2024-12

Health Science Reports

Background Explainable artificial intelligence (XAI) emerged to improve the transparency of machine learning models and increase understanding of how models make actions and decisions. It helps to present complex models in a more digestible form from a human perspective. However, XAI is still in the development stage and must be used carefully in sensitive domains including paediatrics, where misuse might have adverse consequences. Objective This commentary paper discusses concerns and challenges related to implementation and interpretation of XAI methods, with the aim of rising awareness of the main concerns regarding their adoption in paediatrics. Methods A comprehensive literature review was undertaken to explore the challenges of adopting XAI in paediatrics. Results Although XAI has several favorable outcomes, its implementation in paediatrics is prone to challenges including generalizability, trustworthiness, causality and intervention, and XAI evaluation. Conclusion Paediatrics is a very sensitive domain where consequences of misinterpreting AI outcomes might be very significant. XAI should be adopted carefully with focus on evaluating the outcomes primarily by including paediatricians in the loop, enriching the pipeline by injecting domain knowledge promoting a cross-fertilization perspective aiming at filling the gaps still preventing its adoption.

2024-09

European Heart Journal-Imaging Methods and Practice

Background Automated algorithms are regularly used to analyse cardiac magnetic resonance (CMR) images. Validating data output reliability from this method is crucial for enabling widespread adoption. We outline a visual quality control (QC) process for image analysis using automated batch processing. We assess the performance of automated analysis and the reliability of replacing visual checks with statistical outlier removal approach in UK Biobank CMR scans. Methods We included 1,987 CMR scans from the UK Biobank COVID imaging study. We used batch processing software (Circle Cardiovascular Imaging Inc. - CVI42) to automatically extract chamber volumetric data, strain, native T1, and aortic flow data. The automated analysis outputs (∼62,000 videos and 2,000 images) were visually checked by six experienced clinicians using a standardised approach and a custom-built R Shiny app. Inter-observer variability was assessed. Data from scans passing visual QC was compared with a statistical outlier removal QC method in a subset of healthy individuals (n = 1069). Results Automated segmentation was highly rated, with over 95% of scans passing visual QC. Overall inter-observer agreement was very good (Gwet’s AC2 0.91; 95% confidence interval [0.84,0.94]). No difference in overall data derived from visual QC or statistical outlier removal in healthy individuals was observed. Conclusion Automated image analysis using CVI42 prototypes for UK Biobank CMR scans demonstrated high quality. Larger UK Biobank datasets analysed using these automated algorithms do not require in-depth visual QC. Statistical outlier removal is sufficient as a QC measure, with operator discretion for visual checks based on population or research objectives.

2024-08

Artificial Intelligence Review (الحجم : 57)

Explainable artificial intelligence (XAI) elucidates the decision-making process of complex AI models and is important in building trust in model predictions. XAI explanations themselves require evaluation as to accuracy and reasonableness and in the context of use of the underlying AI model. This review details the evaluation of XAI in cardiac AI applications and has found that, of the studies examined, 37% evaluated XAI quality using literature results, 11% used clinicians as domain-experts, 11% used proxies or statistical analysis, with the remaining 43% not assessing the XAI used at all. We aim to inspire additional studies within healthcare, urging researchers not only to apply XAI methods but to systematically assess the resulting explanations, as a step towards developing trustworthy and safe models.

2024-06

Advanced Intelligent Systems (القضية : 2400304)

eXplainable artificial intelligence (XAI) methods have emerged to convert the black box of machine learning (ML) models into a more digestible form. These methods help to communicate how the model works with the aim of making ML models more transparent and increasing the trust of end-users in their output. SHapley Additive exPlanations (SHAP) and Local Interpretable Model Agnostic Explanation (LIME) are two widely used XAI methods, particularly with tabular data. In this perspective piece, the way the explainability metrics of these twomethods are generated is discussed and a framework for the interpretation of their outputs, highlighting their weaknesses and strengths is proposed.Specifically, their outcomes in terms of model-dependency and in the presence of collinearity among the features, relying on a case study from the biomedical domain (classification of individuals with or without myocardial infarction) are discussed. The results indicate that SHAP and LIME are highly affected by the adopted ML model and feature collinearity, raising a note of caution on their usage and interpretation

