Neural network methods for diagnosing patient conditions from cardiopulmonary exercise testing data

Transitioning from stress electrocardiogram to cardiopulmonary exercise testing: a paradigm shift toward comprehensive medical evaluation of exercise function

Cardiopulmonary exercise testing (CPET) has emerged as a powerful diagnostic tool, providing comprehensive physiological insights into the integrated function of cardiovascular, respiratory, and metabolic systems. Exploiting physiological interactions, CPET allows in-depth diagnostic insights. CPET performance entrains several complexities. Interpreting CPET data can be challenging, requiring significant physiological expertise. The advent of artificial intelligence (AI) has introduced a transformative approach to CPET interpretation, enhancing accuracy, efficiency, and clinical decision-making. This review article explores the current state of AI applications in CPET, highlighting AI's potential to replace the traditional stress electrocardiogram (ECG) test as the preferred diagnostic tool in preventive medicine and medical screening. The article discusses the underlying principles of AI, its integration into CPET interpretation, and the associated benefits, including improved diagnostic accuracy, reduced interobserver variability, and expedited decision-making. Additionally, it addresses the challenges and considerations surrounding the implementation of AI in CPET such as data quality, model interpretability, and ethical concerns. The review concludes by emphasizing the significant promise of AI-assisted CPET interpretation in revolutionizing preventive medicine and medical screening settings and enhancing patient care.

MDMNet: Multi-dimensional multi-modal network to identify organ system limitation in cardiopulmonary exercise testing

BACKGROUND AND OBJECTIVE

Cardiopulmonary exercise testing (CPET) serves as an integrative and comprehensive assessment tool for cardiorespiratory fitness. In this paper, we present a novel multi-dimensional multi-modal network (MDMNet) to identify functional limitation of organ systems via CPET, which is of great importance in clinical practice and yet a challenging task due to (1) the intricate intra-variable associations, and (2) the significant inter-individual variability.

METHODS

The proposed model has three compelling characteristics. First, we employ a dedicated embedding strategy for CPET data to map raw inputs into the learned embedding space, facilitating the detection of latent features of physiological variables. Second, we devise a novel multi-dimensional feature extraction module to capture rich features of physiological inputs at different dimensions, which consists of a one-dimensional feature extraction branch unfolding both temporal and spatial patterns of the entire data, and a two-dimensional feature extraction branch based on Gramian Angular Field (GAF) encoding to reveal the complicated temporal correlation relationships between time points within a variable. Third, we integrate these techniques with clinically significant demographic information to establish our MDMNet incorporating multi-dimensional with multi-modal learning, thereby further addressing the issues of complex intra-variable associations and inter-individual variability simultaneously.

RESULTS

We evaluated the proposed method on the publicly available CPET dataset, achieving AUC scores of 0.948, 0.949 and 0.931 for three tasks respectively.

CONCLUSIONS

The superiority of our method in discerning inter-individual differences was further demonstrated through partial least squares discriminant analysis, which holds significant potential for automated clinical application of CPET.

Machine learning models for reinjury risk prediction using cardiopulmonary exercise testing (CPET) data: optimizing athlete recovery

BACKGROUND

Cardiopulmonary Exercise Testing (CPET) provides detailed insights into athletes' cardiovascular and pulmonary function, making it a valuable tool in assessing recovery and injury risks. However, traditional statistical models often fail to leverage the full potential of CPET data in predicting reinjury. Machine learning (ML) algorithms offer promising capabilities in uncovering complex patterns within this data, allowing for more accurate injury risk assessment.

OBJECTIVE

This study aimed to develop machine learning models to predict reinjury risk among elite soccer players using CPET data. Specifically, we sought to identify key physiological and performance variables that correlate with reinjury and to evaluate the performance of various ML algorithms in generating accurate predictions.

METHODS

A dataset of 256 elite soccer players from 16 national and top-tier teams in Iran was analyzed, incorporating physiological variables and categorical data. Several machine learning models, including CatBoost, SVM, Random Forest, and XGBoost, were employed to predict reinjury risk. Model performance was assessed using metrics such as accuracy, precision, recall, F1-score, AUC, and SHAP values to ensure robust evaluation and interpretability.

RESULTS

CatBoost and SVM exhibited the best performance, with CatBoost achieving the highest accuracy (0.9138) and F1-score (0.9148), and SVM achieving the highest AUC (0.9725). A significant association was found between a history of concussion and reinjury risk (χ² = 13.0360, p = 0.0015), highlighting the importance of neurological recovery in preventing future injuries. Heart rate metrics, particularly HRmax and HR2, were also significantly lower in players who experienced reinjury, indicating reduced cardiovascular capacity in this group.

CONCLUSION

Machine learning models, particularly CatBoost and SVM, provide promising tools for predicting reinjury risk using CPET data. These models offer clinicians more precise, data-driven insights into athlete recovery and risk management. Future research should explore the integration of external factors such as training load and psychological readiness to further refine these predictions and enhance injury prevention protocols.

Improving cardiovascular risk stratification through multivariate time-series analysis of cardiopulmonary exercise test data

Nowadays cardiorespiratory fitness (CRF) is assessed using summary indexes of cardiopulmonary exercise tests (CPETs). Yet, raw time-series CPET recordings may hold additional information with clinical relevance. Therefore, we investigated whether analysis of raw CPET data using dynamic time warping combined with k-medoids could identify distinct CRF phenogroups and improve cardiovascular (CV) risk stratification. CPET recordings from 1,399 participants (mean age, 56.4 years; 37.7% women) were separated into 5 groups with distinct patterns. Cluster 5 was associated with the worst CV profile with higher use of antihypertensive medication and a history of CV disease, while cluster 1 represented the most favorable CV profile. Clusters 4 (hazard ratio: 1.30; p = 0.033) and 5 (hazard ratio: 1.36; p = 0.0088) had a significantly higher risk of incident adverse events compared to clusters 1 and 2. The model evaluation in the external validation cohort revealed similar patterns. Therefore, an integrative CRF profiling might facilitate CV risk stratification and management.

