Evaluating explainability for graph neural networks.

As explanations are increasingly used to understand the behavior of graph neural networks (GNNs), evaluating the quality and reliability of GNN explanations is crucial. However, assessing the quality of GNN explanations is challenging as existing graph datasets have no or unreliable ground-truth explanations. Here, we introduce a synthetic graph data generator, SHAPEGGEN, which can generate a variety of benchmark datasets (e.g., varying graph sizes, degree distributions, homophilic vs. heterophilic graphs) accompanied by ground-truth explanations. The flexibility to generate diverse synthetic datasets and corresponding ground-truth explanations allows SHAPEGGEN to mimic the data in various real-world areas. We include SHAPEGGEN and several real-world graph datasets in a graph explainability library, GRAPHXAI. In addition to synthetic and real-world graph datasets with ground-truth explanations, GRAPHXAI provides data loaders, data processing functions, visualizers, GNN model implementations, and evaluation metrics to benchmark GNN explainability methods.

Enhancing convolutional neural network predictions of electrocardiograms with left ventricular dysfunction using a novel sub-waveform representation.

Background: Electrocardiogram (ECG) deep learning (DL) has promise to improve the outcomes of patients with cardiovascular abnormalities. In ECG DL, researchers often use convolutional neural networks (CNNs) and traditionally use the full duration of raw ECG waveforms that create redundancies in feature learning and result in inaccurate predictions with large uncertainties. Objective: For enhancing these predictions, we introduced a sub-waveform representation that leverages the rhythmic pattern of ECG waveforms (data-centric approach) rather than changing the CNN architecture (model-centric approach). Results: We applied the proposed representation to a population with 92,446 patients to identify left ventricular dysfunction. We found that the sub-waveform representation increases the performance metrics compared to the full-waveform representation. We observed a 2% increase for area under the receiver operating characteristic curve and 10% increase for area under the precision-recall curve. We also carefully examined three reliability components of explainability, interpretability, and fairness. We provided an explanation for enhancements obtained by heartbeat alignment mechanism. By developing a new scoring system, we interpreted the clinical relevance of ECG features and showed that sub-waveform representation further pushes the scores towards clinical predictions. Finally, we showed that the new representation significantly reduces prediction uncertainties within subgroups that contributes to individual fairness. Conclusion: We expect that this added control over the granularity of ECG data will improve the DL modeling for new artificial intelligence technologies in the cardiovascular space.

Deep Learning Analyses of Brain MRI to Identify Sustained Attention Deficit in Treated Obstructive Sleep Apnea: A Pilot Study.

Purpose: Persistent sustained attention deficit (SAD) after continuous positive airway pressure (CPAP) treatment is a source of quality of life and occupational impairment in obstructive sleep apnea (OSA). However, persistent SAD is difficult to predict in patients initiated on CPAP treatment. We performed secondary analyses of brain magnetic resonance (MR) images in treated OSA participants, using deep learning, to predict SAD. Methods: 26 middle-aged men with CPAP use of more than 6 hours daily and MR imaging were included. SAD was defined by psychomotor vigilance task lapses of more than 2. 17 participants had SAD and 9 were without SAD. A Convolutional Neural Network (CNN) model was used for classifying the MR images into +SAD and -SAD categories. Results: The CNN model achieved an accuracy of 97.02±0.80% in classifying MR images into +SAD and -SAD categories. Assuming a threshold of 90% probability for the MR image being correctly classified, the model provided a participant-level accuracy of 99.11±0.55% and a stable image level accuracy of 97.45±0.63%. Conclusion: Deep learning methods, such as the proposed CNN model, can accurately predict persistent SAD based on MR images. Further replication of these findings will allow early initiation of adjunctive pharmacologic treatment in high-risk patients, along with CPAP, to improve quality of life and occupational fitness. Future augmentation of this approach with explainable artificial intelligence methods may elucidate the neuroanatomical areas underlying persistent SAD to provide mechanistic insights and novel therapeutic targets.

Outcomes in Patients With COVID-19 Disease and High Oxygen Requirements.

BACKGROUND: Approximately 19% of people infected with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) progress to severe or critical stages of the coronavirus disease 2019 (COVID-19) with a mortality rate exceeding 50%. We aimed to examine the characteristics, mortality rates, intubation rate, and length of stay (LOS) of patients hospitalized with COVID-19 disease with high oxygen requirements (critically ill). METHODS: We conducted a retrospective analysis in a single center in Brooklyn, New York. Adult hospitalized patients with confirmed COVID-19 disease and high oxygen requirements were included. We performed multivariate logistic regression analyses for statistically significant variables to reduce any confounding. RESULTS: A total of 398 patients were identified between March 19th and April 25th, 2020 who met the inclusion criteria, of which 247 (62.1%) required intubation. The overall mortality rate in our study was 57.3% (n = 228). The mean hospital LOS was 19.1 ± 17.4 days. Patients who survived to hospital discharge had a longer mean LOS compared to those who died during hospitalization (25.4 ± 22.03 days versus10.7 ± 1.74 days). In the multivariate analysis, increased age, intubation and increased lactate dehydrogenase (LDH) were each independently associated with increased odds of mortality. Diarrhea was associated with decreased mortality (OR 0.4; CI 0.16, 0.99). Obesity and use of vasopressors were each independently associated with increased intubation. CONCLUSIONS: In patients with COVID-19 disease and high oxygen requirements, advanced age, intubation, and higher LDH levels were associated with increased mortality, while diarrhea was associated with decreased mortality. Gender, diabetes, and hypertension did not have any association with mortality or length of hospital stay.

Deep learning applied to polysomnography to predict blood pressure in obstructive sleep apnea and obesity hypoventilation: a proof-of-concept study.