2024-05

IEEE Journal of Biomedical and Health Informatics

Explainable Artificial Intelligence (XAI) provides tools to help understanding how AI models work and reach a particular decision or outcome. It helps to increase the interpretability of models and makes them more trustworthy and transparent. In this context, many XAI methods have been proposed to make black-box and complex models more digestible from a human perspective. However, one of the main issues that XAI methods have to face especially when dealing with a high number of features is the presence of multicollinearity, which casts shadows on the robustness of the XAI outcomes, such as the ranking of informative features. Most of the current XAI methods either do not consider the collinearity or assume the features are independent which, in general, is not necessarily true. Here, we propose a simple, yet useful, proxy that modifies the outcome of any XAI feature ranking method allowing to account for the dependency among the features, and to reveal their impact on the outcome. The proposed method was applied to SHAP, as an example of XAI method which assume that the features are independent. For this purpose, several models were exploited for a well-known classification task (males versus females) using nine cardiac phenotypes extracted from cardiac magnetic resonance imaging as features. Principal component analysis and biological plausibility were employed to validate the proposed method. Our results showed that the proposed proxy could lead to a more robust list of informative features compared to the original SHAP in presence of collinearity.

2024-04

JACC: Cardiovascular Imaging

Population aging is one of the most important demographic transformations of our time. Increasing the “health span”—the proportion of life spent in good health—is a global priority. Biological aging comprises molecular and cellular modifications over many years, which culminate in gradual physiological decline across multiple organ systems and predispose to age-related illnesses. Cardiovascular disease is a major cause of ill health and premature death in older people. The rate at which biological aging occurs varies across individuals of the same age and is influenced by a wide range of genetic and environmental exposures. The authors review the hallmarks of biological cardiovascular aging and their capture using imaging and other noninvasive techniques and examine how this information may be used to understand aging trajectories, with the aim of guiding individual- and population-level interventions to promote healthy aging.

2024-04

JACC: Cardiovascular Imaging

Background The absence of population-stratified cardiovascular magnetic resonance (CMR) reference ranges from large cohorts is a major shortcoming for clinical care. Objectives This paper provides age-, sex-, and ethnicity-specific CMR reference ranges for atrial and ventricular metrics from the Healthy Hearts Consortium, an international collaborative comprising 9,088 CMR studies from verified healthy individuals, covering the complete adult age spectrum across both sexes, and with the highest ethnic diversity reported to date. Methods CMR studies were analyzed using certified software with batch processing capability (cvi42, version 5.14 prototype, Circle Cardiovascular Imaging) by 2 expert readers. Three segmentation methods (smooth, papillary, anatomic) were used to contour the endocardial and epicardial borders of the ventricles and atria from long- and short-axis cine series. Clinically established ventricular and atrial metrics were extracted and stratified by age, sex, and ethnicity. Variations by segmentation method, scanner vendor, and magnet strength were examined. Reference ranges are reported as 95% prediction intervals. Results The sample included 4,452 (49.0%) men and 4,636 (51.0%) women with average age of 61.1 ± 12.9 years (range: 18-83 years). Among these, 7,424 (81.7%) were from White, 510 (5.6%) South Asian, 478 (5.3%) mixed/other, 341 (3.7%) Black, and 335 (3.7%) Chinese ethnicities. Images were acquired using 1.5-T (n = 8,779; 96.6%) and 3.0-T (n = 309; 3.4%) scanners from Siemens (n = 8,299; 91.3%), Philips (n = 498; 5.5%), and GE (n = 291, 3.2%). Conclusions This work represents a resource with healthy CMR-derived volumetric reference ranges ready for clinical implementation.

