Decision Support Tools, Systems, and Artificial Intelligence in Cardiac Imaging

Evaluating artificial intelligence-driven stress echocardiography analysis system (EASE study): A mixed method study

INTRODUCTION

The use and value of artificial intelligence (AI)-driven tools and techniques are under investigation in detecting coronary artery disease (CAD). EchoGo Pro is a patented AI-driven stress echocardiography analysis system produced by Ultromics Ltd. (henceforth Ultromics) to support clinicians in detecting cardiac ischaemia and potential CAD. This manuscript presents the research protocol for a field study to independently evaluate the accuracy, acceptability, implementation barriers, users' experience and willingness to pay, cost-effectiveness and value of EchoGo Pro.

METHODS AND ANALYSIS

The 'Evaluating AI-driven stress echocardiography analysis system' (EASE) study is a mixed-method evaluation, which will be conducted in five work packages (WPs). In WP1, we will examine the diagnostic accuracy by comparing test reports generated by EchoGo Pro and three manual raters. In WP2, we will focus on interviewing clinicians, innovation/transformation staff, and patients within the National Health Service (NHS), and staff within Ultromics, to assess the acceptability of this technology. In this WP, we will determine convergence and divergence between EchoGo Pro recommendations and cardiologists' interpretations and will assess what profile of cases is linked with convergence and divergence between EchoGo Pro recommendations and cardiologists' interpretations and how these link to outcomes. In WP4, we will conduct a quantitative cross-sectional survey of trust in AI tools applied to cardiac care settings among clinicians, healthcare commissioners and the general public. Lastly, in WP5, we will estimate the cost of deploying the EchoGo Pro technology, cost-effectiveness and willingness to pay cardiologists, healthcare commissioners and the general public. The results of this evaluation will support evidence-informed decision-making around the widespread adoption of EchoGo Pro and similar technologies in the NHS and other health systems.

ETHICS APPROVAL AND DISSEMINATION

This research has been approved by the NHS Health Research Authority (IRAS No: 315284) and the London South Bank University Ethics Panel (ETH2223-0164). Alongside journal publications, we will disseminate study methods and findings in conferences, seminars and social media. We will produce additional outputs in appropriate forms, for example, research summaries and policy briefs, for diverse audiences in NHS.

Computed Tomography Evaluation of Coronary Atherosclerosis: The Road Travelled, and What Lies Ahead

Coronary CT angiography (CCTA) is now endorsed by all major cardiology guidelines for the investigation of chest pain and assessment for coronary artery disease (CAD) in appropriately selected patients. CAD is a leading cause of morbidity and mortality. There is extensive literature to support CCTA diagnostic and prognostic value both for stable and acute symptoms. It enables rapid and cost-effective rule-out of CAD, and permits quantification and characterization of coronary plaque and associated significance. In this comprehensive review, we detail the road traveled as CCTA evolved to include quantitative assessment of plaque stenosis and extent, characterization of plaque characteristics including high-risk features, functional assessment including fractional flow reserve-CT (FFR-CT), and CT perfusion techniques. The state of current guideline recommendations and clinical applications are reviewed, as well as future directions in the rapidly advancing field of CT technology, including photon counting and applications of artificial intelligence (AI).

Optimizing Object Detection Algorithms for Congenital Heart Diseases in Echocardiography: Exploring Bounding Box Sizes and Data Augmentation Techniques

Background

Congenital heart diseases (CHDs), particularly atrial and ventricular septal defects, pose significant health risks and common challenges in detection via echocardiography. Doctors often employ the cardiac structural information during the diagnostic process. However, prior CHD research has not determined the influence of including cardiac structural information during the labeling process and the application of data augmentation techniques.

Methods

This study utilizes advanced artificial intelligence (AI)-driven object detection frameworks, specifically You Look Only Once (YOLO)v5, YOLOv7, and YOLOv9, to assess the impact of including cardiac structural information and data augmentation techniques on the identification of septal defects in echocardiographic images.