STUDY OBJECTIVES: Nocturnal blood pressure (BP) profile shows characteristic abnormalities in OSA, namely acute postapnea BP surges and nondipping BP. These abnormal BP profiles provide prognostic clues indicating increased cardiovascular disease risk. We developed a deep neural network model to perform computerized analysis of polysomnography data and predict nocturnal BP profile. METHODS: We analyzed concurrently performed polysomnography and noninvasive beat-to-beat BP measurement with a deep neural network model to predict nocturnal BP profiles from polysomnography data in 13 patients with severe OSA. RESULTS: A good correlation was noted between measured and predicted postapnea systolic and diastolic BP (Pearson r ≥ .75). Moreover, Bland-Altman analyses showed good agreement between the 2 values. Continuous systolic and diastolic BP prediction by the deep neural network model was also well correlated with measured BP values (Pearson r ≥ .83). CONCLUSIONS: We developed a deep neural network model to predict nocturnal BP profile from clinical polysomnography signals and provide a potential prognostic tool in OSA. Validation of the model in larger samples and examination of its utility in predicting CVD risk in future studies is warranted.

Using Adipose Measures from Health Care Provider-Based Imaging Data for Discovery.

The location and type of adipose tissue is an important factor in metabolic syndrome. A database of picture archiving and communication system (PACS) derived abdominal computerized tomography (CT) images from a large health care provider, Geisinger, was used for large-scale research of the relationship of volume of subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) with obesity-related diseases and clinical laboratory measures. Using a "greedy snake" algorithm and 2,545 CT images from the Geisinger PACS, we measured levels of VAT, SAT, total adipose tissue (TAT), and adipose ratio volumes. Sex-combined and sex-stratified association testing was done between adipose measures and 1,233 disease diagnoses and 37 clinical laboratory measures. A genome-wide association study (GWAS) for adipose measures was also performed. SAT was strongly associated with obesity and morbid obesity. VAT levels were strongly associated with type 2 diabetes-related diagnoses (p = 1.5 × 10-58), obstructive sleep apnea (p = 7.7 × 10-37), high-density lipoprotein (HDL) levels (p = 1.42 × 10-36), triglyceride levels (p = 1.44 × 10-43), and white blood cell (WBC) counts (p = 7.37 × 10-9). Sex-stratified tests revealed stronger associations among women, indicating the increased influence of VAT on obesity-related disease outcomes particularly among women. The GWAS identified some suggestive associations. This study supports the utility of pursuing future clinical and genetic discoveries with existing imaging data-derived adipose tissue measures deployed at a larger scale.

Clinical Application of Biomarkers in Heart Failure with a Preserved Ejection Fraction: A Review.

Heart failure with a preserved ejection fraction (HFpEF) is increasingly prevalent and a leading cause of morbidity and mortality worldwide. HFpEF has a complex pathophysiology, with recent evidence suggesting that an interaction of cardiovascular and noncardiovascular comorbidities (e.g. obesity, hypertension, diabetes, coronary artery disease, and chronic kidney disease) induces an inflammatory state that eventually leads to myocardial structural and functional alterations. Current ACCF/AHA guidelines suggest incorporation of biomarkers along with clinical and imaging tools to establish the diagnosis and disease severity in heart failure (HF). However, the majority of data on biomarkers relating to their levels, or their role in accurate diagnosis, prognostication, and disease activity, has been derived from studies in undifferentiated HF or HF with a reduced EF (HFrEF). As the understanding of the mechanisms underlying HFpEF continues to evolve, biomarkers reflecting different pathways including neurohormonal activation, myocardial injury, inflammation, and fibrosis have a clinical utility beyond the diagnostic scope. Accordingly, in this review article we describe the various established and novel plasma biomarkers and their emerging value in diagnosis, prognosis, response, and guiding of targeted therapy in patients with HFpEF.

Carcinoid Heart Disease without Liver Involvement Caused by a Primary Ovarian Carcinoid Tumour.

Carcinoid heart disease, caused by primary ovarian carcinoid tumour, is a rare form of valvular heart disease. This form of heart disease usually presents with symptoms of right-sided valvular dysfunction, ultimately leading to right-sided heart failure. This entity is unique as it develops in the absence of liver metastasis. We report a case of 75 year-old woman with primary ovarian carcinoid tumour who presented with symptoms of severe right-sided heart failure and successfully underwent pulmonic and tricuspid valve replacement along with a right ventricular (RV) outflow patch enlargement. This patient later underwent uneventful resection of the primary ovarian carcinoid tumour, with complete resolution of her symptoms.

CT imaging of post-myocardial infarction ventricular septal defect with a contained rupture/pseudoaneurysm.

This is a CT imaging study of a 63-year-old female who presented to our center with ST segment elevation MI and was found to have life threatening post-MI ventricular septal defect with associated pseudoaneurysm, which was detected on cardiac CTA. The patient refused surgical management and had a successful percutaneous VSD repair.

Imaging Modalities to Identity Inflammation in an Atherosclerotic Plaque.

Atherosclerosis is a chronic, progressive, multifocal arterial wall disease caused by local and systemic inflammation responsible for major cardiovascular complications such as myocardial infarction and stroke. With the recent understanding that vulnerable plaque erosion and rupture, with subsequent thrombosis, rather than luminal stenosis, is the underlying cause of acute ischemic events, there has been a shift of focus to understand the mechanisms that make an atherosclerotic plaque unstable or vulnerable to rupture. The presence of inflammation in the atherosclerotic plaque has been considered as one of the initial events which convert a stable plaque into an unstable and vulnerable plaque. This paper systemically reviews the noninvasive and invasive imaging modalities that are currently available to detect this inflammatory process, at least in the intermediate stages, and discusses the ongoing studies that will help us to better understand and identify it at the molecular level.