Extracorporeal membrane oxygenation (ECMO) is primarily used in clinical practice to provide continuous extracorporeal respiratory and circulatory support for patients with severe heart and lung failure, thereby sustaining life. It is a key technology for managing severe heart failure and respiratory failure that are difficult to control. With the accumulation of clinical experience in ECMO for circulatory and/or respiratory support, as well as advancements in biomedical engineering technology, more portable and stable ECMO devices have been introduced into clinical use, benefiting an increasing number of critically ill patients. Although ECMO technology has become relatively mature, the timing of ECMO initiation, management of sudden complications, and monitoring and early warning of physiological indicators are critical factors that greatly affect the therapeutic outcomes of ECMO. This article reviews traditional methods and artificial intelligence techniques used in risk assessment related to ECMO, including the latest achievements and research hotspots. Additionally, it discusses future trends in ECMO risk management, focusing on six key areas: multi-center and prospective studies, external validation and standardization of model performance, long-term prognosis considerations, integration of innovative technologies, enhancing model interpretability, and economic cost-effectiveness analysis. This provides a reference for future researchers to build models and explore new research directions.
Extracorporeal Membrane Oxygenation ; Humans ; Artificial Intelligence ; Risk Assessment ; Respiratory Insufficiency/therapy* ; Heart Failure/therapy*

Cardiovascular risk assessment is a basic tenet of the prevention of cardiovascular disease. Conventional risk assessment models require measurements of blood pressure, blood lipids, and other health-related information prior to assessment of risk via regression models. Compared with traditional approaches, fundus photograph-based cardiovascular risk assessment using artificial intelligence (AI) technology is novel, and has the advantages of immediacy, non-invasiveness, easy performance, and low cost. The Health Risk Assessment and Control Committee of the Chinese Preventive Medicine Association, in collaboration with the Chinese Society of Cardiology and the Society of Health Examination, invited multi-disciplinary experts to form a panel to develop the present consensus, which includes relevant theories, progress in research, and requirements for AI model development, as well as applicable scenarios, applicable subjects, assessment processes, and other issues associated with applying AI technology to assess cardiovascular risk based on fundus photographs. A consensus was reached after multiple careful discussions on the relevant research, and the needs of the health management industry in China and abroad, in order to guide the development and promotion of this new technology.
Humans ; Cardiovascular Diseases/prevention & control* ; Artificial Intelligence ; Consensus ; Risk Factors ; Heart Disease Risk Factors

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.
Male ; Humans ; Artificial Intelligence ; Retrospective Studies ; Echocardiography/methods* ; Heart Diseases ; Hypertension ; Hypothyroidism

OBJECTIVE: To use the low-cost anesthesia monitor for realizing anesthesia depth monitoring, effectively assist anesthesiologists in diagnosis and reduce the cost of anesthesia operation. METHODS: Propose a monitoring method of anesthesia depth based on artificial intelligence. The monitoring method is designed based on convolutional neural network (CNN) and long and short-term memory (LSTM) network. The input data of the model include electrocardiogram (ECG) and pulse wave photoplethysmography (PPG) recorded in the anesthesia monitor, as well as heart rate variability (HRV) calculated from ECG, The output of the model is in three states of anesthesia induction, anesthesia maintenance and anesthesia awakening. RESULTS: The accuracy of anesthesia depth monitoring model under transfer learning is 94.1%, which is better than all comparison methods. CONCLUSIONS The accuracy of this study meets the needs of perioperative anesthesia depth monitoring and the study reduces the operation cost.
Artificial Intelligence ; Neural Networks, Computer ; Heart Rate ; Electrocardiography ; Photoplethysmography/methods* ; Anesthesia

Humans ; Artificial Intelligence ; Heart Failure/therapy* ; Electrocardiography

The number of patients with heart failure in China is large, and the proportion of patients with end-stage heart failure continues to increase. The clinical effect of guideline-directed medications therapy for end-stage heart failure is poor. Heart transplantation is the most effective treatment for end-stage heart failure. But it is faced with many limitations such as the shortage of donors. In recent years, the research and development of artificial heart in China has made great progress. Three devices have been approved by the National Medical Products Administration for marketing, and another one is undergoing pre-marketing clinical trial. Since 2017, more than 200 cases of ventricular assist device implantation have been carried out in more than 34 hospitals in China. Among them, 70 patients in Fuwai Hospital, Chinese Academy of Medical Sciences had a 2-year survival rate of 90%. The first patient has survived more than 5 years with the device. More efforts should be put into the training of standardized technical team and quality control. Further research should be carried out in the aspects of pulsatile blood flow pump, fully implanted cable-free device, and improved biomaterial with better blood compatibility.
Humans ; Heart-Assist Devices ; Heart Failure/surgery* ; Heart, Artificial ; Heart Transplantation ; Pulsatile Flow