Results

The experimental results reveal that different labeling strategies substantially affect the performance of the detection models. Notably, adjustments in bounding box dimensions and the inclusion of cardiac structural details in the annotations are key factors influencing the accuracy of the model. The application of deep learning techniques in echocardiography enhances the precision of detecting septal heart defects.

Conclusions

This study confirms that careful annotation of imaging data is crucial for optimizing the performance of object detection algorithms in medical imaging. These findings suggest potential pathways for refining AI applications in diagnostic cardiology studies.

Machine Learning in Cardiology: A Potential Real-World Solution in Low- and Middle-Income Countries

Artificial intelligence (AI) and machine learning (ML) is a promising field of cardiovascular medicine. Many AI tools have been shown to be efficacious with a high level of accuracy. Yet, their use in real life is not well established. In the era of health technology and data science, it is crucial to consider how these tools could improve healthcare delivery. This is particularly important in countries with limited resources, such as low- and middle-income countries (LMICs). LMICs have many barriers in the care continuum of cardiovascular diseases (CVD), and big portion of these barriers come from scarcity of resources, mainly financial and human power constraints. AI/ML could potentially improve healthcare delivery if appropriately applied in these countries. Expectedly, the current literature lacks original articles about AI/ML originating from these countries. It is important to start early with a stepwise approach to understand the obstacles these countries face in order to develop AI/ML-based solutions. This could be detrimental to many patients' lives, in addition to other expected advantages in other sectors, including the economy sector. In this report, we aim to review what is known about AI/ML in cardiovascular medicine, and to discuss how it could benefit LMICs.

The Merits, Limitations, and Future Directions of Cost-Effectiveness Analysis in Cardiac MRI with a Focus on Coronary Artery Disease: A Literature Review

Cardiac magnetic resonance (CMR) imaging has a wide range of clinical applications with a high degree of accuracy for many myocardial pathologies. Recent literature has shown great utility of CMR in diagnosing many diseases, often changing the course of treatment. Despite this, it is often underutilized possibly due to perceived costs, limiting patient factors and comfort, and longer examination periods compared to other imaging modalities. In this regard, we conducted a literature review using keywords “Cost-Effectiveness” and “Cardiac MRI” and selected articles from the PubMed MEDLINE database that met our inclusion and exclusion criteria to examine the cost-effectiveness of CMR. Our search result yielded 17 articles included in our review. We found that CMR can be cost-effective in quality-adjusted life years (QALYs) in select patient populations with various cardiac pathologies. Specifically, the use of CMR in coronary artery disease (CAD) patients with a pretest probability below a certain threshold may be more cost-effective compared to patients with a higher pretest probability, although its use can be limited based on geographic location, professional society guidelines, and differing reimbursement patterns. In addition, a stepwise combination of different imaging modalities, with conjunction of AHA/ACC guidelines can further enhance the cost-effectiveness of CMR.

Artificial intelligence and the future of midwifery: What do midwives think about artificial intelligence? A qualitative study

The evidence on how AI will make a revolution is insufficient. Our aim was to investigate opinions of midwives on the future of AI and midwifery. Semi-structured interviews were done with 18 midwives in Turkey. Themes were identified: expectations included the advantages and conditional acceptance of robotic technology, prejudices reflected perceived shortcomings, lack of human competencies, and trust issues. Concerns included midwifery care and concerns about her future. Midwives were overwhelmingly skeptical about the replacement of human capabilities by AI and found the technology's potential limited.