Recent advancements and applications of deep learning in heart failure: Α systematic review

BACKGROUND

Heart failure (HF), a global health challenge, requires innovative diagnostic and management approaches. The rapid evolution of deep learning (DL) in healthcare necessitates a comprehensive review to evaluate these developments and their potential to enhance HF evaluation, aligning clinical practices with technological advancements.

OBJECTIVE

This review aims to systematically explore the contributions of DL technologies in the assessment of HF, focusing on their potential to improve diagnostic accuracy, personalize treatment strategies, and address the impact of comorbidities.

METHODS

A thorough literature search was conducted across four major electronic databases: PubMed, Scopus, Web of Science and IEEE Xplore, yielding 137 articles that were subsequently categorized into five primary application areas: cardiovascular disease (CVD) classification, HF detection, image analysis, risk assessment, and other clinical analyses. The selection criteria focused on studies utilizing DL algorithms for HF assessment, not limited to HF detection but extending to any attempt in analyzing and interpreting HF-related data.

RESULTS

The analysis revealed a notable emphasis on CVD classification and HF detection, with DL algorithms showing significant promise in distinguishing between affected individuals and healthy subjects. Furthermore, the review highlights DL's capacity to identify underlying cardiomyopathies and other comorbidities, underscoring its utility in refining diagnostic processes and tailoring treatment plans to individual patient needs.

CONCLUSIONS

This review establishes DL as a key innovation in HF management, highlighting its role in advancing diagnostic accuracy and personalized care. The insights provided advocate for the integration of DL in clinical settings and suggest directions for future research to enhance patient outcomes in HF care.

Universal representation learning for multivariate time series using the instance-level and cluster-level supervised contrastive learning

The multivariate time series classification (MTSC) task aims to predict a class label for a given time series. Recently, modern deep learning-based approaches have achieved promising performance over traditional methods for MTSC tasks. The success of these approaches relies on access to the massive amount of labeled data (i.e., annotating or assigning tags to each sample that shows its corresponding category). However, obtaining a massive amount of labeled data is usually very time-consuming and expensive in many real-world applications such as medicine, because it requires domain experts' knowledge to annotate data. Insufficient labeled data prevents these models from learning discriminative features, resulting in poor margins that reduce generalization performance. To address this challenge, we propose a novel approach: supervised contrastive learning for time series classification (SupCon-TSC). This approach improves the classification performance by learning the discriminative low-dimensional representations of multivariate time series, and its end-to-end structure allows for interpretable outcomes. It is based on supervised contrastive (SupCon) loss to learn the inherent structure of multivariate time series. First, two separate augmentation families, including strong and weak augmentation methods, are utilized to generate augmented data for the source and target networks, respectively. Second, we propose the instance-level, and cluster-level SupCon learning approaches to capture contextual information to learn the discriminative and universal representation for multivariate time series datasets. In the instance-level SupCon learning approach, for each given anchor instance that comes from the source network, the low-variance output encodings from the target network are sampled as positive and negative instances based on their labels. However, the cluster-level approach is performed between each instance and cluster centers among batches, as opposed to the instance-level approach. The cluster-level SupCon loss attempts to maximize the similarities between each instance and cluster centers among batches. We tested this novel approach on two small cardiopulmonary exercise testing (CPET) datasets and the real-world UEA Multivariate time series archive. The results of the SupCon-TSC model on CPET datasets indicate its capability to learn more discriminative features than existing approaches in situations where the size of the dataset is small. Moreover, the results on the UEA archive show that training a classifier on top of the universal representation features learned by our proposed method outperforms the state-of-the-art approaches.

Genetic features of bovine viral diarrhea virus subgenotype 1c in newborn calves at nucleotide and synonymous codon usages

Bovine viral diarrhea virus (BVDV), serving as an important pathogen for newborn calves, poses threat to reproductive and economic losses in the cattle industry. To survey the infection rate and genetic diversity of BVDV in newborn calves in northern China, a total of 676 sera samples of newborn calves were collected from four provinces between 2021 and 2022. All sera samples were individually detected for BVDV infection by RT-PCR and ELISA. Our results showed that the overall serological rate was 9.76% (66/676) and the average positive rate of BVDV RNA was 8.14% (55/676) in the newborn calves. Eight BVDV strains were successfully isolated from RT-PCR positive sera samples, and four isolates displayed the cytopathic effect (CPE). Based on phylogenetic tree at the genome level, the eight strains were classified into subgenotype 1c. Moreover, the BVDV isolates had a close genetic relationship with the GSTZ strain at either nucleotide or codon usage level. Interestingly, in comparison of synonymous codon usage patterns between the BVDV isolates with CPE and ones without CPE, there were four synonymous codons (UCG, CCC, GCA, and AAC) which displayed the significant differences (p < 0.05) at codon usage pattern, suggesting that synonymous codon usage bias might play a role in BVDV-1c biotypes. In addition, the usage of synonymous codons containing CpG dinucleotides was suppressed by the BVDV-1c isolates, reflecting one of strategies of immune evasion of BVDV to its host. Taken together, our study provided data for monitoring and vaccination strategies of BVDV for newborn calves in northern China.