrowing of pial arterioles during ischemia. This improved flow was associated with reduced early ischemic damage in RIPerC treated aged rats relative to controls. Thus, RIPerC is an easily administered, non-invasive neuroprotective strategy that can improve penumbral blood flow via collaterals. Enhanced collateral flow supports further investigation as an adjuvant therapy to recanalization therapy and a protective treatment to maintain tissue viability prior to reperfusion.

Impaired Collateral Flow in Pial Arterioles of Aged Rats During Ischemic Stroke.

Cerebral collateral circulation and age are critical factors in determining outcome from acute ischemic stroke. Aging may lead to rarefaction of cerebral collaterals, and thereby accelerate ischemic injury by reducing penumbral blood flow. Dynamic changes in pial collaterals after onset of cerebral ischemia may vary with age but have not been extensively studied. Here, laser speckle contrast imaging (LSCI) and two-photon laser scanning microscopy (TPLSM) were combined to monitor cerebral pial collaterals between the anterior cerebral artery (ACA) and the middle cerebral artery (MCA) in young adult and aged male Sprague Dawley rats during distal middle cerebral artery occlusion (dMCAo). Histological analysis showed that aged rats had significantly greater volumes of ischemic damage than young rats. LSCI showed that cerebral collateral perfusion declined over time after stroke in aged and young rats, and that this decline was significantly greater in aged rats. TPLSM demonstrated that pial arterioles narrowed faster after dMCAo in aged rats compared to young adult rats. Notably, while arteriole vessel narrowing was comparable 4.5 h after ischemic onset in aged and young adult rats, red blood cell velocity was stable in young adults but declined over time in aged rats. Overall, red blood cell flux through pial arterioles was significantly reduced at all time-points after 90 min post-dMCAo in aged rats relative to young adult rats. Thus, collateral failure is more severe in aged rats with significantly impaired pial collateral dynamics (reduced diameter, red blood cell velocity, and red blood cell flux) relative to young adult rats.

Prevention of the collapse of pial collaterals by remote ischemic perconditioning during acute ischemic stroke.

Collateral circulation is a key variable determining prognosis and response to recanalization therapy during acute ischemic stroke. Remote ischemic perconditioning (RIPerC) involves inducing peripheral ischemia (typically in the limbs) during stroke and may reduce perfusion deficits and brain damage due to cerebral ischemia. In this study, we directly investigated pial collateral flow augmentation due to RIPerC during distal middle cerebral artery occlusion (MCAo) in rats. Blood flow through pial collaterals between the anterior cerebral artery (ACA) and the MCA was assessed in male Sprague Dawley rats using in vivo laser speckle contrast imaging (LSCI) and two photon laser scanning microscopy (TPLSM) during distal MCAo. LSCI and TPLSM revealed that RIPerC augmented collateral flow into distal MCA segments. Notably, while control rats exhibited an initial dilation followed by a progressive narrowing of pial arterioles 60 to 150-min post-MCAo (constricting to 80-90% of post-MCAo peak diameter), this constriction was prevented or reversed by RIPerC (such that vessel diameters increased to 105-110% of post-MCAo, pre-RIPerC diameter). RIPerC significantly reduced early ischemic damage measured 6 h after stroke onset. Thus, prevention of collateral collapse via RIPerC is neuroprotective and may facilitate other protective or recanalization therapies by improving blood flow in penumbral tissue.

[Research on Application of Fourier Transform Infrared Spectrometry in the Diagnosis of Lymph Node Metastasis in Gastric Cancer].

To explore the feasibility of quick intraoperative in situ and noninvasive diagnosis of lymph node metastasis in gastric cancer by Fourier transform infrared (FTIR) spectrometry. FTIR spectra of surgically removed fresh lymph nodes were measured by FTIR via probe of attenuated total reflection (ATR). For each spectrum, 13 bands were indentified and assigned between 3 000 and 1 000 cm(-1). Peaks in the spectra were measured and relative intensity ratios were calculated and compared between the spectra of Metastatic lymph nodes (MLN) and Non-metastatic lymph nodes (NMLN). Standard statistic analysis was performed. 720 lymph nodes were measured in 38 gastric cancer patients. Results show that there were significant differences between the FTIR of 540 MLN and 180 NMLN. (1) For the band related to nucleic acid: The ratios of I1240/I1460 (p = 0.015) and I1080/I1460 (p = 0.034) increased in MLN, which shows that the relative quantity of nucleic acid was more in MLN than that in NMLN. (2) For the bands related to protein: The ratios of I1640 /I1460 (p = 0.001) and I146/I1460 (p = 0.027) increased in MLN, which shows that the relative quantity of protein was more in MLN. (3) For the bands related to lipid: The ratio of I2855/I460 and I1740/I1460 decreased in MLN FTIR spectrum, indicating the lower relative quantity of lipid in MLN. (4) For the bands related to carbohydrate: The ratio of I1160/I1460 (p = 0.023) decreased in MLN FTIR spectrum, indicating the lower relative quantity of carbohydrate in MLN. The results demonstrate that the FTIR spectroscopy technique maybe develop into a promising method for in situ and quick intraoperative differential diagnosis of lymph node metastasis in gastric cancer.