Machine Learning Algorithms to Distinguish Myocardial Perfusion SPECT Polar Maps

Myocardial perfusion imaging (MPI) plays an important role in patients with suspected and documented coronary artery disease (CAD). Machine Learning (ML) algorithms have been developed for many medical applications with excellent performance. This study used ML algorithms to discern normal and abnormal gated Single Photon Emission Computed Tomography (SPECT) images. We analyzed one thousand and seven polar maps from a database of patients referred to a university hospital for clinically indicated MPI between January 2016 and December 2018. These studies were reported and evaluated by two different expert readers. The image features were extracted from a specific type of polar map segmentation based on horizontal and vertical slices. A senior expert reading was the comparator (gold standard). We used cross-validation to divide the dataset into training and testing subsets, using data augmentation in the training set, and evaluated 04 ML models. All models had accuracy >90% and area under the receiver operating characteristics curve (AUC) >0.80 except for Adaptive Boosting (AUC = 0.77), while all precision and sensitivity obtained were >96 and 92%, respectively. Random Forest had the best performance (AUC: 0.853; accuracy: 0,938; precision: 0.968; sensitivity: 0.963). ML algorithms performed very well in image classification. These models were capable of distinguishing polar maps remarkably into normal and abnormal.

Automated Echocardiographic Detection of Severe Coronary Artery Disease Using Artificial Intelligence

OBJECTIVES

The purpose of this study was to establish whether an artificially intelligent (AI) system can be developed to automate stress echocardiography analysis and support clinician interpretation.

BACKGROUND

Coronary artery disease is the leading global cause of mortality and morbidity and stress echocardiography remains one of the most commonly used diagnostic imaging tests.

METHODS

An automated image processing pipeline was developed to extract novel geometric and kinematic features from stress echocardiograms collected as part of a large, United Kingdom-based prospective, multicenter, multivendor study. An ensemble machine learning classifier was trained, using the extracted features, to identify patients with severe coronary artery disease on invasive coronary angiography. The model was tested in an independent U.S.

STUDY

How availability of an AI classification might impact clinical interpretation of stress echocardiograms was evaluated in a randomized crossover reader study.

RESULTS

Acceptable classification accuracy for identification of patients with severe coronary artery disease in the training data set was achieved on cross-fold validation based on 31 unique geometric and kinematic features, with a specificity of 92.7% and a sensitivity of 84.4%. This accuracy was maintained in the independent validation data set. The use of the AI classification tool by clinicians increased inter-reader agreement and confidence as well as sensitivity for detection of disease by 10% to achieve an area under the receiver-operating characteristic curve of 0.93.

CONCLUSION

Automated analysis of stress echocardiograms is possible using AI and provision of automated classifications to clinicians when reading stress echocardiograms could improve accuracy, inter-reader agreement, and reader confidence.

New Hybrid Method for Left Ventricular Ejection Fraction Assessment from Radionuclide Ventriculography Images

BACKGROUND

In this paper, we propose a new efficient method of radionuclide ventriculography image segmentation to estimate the left ventricular ejection fraction. This parameter is an important prognostic factor for diagnosing abnormal cardiac function.

METHODS

The proposed method combines the Chan-Vese and the mathematical morphology algorithms. It was applied to diastolic and systolic images obtained from the Nuclear Medicine Department of Salah AZAIEZ Institute. In order to validate our proposed method, we compare the obtained results to those of two methods present in the literature. The first one is based on mathematical morphology, while the second one uses the basic Chan-Vese algorithm. To evaluate the quality of segmentation, we compute accuracy, positive predictive value and area under the ROC curve. We also compare the left ventricle ejection fraction estimated by our method to that of the reference given by the software of the gamma-camera and validated by the expert, using Pearson's correlation coefficient, ANOVA test and linear regression.

RESULTS

Static results show that the proposed method is very efficient for the detection of the left ventricle. The accuracy was 98.60%, higher than that of the other two methods (95.52% and 98.50%).

CONCLUSION

Likewise, the positive predictive value was the highest (86.40% vs. 83.63% 71.82%). The area under the ROC curve was also the most important (0.998% vs. 0.926% 0.919%). On the other hand, Pearson's correlation coefficient was the highest (99% vs. 98% 37%). The correlation was significantly positive (p<0.001).

