Arunachalam, S P

The analysis and interpretation of cardiac magnetic resonance (CMR) images are often time-consuming. The automated segmentation of cardiac structures can reduce the time required for image analysis. Spatial similarities between different CMR image types were leveraged to jointly segment multiple sequences using a segmentation model termed a multi-image type UNet (MI-UNet). This model was developed from 72 exams (46% female, mean age 63 ± 11 years) performed on patients with hypertrophic cardiomyopathy. The MI-UNet for steady-state free precession (SSFP) images achieved a superior Dice similarity coefficient (DSC) of 0.92 ± 0.06 compared to 0.87 ± 0.08 for a single-image type UNet (p < 0.001). The MI-UNet for late gadolinium enhancement (LGE) images also had a superior DSC of 0.86 ± 0.11 compared to 0.78 ± 0.11 for a single-image type UNet (p = 0.001). The difference across image types was most evident for the left ventricular myocardium in SSFP images and for both the left ventricular cavity and the left ventricular myocardium in LGE images. For the right ventricle, there were no differences in DCS when comparing the MI-UNet with single-image type UNets. The joint segmentation of multiple image types increases segmentation accuracy for CMR images of the left ventricle compared to single-image models. In clinical practice, the MI-UNet model may expedite the analysis and interpretation of CMR images of multiple types.

Abstract Deep vein thrombosis (DVT) is the formation of thrombosis or blood clot in the deep veins of the body, usually in the lower extremities or pelvic vein. During this time, a hypercoagulable state results in a higher rate of deep vein thrombosis (DVT) and pulmonary embolism (PE) during gestation and the postnatal stage. PE is the leading cause of maternal death, making early diagnosis and clinical treatment vital to both the mother’s and fetal life. The current technologies for the diagnosis of DVT reduced detection accuracy with increased depth, false negatives when parallel veins are present, and the incompetence of imaging due to factors such as obesity and edema. The drawback of compression ultrasound is that clots may emboli during diagnosis and travel to vital organs such as the heart. In this study a new technological advancement is explored to superior diagnostic methods, and a near infrared device will be able to provide relative measurements of the oxygenation levels of calf veins and can help identify excessive deoxygenation and specific locations. Oxyhemoglobin & deoxyhemoglobin absorbs red light and infra-red lights differently. Oxygenated blood gets absorbed by IR light while deoxygenated gets transmitted. When infrared light is emitted onto a vein, it gets absorbed by the oxygenated hemoglobin and gives an absorbance ratio of output to the input close to 1. Infrared radiation spectroscopy measurements can indicate Δ½HbO2 and Δ½Hb to show a comparison, helping to indicate the presence or absence of a clot. . The circuit is designed with capacitors, resistors, and transistors to act as filters and triggers to pulsate at 20Hz, to pick up biological signals. An Arduino Uno microcontroller helped to process data in order to analyze signal proximity to 0 or 1 to classify DVT and validate the efficacy of using IR for DVT detection Testing of the device allowed for a functionality check to understand whether it was able to pick up signals of oxygenation levels. The 3 tests have insight on the potential of the device; however, the results were inconclusive, due to a lack of sufficient testing. The limitation has been the inability to test on a large sample size and insufficient data, therefore it can’t be said for now whether IR is an effective way of diagnosing DVT. However, since on localized testing, the device seems to be gathering programmed signals, the research can be furthered, and the efficacy can be proved by testing on expecting patients.

Abstract Hypertrophic Cardiomyopathy (HCM) is the most common genetic heart disease in the US and is known to cause sudden death (SCD) in young adults. While significant advancements have been made in HCM diagnosis and management, there is a need to identify HCM cases from electronic health record (EHR) data to develop automated tools based on natural language processing guided machine learning (ML) models for accurate HCM case identification to improve management and reduce adverse outcomes of HCM patients. Cardiac Magnetic Resonance (CMR) Imaging, plays a significant role in HCM diagnosis and risk stratification. CMR reports, generated by clinician annotation, offer rich data in the form of cardiac measurements as well as narratives describing interpretation and phenotypic description. The purpose of this study is to develop an NLP-based interpretable model utilizing impressions extracted from CMR reports to automatically identify HCM patients. CMR reports of patients with suspected HCM diagnosis between the years 1995 to 2019 were used in this study. Patients were classified into three categories of yes HCM, no HCM and, possible HCM. A random forest (RF) model was developed to predict the performance of both CMR measurements and impression features to identify HCM patients. The RF model yielded an accuracy of 86% (608 features) and 85% (30 features). These results offer promise for accurate identification of HCM patients using CMR reports from EHR for efficient clinical management transforming health care delivery for these patients.

