Automatic Identification of Patients With Unexplained Left Ventricular Hypertrophy in Electronic Health Record Data to Improve Targeted Treatment and Family Screening

Applying artificial intelligence to rare diseases: a literature review highlighting lessons from Fabry disease

BACKGROUND

Use of artificial intelligence (AI) in rare diseases has grown rapidly in recent years. In this review we have outlined the most common machine-learning and deep-learning methods currently being used to classify and analyse large amounts of data, such as standardized images or specific text in electronic health records. To illustrate how these methods have been adapted or developed for use with rare diseases, we have focused on Fabry disease, an X-linked genetic disorder caused by lysosomal α-galactosidase. A deficiency that can result in multiple organ damage.

METHODS

We searched PubMed for articles focusing on AI, rare diseases, and Fabry disease published anytime up to 08 January 2025. Further searches, limited to articles published between 01 January 2021 and 31 December 2023, were also performed using double combinations of keywords related to AI and each organ affected in Fabry disease, and AI and rare diseases.

RESULTS

In total, 20 articles on AI and Fabry disease were included. In the rare disease field, AI methods may be applied prospectively to large populations to identify specific patients, or retrospectively to large data sets to diagnose a previously overlooked rare disease. Different AI methods may facilitate Fabry disease diagnosis, help monitor progression in affected organs, and potentially contribute to personalized therapy development. The implementation of AI methods in general healthcare and medical imaging centres may help raise awareness of rare diseases and prompt general practitioners to consider these conditions earlier in the diagnostic pathway, while chatbots and telemedicine may accelerate patient referral to rare disease experts. The use of AI technologies in healthcare may generate specific ethical risks, prompting new AI regulatory frameworks aimed at addressing these issues to be established in Europe and the United States.

CONCLUSION

AI-based methods will lead to substantial improvements in the diagnosis and management of rare diseases. The need for a human guarantee of AI is a key issue in pursuing innovation while ensuring that human involvement remains at the centre of patient care during this technological revolution.

Deep learning for electrocardiogram interpretation: Bench to bedside

BACKGROUND

Recent advancements in deep learning (DL), a subset of artificial intelligence, have shown the potential to automate and improve disease recognition, phenotyping and prediction of disease onset and outcomes by analysing various sources of medical data. The electrocardiogram (ECG) is a valuable tool for diagnosing and monitoring cardiovascular conditions.

METHODS

The implementation of DL in ECG analysis has been used to detect and predict rhythm abnormalities and conduction abnormalities, ischemic and structural heart diseases, with performance comparable to physicians. However, despite promising development of DL algorithms for automatic ECG analysis, the integration of DL-based ECG analysis and deployment of medical devices incorporating these algorithms into routine clinical practice remains limited.

RESULTS

This narrative review highlights the applications of DL in 12-lead ECG analysis. Furthermore, we review randomized controlled trials that assess the clinical effectiveness of these DL tools. Finally, it addresses different key barriers to widespread implementation in clinical practice, including regulatory hurdles, algorithm transparency and data privacy concerns.

CONCLUSIONS

By outlining both the progress and the obstacles in this field, this review aims to provide insights into how DL could shape the future of ECG analysis and enhance cardiovascular care in daily clinical practice.

Screening for Left Ventricular Hypertrophy Using Artificial Intelligence Algorithms Based on 12 Leads of the Electrocardiogram-Applicable in Clinical Practice?-Critical Literature Review with Meta-Analysis

Background/Objectives: The increasing utilization of artificial intelligence (AI) in the medical field holds the potential to address the global shortage of doctors. However, various challenges, such as usability, privacy, inequality, and misdiagnosis, complicate its application. This literature review focuses on AI's role in cardiology, specifically its impact on the diagnostic accuracy of AI algorithms analyzing 12-lead electrocardiograms (ECGs) to detect left ventricular hypertrophy (LVH). Methods: Following PRISMA 2020 guidelines, we conducted a comprehensive search of PubMed, CENTRAL, Google Scholar, Web of Science, and Cochrane Library. Eligible studies included randomized controlled trials (RCTs), observational studies, and case-control studies across various settings. This review is registered in the PROSPERO database (registration number 531468). Results: Seven significant studies were selected and included in our review. Meta-analysis was performed using RevMan. Co-CNN (with incorporated demographic data and clinical variables) demonstrated the highest weighted average sensitivity at 0.84. 2D-CNN models (with demographic features) showed a balanced performance with good sensitivity (0.62) and high specificity (0.82); Co-CNN models excelled in sensitivity (0.84) but had lower specificity (0.71). Traditional ECG criteria (SLV and CV) maintained high specificities but low sensitivities. Scatter plots revealed trends between demographic factors and performance metrics. Conclusions: AI algorithms can rapidly analyze ECG data with high sensitivity. The diagnostic accuracy of AI models is variable but generally comparable to classical criteria. Clinical data and the training population of AI algorithms play a critical role in their efficacy. Future research should focus on collecting diverse ECG data across different populations to improve the generalizability of AI algorithms.