Accuracy of automated three-dimensional ultrasound imaging technique for fetal head biometry

OBJECTIVES

To evaluate the accuracy of an automated three-dimensional (3D) ultrasound technique for fetal intracranial measurements compared with manual acquisition.

METHODS

This was a prospective observational study of patients presenting for routine anatomical survey between 18 + 0 and 22 + 6 weeks' gestation. After providing informed consent, each patient underwent two consecutive ultrasound examinations of the fetal head, one by a sonographer and one by a physician. Each operator obtained manual measurements of the biparietal diameter (BPD), head circumference (HC), transcerebellar diameter (TCD), cisterna magna (CM) and posterior horn of the lateral ventricle (Vp), followed by automated measurements of these structures using an artificial intelligence-based tool, SonoCNS® Fetal Brain. Both operators repeated the automated approach until all five measurements were obtained in a single sweep, up to a maximum of three attempts. The accuracy of automated measurements was compared with that of manual measurements using intraclass correlation coefficients (ICC) by operator type, accounting for patient and ultrasound characteristics.

RESULTS

One hundred and forty-three women were enrolled in the study. Median body mass index was 24.0 kg/m² (interquartile range (IQR), 22.5-26.8 kg/m² ) and median subcutaneous thickness was 1.6 cm (IQR, 1.3-2.0 cm). Fifteen (10%) patients had at least one prior Cesarean delivery, 17 (12%) had other abdominal surgery and 78 (55%) had an anterior placenta. Successful acquisition of the automated measurements was achieved on the first, second and third attempts in 70%, 22% and 3% of patients, respectively, by sonographers and in 76%, 16% and 3% of cases, respectively, by physicians. The automated algorithm was not able to identify and measure all five structures correctly in six (4%) and seven (5%) patients scanned by the sonographers and physicians, respectively. The ICCs reflected good reliability (0.80-0.88) of the automated compared with the manual approach for BPD and HC and poor to moderate reliability (0.23-0.50) for TCD, CM and Vp. Fetal lie, head position, placental location, maternal subcutaneous thickness and prior Cesarean section were not associated with the success or accuracy of the automated technique.

CONCLUSIONS

Automated 3D ultrasound imaging of the fetal head using SonoCNS reliably identified and measured BPD and HC but was less consistent in accurately identifying and measuring TCD, CM and Vp. While these results are encouraging, further optimization of the automated technology is necessary prior to incorporation of the technique into routine sonographic protocols. © 2020 International Society of Ultrasound in Obstetrics and Gynecology.

Deep learning analysis in coronary computed tomographic angiography imaging for the assessment of patients with coronary artery stenosis

BACKGROUND AND OBJECTIVE

Recently, deep convolutional neural network has significantly improved image classification and image segmentation. If coronary artery disease (CAD) can be diagnosed through machine learning and deep learning, it will significantly reduce the burdens of the doctors and accelerate the critical patient diagnoses. The purpose of the study is to assess the practicability of utilizing deep learning approaches to process coronary computed tomographic angiography (CCTA) imaging (termed CCTA-artificial intelligence, CCTA-AI) in coronary artery stenosis.

MATERIALS AND METHODS

A CCTA reconstruction pipeline was built by utilizing deep learning and transfer learning approaches to generate auto-reconstructed CCTA images based on a series of two-dimensional (2D) CT images. 150 patients who underwent successively CCTA and digital subtraction angiography (DSA) from June 2017 to December 2017 were retrospectively analyzed. The dataset was divided into two parts comprising training dataset and testing dataset. The training dataset included the CCTA images of 100 patients which are trained using convolutional neural networks (CNN) in order to further identify various plaque classifications and coronary stenosis. The other 50 CAD patients acted as testing dataset that is evaluated by comparing the auto-reconstructed CCTA images with traditional CCTA images on the condition that DSA images are regarded as the reference method. Receiver operating characteristic (ROC) analysis was used for statistical analysis to compare CCTA-AI with DSA and traditional CCTA in the aspect of detecting coronary stenosis and plaque features.