Heart failure is a disease that seriously threatens human health and has become a global public health problem. Diagnostic and prognostic analysis of heart failure based on medical imaging and clinical data can reveal the progression of heart failure and reduce the risk of death of patients, which has important research value. The traditional analysis methods based on statistics and machine learning have some problems, such as insufficient model capability, poor accuracy due to prior dependence, and poor model adaptability. In recent years, with the development of artificial intelligence technology, deep learning has been gradually applied to clinical data analysis in the field of heart failure, showing a new perspective. This paper reviews the main progress, application methods and major achievements of deep learning in heart failure diagnosis, heart failure mortality and heart failure readmission, summarizes the existing problems and presents the prospects of related research to promote the clinical application of deep learning in heart failure clinical research.
Humans ; Artificial Intelligence ; Deep Learning ; Heart Failure/diagnosis* ; Machine Learning ; Diagnostic Imaging

Humans ; Cardiac Resynchronization Therapy ; Cardiac Pacing, Artificial ; Heart Ventricles ; Heart Failure/therapy* ; Treatment Outcome ; Electrocardiography ; Ventricular Function, Left

Objective: Explore the relationship between tip of the left bundle branch pacing lead and anatomic location of left bundle branch as well as the mechanism of left bundle branch current of injury. To clarify the clinical value of left bundle branch current of injury during operation. Methods: The pacing leads were implanted in the hearts of two living swines. Intraoperative electrophysiological study confirmed that the left bundle branch or only the deep left ventricular septum was captured at low output. Immediately after operation, the gross specimen of swine hearts was stained with iodine to observe the gross distribution of His-purkinje conduction system on the left ventricular endocardium and its relationship with the leads. Subsequently, the swine hearts were fixed with formalin solution, and the pacing leads were removed after the positions were marked. The swine hearts were then sectioned and stained with Masson and Goldner trichrome, and the relationship between the anatomic location of the conduction system and the tip of the lead was observed under a light microscope. Results: After iodine staining of the specimen, the His-purkinje conduction system was observed with the naked eye in a net-like distribution, and the lead tip was screwed deeply and fixed in the left bundle branch area of the left ventricular subendocardium in the ventricular septum. Masson and Goldner trichrome staining showed that left bundle branch pacing lead directly passed through the left bundle branch when there was left bundle branch potential with left bundle branch current of injury, while it was not directly contact the left bundle branch when there was left bundle branch potential without left bundle branch current of injury. Conclusion: The left bundle branch current of injury observed on intracardiac electrocardiogram during His-purkinje conduction system pacing suggests that the pacing lead directly contacted the conduction bundle or its branches, therefore, the captured threshold was relatively low. Left bundle branch current of injury can be used as an important anatomic and electrophysiological evidence of left bundle branch capture.
Animals ; Swine ; Bundle of His/physiology* ; Ventricular Septum ; Cardiac Pacing, Artificial ; Heart Conduction System ; Electrocardiography ; Iodine

Objective: Explore the relationship between tip of the left bundle branch pacing lead and anatomic location of left bundle branch as well as the mechanism of left bundle branch current of injury. To clarify the clinical value of left bundle branch current of injury during operation. Methods: The pacing leads were implanted in the hearts of two living swines. Intraoperative electrophysiological study confirmed that the left bundle branch or only the deep left ventricular septum was captured at low output. Immediately after operation, the gross specimen of swine hearts was stained with iodine to observe the gross distribution of His-purkinje conduction system on the left ventricular endocardium and its relationship with the leads. Subsequently, the swine hearts were fixed with formalin solution, and the pacing leads were removed after the positions were marked. The swine hearts were then sectioned and stained with Masson and Goldner trichrome, and the relationship between the anatomic location of the conduction system and the tip of the lead was observed under a light microscope. Results: After iodine staining of the specimen, the His-purkinje conduction system was observed with the naked eye in a net-like distribution, and the lead tip was screwed deeply and fixed in the left bundle branch area of the left ventricular subendocardium in the ventricular septum. Masson and Goldner trichrome staining showed that left bundle branch pacing lead directly passed through the left bundle branch when there was left bundle branch potential with left bundle branch current of injury, while it was not directly contact the left bundle branch when there was left bundle branch potential without left bundle branch current of injury. Conclusion: The left bundle branch current of injury observed on intracardiac electrocardiogram during His-purkinje conduction system pacing suggests that the pacing lead directly contacted the conduction bundle or its branches, therefore, the captured threshold was relatively low. Left bundle branch current of injury can be used as an important anatomic and electrophysiological evidence of left bundle branch capture.
Animals ; Swine ; Bundle of His/physiology* ; Ventricular Septum ; Cardiac Pacing, Artificial ; Heart Conduction System ; Electrocardiography ; Iodine