Searching for the Best Machine Learning Algorithm for the Detection of Left Ventricular Hypertrophy from the ECG: A Review

Background: Left ventricular hypertrophy (LVH) is a powerful predictor of future cardiovascular events. Objectives: The objectives of this study were to conduct a systematic review of machine learning (ML) algorithms for the identification of LVH and compare them with respect to the classical features of test sensitivity, specificity, accuracy, ROC and the traditional ECG criteria for LVH. Methods: A search string was constructed with the operators "left ventricular hypertrophy, electrocardiogram" AND machine learning; then, Medline and PubMed were systematically searched. Results: There were 14 studies that examined the detection of LVH utilizing the ECG and utilized at least one ML approach. ML approaches encompassed support vector machines, logistic regression, Random Forest, GLMNet, Gradient Boosting Machine, XGBoost, AdaBoost, ensemble neural networks, convolutional neural networks, deep neural networks and a back-propagation neural network. Sensitivity ranged from 0.29 to 0.966 and specificity ranged from 0.53 to 0.99. A comparison with the classical ECG criteria for LVH was performed in nine studies. ML algorithms were universally more sensitive than the Cornell voltage, Cornell product, Sokolow-Lyons or Romhilt-Estes criteria. However, none of the ML algorithms had meaningfully better specificity, and four were worse. Many of the ML algorithms included a large number of clinical (age, sex, height, weight), laboratory and detailed ECG waveform data (P, QRS and T wave), making them difficult to utilize in a clinical screening situation. Conclusions: There are over a dozen different ML algorithms for the detection of LVH on a 12-lead ECG that use various ECG signal analyses and/or the inclusion of clinical and laboratory variables. Most improved in terms of sensitivity, but most also failed to outperform specificity compared to the classic ECG criteria. ML algorithms should be compared or tested on the same (standard) database.

World Heart Federation Consensus on Transthyretin Amyloidosis Cardiomyopathy (ATTR-CM)

Transthyretin amyloid cardiomyopathy (ATTR-CM) is a progressive and fatal condition that requires early diagnosis, management, and specific treatment. The availability of new disease-modifying therapies has made successful treatment a reality. Transthyretin amyloid cardiomyopathy can be either age-related (wild-type form) or caused by mutations in the TTR gene (genetic, hereditary forms). It is a systemic disease, and while the genetic forms may exhibit a variety of symptoms, a predominant cardiac phenotype is often present. This document aims to provide an overview of ATTR-CM amyloidosis focusing on cardiac involvement, which is the most critical factor for prognosis. It will discuss the available tools for early diagnosis and patient management, given that specific treatments are more effective in the early stages of the disease, and will highlight the importance of a multidisciplinary approach and of specialized amyloidosis centres. To accomplish these goals, the World Heart Federation assembled a panel of 18 expert clinicians specialized in TTR amyloidosis from 13 countries, along with a representative from the Amyloidosis Alliance, a patient advocacy group. This document is based on a review of published literature, expert opinions, registries data, patients' perspectives, treatment options, and ongoing developments, as well as the progress made possible via the existence of centres of excellence. From the patients' perspective, increasing disease awareness is crucial to achieving an early and accurate diagnosis. Patients also seek to receive care at specialized amyloidosis centres and be fully informed about their treatment and prognosis.

A Machine Learning Framework for Diagnosing and Predicting the Severity of Coronary Artery Disease

BACKGROUND

Although machine learning (ML)-based prediction of coronary artery disease (CAD) has gained increasing attention, assessment of the severity of suspected CAD in symptomatic patients remains challenging.