RESULTS

AI significantly reduces time for post-processing and diagnosis comparing to the traditional methods. In identifying various degrees of coronary stenosis, the diagnostic accuracy of CCTA-AI is better than traditional CCTA (AUC_AI = 0.870, AUC_CCTA = 0.781, P < 0.001). In identifying ≥ 50% stenotic vessels, the accuracy, sensitivity, specificity, positive predictive value and negative predictive value of CCTA-AI and traditional method are 86% and 83%, 88% and 59%, 85% and 94%, 73% and 84%, 94% and 83%, respectively. In the aspect of identifying plaque classification, accuracy of CCTA-AI is moderate compared to traditional CCTA (AUC = 0.750, P < 0.001).

CONCLUSION

The proposed CCTA-AI allows the generation of auto-reconstructed CCTA images from a series of 2D CT images. This approach is relatively accurate for detecting ≥50% stenosis and analyzing plaque features compared to traditional CCTA.

Artificial Intelligence and Health Technology Assessment: Anticipating a New Level of Complexity

Artificial intelligence (AI) is seen as a strategic lever to improve access, quality, and efficiency of care and services and to build learning and value-based health systems. Many studies have examined the technical performance of AI within an experimental context. These studies provide limited insights into the issues that its use in a real-world context of care and services raises. To help decision makers address these issues in a systemic and holistic manner, this viewpoint paper relies on the health technology assessment core model to contrast the expectations of the health sector toward the use of AI with the risks that should be mitigated for its responsible deployment. The analysis adopts the perspective of payers (ie, health system organizations and agencies) because of their central role in regulating, financing, and reimbursing novel technologies. This paper suggests that AI-based systems should be seen as a health system transformation lever, rather than a discrete set of technological devices. Their use could bring significant changes and impacts at several levels: technological, clinical, human and cognitive (patient and clinician), professional and organizational, economic, legal, and ethical. The assessment of AI's value proposition should thus go beyond technical performance and cost logic by performing a holistic analysis of its value in a real-world context of care and services. To guide AI development, generate knowledge, and draw lessons that can be translated into action, the right political, regulatory, organizational, clinical, and technological conditions for innovation should be created as a first step.

Assistive game controller for artificial intelligence-enhanced telerehabilitation post-stroke

Off-the-shelf gaming technology is designed for young, fit, and motor intact individuals. Artificial intelligence (AI) has a role in making controllers and therapeutic games adaptable to the disabled. Post-stroke rehabilitation outcomes can be enhanced by gaming technology within the home to enable engaging telerehabilitation. BrightBrainer™ Grasp (BBG) is a novel therapeutic game controller designed to adapt to arm and hand impairments post-stroke. It mediates intensive arm reach, grasp and finger extension training and has the ability to track relevant outcomes. The newly designed controller uses BrightBrainer gamification system with AI technology to provide automatic adaptation, requiring minimal clinician input. This article describes the BBG design, hardware, force and movement detection and calibration, and its integration with the therapeutic games. The use of AI in adapting a library of custom therapeutic games is also described. Results of a usability study with healthy individuals and related design modifications are presented, with implications for future trials.

The 21st Century Echocardiography Laboratory in Australia and New Zealand: Rapid Evolution of Training and Workforce, Practice and Technology

Echocardiography is a common and increasingly used noninvasive imaging tool in medicine. In this paper, we imagine the echocardiography laboratory of the future and consider the challenges we face currently, and may face in the future, and how these might be overcome; challenges such as training enough sonographers to meet the increasing demands of the ageing population living with chronic cardiovascular disease and the need for surveillance in other clinical scenarios. We consider the changing qualification framework and the requirements for accreditation and registration in Australia and New Zealand and the potential for migrant sonographers to meet some of the increasing demand. Advanced scopes of practice are likely to be a feature of the future workforce and we consider some of the ways these may evolve. Lastly, we consider how the evolving clinical landscape and technology may change the way echocardiography is delivered.