2024-04

JBMR Plus

This study examined the association of estimated heel bone mineral density (eBMD, derived from quantitative ultrasound) with: 1) prevalent and incident cardiovascular diseases (CVDs: ischaemic heart disease (IHD), myocardial infarction (MI), heart failure, non-ischaemic cardiomyopathy (NICM), arrhythmia), 2) mortality (all-cause, CVD, IHD), and 3) cardiovascular magnetic resonance (CMR) measures of left ventricular and atrial structure and function and aortic distensibility, in the UK Biobank. Clinical outcomes were ascertained using health record linkage over 12.3 years of prospective follow-up. Two-sample Mendelian randomization (MR) was conducted to assess causal associations between BMD and CMR metrics using genetic instrumental variables identified from published genome-wide association studies. The analysis included 485 257 participants (55% women, mean age 56.5 ± 8.1 years). Higher heel eBMD was associated with lower odds of all prevalent CVDs considered. The greatest magnitude of effect was seen in association with heart failure and NICM, where 1-SD increase in eBMD was associated with 15% lower odds of heart failure and 16% lower odds of NICM. Association between eBMD and incident IHD and MI were non-significant; the strongest relationship was with incident heart failure (SHR: 0.90 [95% CI: 0.89-0.92]). Higher eBMD was associated with a decreased risk in all-cause, CVD, and IHD mortality, in the fully adjusted model. Higher eBMD was associated with greater aortic distensibility; associations with other CMR metrics were null. Higher heel eBMD is linked to reduced risk of a range of prevalent and incident CVD and mortality outcomes. While observational analyses suggest associations between higher eBMD and greater aortic compliance, MR analysis did not support a causal relationship between genetically predicted BMD and CMR phenotypes. These findings support the notion that bone-cardiovascular associations reflect shared risk factors/mechanisms rather than direct causal pathways.

2024-02

Journal of the American Heart Association (القضية : 3) (الحجم : 13)

BACKGROUNDExisting research demonstrates the association of shorter leukocyte telomere length with increased risk of age‐related health outcomes including cardiovascular diseases. However, the direct causality of these relationships has not been definitively established. Cardiovascular aging at an organ level may be captured using image‐derived phenotypes of cardiac anatomy and function. METHODS AND RESULTSIn the current study, we use 2‐sample Mendelian randomization to assess the causal link between leukocyte telomere length and 54 cardiac magnetic resonance imaging measures representing structure and function across the 4 cardiac chambers. Genetically predicted shorter leukocyte telomere length was causally linked to smaller ventricular cavity sizes including left ventricular end‐systolic volume, left ventricular end‐diastolic volume, lower left ventricular mass, and pulmonary artery. The association with left ventricular mass (β =0.217, Pfalse discovery rate=0.016) remained significant after multiple testing adjustment, whereas other associations were attenuated. CONCLUSIONSOur findings support a causal role for shorter leukocyte telomere length and faster cardiac aging, with the most prominent relationship with left ventricular mass.

2018-08

International Journal of Advanced Computer Science and Applications (القضية : 8) (الحجم : 9)

As the use of technology is increasing rapidly, the amount of shared, sent, and received information is also increas-ing in the same way. As a result, this necessitates the need for finding techniques that can save and secure the information over the net. There are many methods that have been used to protect the information such as hiding information and encryption. In this study, we propose a new encryption method making use of amino acid and DNA sequences. In addition, several criteria including data size, key size and the probability of cracking are used to evaluate the proposed method. The results show that the performance of the proposed method is better than many common encryption methods, such as RSA in terms of evaluation criteria.

2016-09

European Scientific Journal, ESJ (القضية : 27) (الحجم : 12)

XML files are used widely to save the information especially in the field of bioinformatics about the whole genome. There are many programming languages and modules used to parse XML files in different platforms such as Linux, Macintosh and Windows. The aim of this report is to reveal and determine which common programming language to use and on which platform is better to parse XML files in terms of memory usage, CPU time consumption and executing time.

2016-05

European Scientific Journal, ESJ (القضية : 15) (الحجم : 12)

The process involve in building a successful website goes beyond knowing web developing languages such as HTML and CSS. The success or failure of any website depends heavily on the use of some aspects of Graphical User Interface. These aspects are usually reserved for client’s wishes and desires. The content features of the GUI are used to stay in touch with visitors. Therefore, this article will highlight some important GUI features used in building an attractive and a successful website.