METHODS

The training set for this study consisted of 284 retrospective participants, while the test set included 116 prospectively enrolled participants from whom we collected 53 baseline variables and coronary angiography results. The data was pre-processed with outlier processing and One-Hot coding. In the first stage, we constructed a ML model that used baseline information to predict the presence of CAD with a dichotomous model. In the second stage, baseline information was used to construct ML regression models for predicting the severity of CAD. The non-CAD population was included, and two different scores were used as output variables. Finally, statistical analysis and SHAP plot visualization methods were employed to explore the relationship between baseline information and CAD.

RESULTS

The study included 269 CAD patients and 131 healthy controls. The eXtreme Gradient Boosting (XGBoost) model exhibited the best performance amongst the different models for predicting CAD, with an area under the receiver operating characteristic curve of 0.728 (95% CI 0.623-0.824). The main correlates were left ventricular ejection fraction, homocysteine, and hemoglobin (p < 0.001). The XGBoost model performed best for predicting the SYNTAX score, with the main correlates being brain natriuretic peptide (BNP), left ventricular ejection fraction, and glycated hemoglobin (p < 0.001). The main relevant features in the model predictive for the GENSINI score were BNP, high density lipoprotein, and homocysteine (p < 0.001).

CONCLUSIONS

This data-driven approach provides a foundation for the risk stratification and severity assessment of CAD.

CLINICAL TRIAL REGISTRATION

The study was registered in www.clinicaltrials.gov protocol registration system (number NCT05018715).

Machine Learning Approaches in Diagnosis, Prognosis and Treatment Selection of Cardiac Amyloidosis

Cardiac amyloidosis is an uncommon restrictive cardiomyopathy featuring an unregulated amyloid protein deposition that impairs organic function. Early cardiac amyloidosis diagnosis is generally delayed by indistinguishable clinical findings of more frequent hypertrophic diseases. Furthermore, amyloidosis is divided into various groups, according to a generally accepted taxonomy, based on the proteins that make up the amyloid deposits; a careful differentiation between the various forms of amyloidosis is necessary to undertake an adequate therapeutic treatment. Thus, cardiac amyloidosis is thought to be underdiagnosed, which delays necessary therapeutic procedures, diminishing quality of life and impairing clinical prognosis. The diagnostic work-up for cardiac amyloidosis begins with the identification of clinical features, electrocardiographic and imaging findings suggestive or compatible with cardiac amyloidosis, and often requires the histological demonstration of amyloid deposition. One approach to overcome the difficulty of an early diagnosis is the use of automated diagnostic algorithms. Machine learning enables the automatic extraction of salient information from “raw data” without the need for pre-processing methods based on the a priori knowledge of the human operator. This review attempts to assess the various diagnostic approaches and artificial intelligence computational techniques in the detection of cardiac amyloidosis.

Estimating the Prevalence of Cardiac Amyloidosis in Old Patients with Heart Failure—Barriers and Opportunities for Improvement: The PREVAMIC Study

Background: Cardiac amyloidosis (CA) could be a common cause of heart failure (HF). The objective of the study was to estimate the prevalence of CA in patients with HF. Methods: Observational, prospective, and multicenter study involving 30 Spanish hospitals. A total of 453 patients ≥ 65 years with HF and an interventricular septum or posterior wall thickness > 12 mm were included. All patients underwent a 99mTc-DPD/PYP/HMDP scintigraphy and monoclonal bands were studied, following the current criteria for non-invasive diagnosis. In inconclusive cases, biopsies were performed. Results: The vast majority of CA were diagnosed non-invasively. The prevalence was 20.1%. Most of the CA were transthyretin (ATTR-CM, 84.6%), with a minority of cardiac light-chain amyloidosis (AL-CM, 2.2%). The remaining (13.2%) was untyped. The prevalence was significantly higher in men (60.1% vs 39.9%, p = 0.019). Of the patients with CA, 26.5% had a left ventricular ejection fraction less than 50%. Conclusions: CA was the cause of HF in one out of five patients and should be screened in the elderly with HF and myocardial thickening, regardless of sex and LVEF. Few transthyretin-gene-sequencing studies were performed in older patients. In many patients, it was not possible to determine the amyloid subtype.