Abstract Hypertrophic cardiomyopathy (HCM) is a heritable, phenotypically diverse, and often asymptomatic heart muscle disease which is a major cause of sudden cardiac death (SCD) in young adults. The gold-standard for the diagnosis of HCM is echocardiography (echo), which is an ultrasound-based cardiac imaging modality. Across all sites of the Mayo Clinic enterprise, echo images and measurement data are reviewed, interpreted, and reported via the Echo Information Management System (EIMS). The objective of this paper is to develop a machine learning model for the identification of HCM from cardiac measurements obtained by the echo. We developed a novel machine learning model on patient demographic information and echo measurements that were retrieved from the EIMS digital data registry and selected by cardiologists. Random forest (RF) was utilized to investigate the predictive performance of these features on the identification of HCM patients. The HCM cohort consists of 3,548 patients with at least one HCM diagnostic billing code (ICD-9 or ICD-10), from 2014 to 2019. The class labels “HCM yes” and “HCM no” were assigned by manual review of medical records as well as the outcomes of the gold standard imaging tests for HCM diagnosis. The developed model performed well in finding HCM patients with an accuracy of 95%, recall of 99%, and precision of 97%. The F1 score was 98 %, while 4% of patients were misclassified. This model will be translated into clinical practice for a clinical decision support system in EIMS to assist providers in the accurate diagnosis of HCM from echo data automatically while ensuring high-quality echo interpretation.

Background: Heart failure with preserved ejection fraction is increasing in prevalence, but few effective treatments are available. Elevated left ventricular (LV) diastolic filling pressures represent a key therapeutic target. Pericardial restraint contributes to elevated LV end-diastolic pressure, and acute studies have shown that pericardiotomy attenuates the rise in LV end-diastolic pressure with volume loading. However, whether these acute effects are sustained chronically remains unknown. Methods: Minimally invasive pericardiotomy was performed percutaneously using a novel device in a porcine model of heart failure with preserved ejection fraction. Hemodynamics were assessed at baseline and following volume loading with pericardium intact, acutely following pericardiotomy, and then again chronically after 4 weeks. Cardiac structure was assessed by magnetic resonance imaging. Results: The increase in LV end-diastolic pressure with volume loading was mitigated by 41% (95% CI, 27%–45%, P <0.0001; ΔLV end-diastolic pressure reduced from +9±3 mm Hg to +5±3 mm Hg, P =0.0003, 95% CI, −2.2 to −5.5). The effect was sustained at 4 weeks (+5±2 mm Hg, P =0.28 versus acute). There was no statistically significant effect of pericardiotomy on ventricular remodeling compared with age-matched controls. None of the animals developed hemodynamic or pathological indicators of pericardial constriction or frank systolic dysfunction. Conclusions: The acute hemodynamic benefits of pericardiotomy are sustained for at least 4 weeks in a swine model of heart failure with preserved ejection fraction, without excessive chamber remodeling, pericarditis, or clinically significant systolic dysfunction. These data support trials evaluating minimally invasive pericardiotomy as a novel treatment for heart failure with preserved ejection fraction in humans.

Abstract Phonocardiogram (PCG) signals are electrical recording of heart sounds containing vital information of diagnostic importance. Several signal processing methods exist to characterize PCG, however suffers in terms of sensitivity and specificity in accurately discriminating normal and abnormal heart sounds. Recently, a multiscale frequency (MSF) analysis of normal PCG was reported to characterize subtle frequency content changes in PCG which can aid in differentiating normal and abnormal heart sounds. In this work, it was hypothesized that MSF can discriminate normal PCG signal compared to an artifact, PCG with extra systolic heart sounds and murmur based on their varying frequency content. Various samples of PCG with normal and abnormal heart sounds were obtained from Peter Bentley Heart Sounds Database sampled at 44.1 kHz for analysis. The signal was filtered using a 4th order Butterworth lowpass filter with cutoff frequency at 200 Hz to remove higher frequency noise and MSF estimation was performed on the filtered dataset using custom MATLAB software. Mann-Whitney test was performed for statistical significance at p &lt; 0.05. Results indicate that MSF successfully discriminated normal and abnormal heart sounds, which can aid in PCG classification with more sophisticated analysis. Validation of this technique with larger dataset is required.