An echo score raises the suspicion of cardiac amyloidosis in Chinese with heart failure with preserved ejection fraction

AIMS

Transthyretin cardiac amyloidosis (ATTR-CA) has been realized as an important cause of heart failure with preserved ejection fraction (HFpEF). We aim to provide insights into its prevalence in Chinese HFpEF patients, which is not known to date, using increased wall thickness (IWT) score by echocardiography.

METHODS

Consecutive patients with HFpEF (EF ≥ 40%) and IWT (≥12 mm) were prospectively screened. Echocardiography was performed, and the IWT score incorporated relative wall thickness, E/e' ratio, longitudinal strains, and tricuspid annular plane systolic excursion, and septal apical-to-base ratio was calculated. ATTR-CA was defined as score ≥8 in the absence of serum and urine free light chain.

RESULTS

Six hundred twenty-four HFpEF patients from January 2019 to December 2021 were enrolled, of which 65.2% were males and the median (interquartile range [IQR]) age was 66 (IQR 57, 73) years. Thirty-three patients (5.3%, 95% CI 3.5-7.0%) were with score ≥8, and 33.3% were females. They were younger (58 vs. 69 years, P < 0.001), had higher NT-proBNP (6525.0 vs. 1741.5 pg/mL, P < 0.001) and troponin I (105.2 vs. 27.7 pg/mL, P = 0.001) level, and lower LVEF (47% vs. 57%, P < 0.001) compared with the patients with score <5. In the internal cohort (82 patients) who had undergone scintigraphy, the IWT score ≥8 was shown to have a sensitivity of 85.7% (95% CI 56.2-97.5%) and a specificity of 92.6% (95% CI 83.0-97.3%) for diagnosing CA, and the IWT score <5 had great accuracy in excluding CA with the negative predictive value of 100%, supporting the clinical usefulness of the IWT score to guide further dedicated testing for ATTR-CA.

CONCLUSIONS

The IWT score by echocardiography was an excellent tool for screening ATTR-CA in HFpEF. In Chinese HFpEF patients associated with a hypertrophic phenotype, the proportion of highly suspected ATTR-CA as detected by IWT score ≥8 was 5.3%, lower than the reported prevalence of ATTR-CA in non-Asian patients with the disease.

Restrictive Atrial Dysfunction in Cardiac Amyloidosis: Differences between Immunoglobulin Light Chain and Transthyretin Cardiac Amyloidosis Patients

Background: In cardiac amyloidosis, the prevalence of thromboembolic events and atrial fibrillation is higher in transthyretin amyloidosis compared to immunoglobulin light chain amyloidosis. Therefore, we hypothesize that transthyretin cardiac amyloidosis patients have worse atrial function. Purpose: To explore the left atrial function by conventional ultrasound and strain analysis in immunoglobulin light chain- and transthyretin cardiac amyloidosis patients. Methods: In cardiac amyloidosis patients in our Amyloidosis Expert Center, echocardiographic strain analysis was performed using speckle tracking. Results: The data of 53 cardiac amyloidosis patients (83% male, mean age 70 years) were analyzed. Transthyretin cardiac amyloidosis patients (n = 24, 45%) were older (75 ± 5.6 vs. 65 ± 7.2 years, p < 0.001) and had more left ventricular (LV) hypertrophy than immunoglobulin light chain cardiac amyloidosis patients (n = 29, 55%). However, LV systolic and diastolic function did not differ, nor did left atrial dimensions (LAVI 56(24) vs. 50(31) mL/m2). Left atrial reservoir strain was markedly lower in transthyretin cardiac amyloidosis (7.4(6.2) vs. 13.6(14.7), p = 0.017). This association was independent of other measurements of the left atrial and ventricular function. Conclusions: Transthyretin cardiac amyloidosis patients had lower left atrial reservoir function compared to immunoglobulin light chain cardiac amyloidosis patients although the left atrial geometry was similar. Interestingly, this association was independent of left atrial- and LV ejection fraction and global longitudinal strain. Further research is warranted to assess the impact of impaired left atrial dysfunction in transthyretin cardiac amyloidosis on atrial fibrillation burden and prognosis.