This work attempts to reduce the number of false alarms generated by bedside monitors in the intensive care unit (ICU), as a majority of current alarms are false. In this study, we applied methods that can be categorized into three stages: signal processing, feature extraction, and optimized machine learning. At the stage of signal processing, we ensured that the heartbeats were properly annotated. During feature extraction, besides extracting features that are relevant to the arrhythmic alarms, we also extracted a set of signal quality indices (SQIs), which we used to distinguish noise/artifact from normal physiological signals. When applying a machine learning algorithm (Random Forest), we performed feature selection in order to reduce the complexity of the models and improve the efficiency of the algorithm. The dataset used is from Reducing False Arrhythmia Alarms in the ICU: the PhysioNet/Computing in Cardiology Challenge 2015. Using the performance metric “score” from the Challenge, we achieved a score of 83.08 in the real-time category on the hidden test set, which is the highest in all published work.

Cardio-respiratory monitoring is one of the most demanding areas in the rapidly growing, mobile-device, based health care delivery. We developed a 12-lead smartphone-based electrocardiogram (ECG) acquisition and monitoring system (called “cvrPhone”), and an application to assess underlying ischemia, and estimate the respiration rate (RR) and tidal volume (TV) from analysis of electrocardiographic (ECG) signals only. During in-vivo swine studies (n = 6), 12-lead ECG signals were recorded at baseline and following coronary artery occlusion. Ischemic indices calculated from each lead showed statistically significant (p < 0.05) increase within 2 min of occlusion compared to baseline. Following myocardial infarction, spontaneous ventricular tachycardia episodes (n = 3) were preceded by significant (p < 0.05) increase of the ischemic index ~1–4 min prior to the onset of the tachy-arrhythmias. In order to assess the respiratory status during apnea, the mechanical ventilator was paused for up to 2 min during normal breathing. We observed that the RR and TV estimation algorithms detected apnea within 7.9 ± 1.1 sec and 5.5 ± 2.2 sec, respectively, while the estimated RR and TV values were 0 breaths/min and less than 100 ml, respectively. In conclusion, the cvrPhone can be used to detect myocardial ischemia and periods of respiratory apnea using a readily available mobile platform.

Chikungunya is a relatively benign disease, and a paucity of literature on severe manifestations in children exits. We describe a cohort of pediatric chikungunya fever patients in New Delhi, India, who had severe sepsis and septic shock, which can develop during the acute phase of illness.

No abstract

Arterial stiffness (AS) has been shown to be an important marker for risk assessment of cardiovascular events. Local arterial stiffness (LAS) is conventionally measured by evaluating arterial distensibility at particular arterial sites through ultrasound imaging systems. Regional arterial stiffness (RAS) is generally obtained by evaluating carotid to femoral pulse wave velocity (cfPWV) through tonometric devices. RAS has a better prognostic value than LAS and cfPWV is considered as the gold standard of AS. Over the past few years our group has been developing ARTerial Stiffness Evaluation for Non-Invasive Screening (ARTSENS), an inexpensive and portable device to measure the LAS. It uses a single element ultrasound transducer to obtain A-Mode frames from the desired artery and is fully automated to enable a non-expert to perform measurements. In this work, we report an extension of ARTSENS to enable measurement of cfPWV that now makes it the only fully automatic device that can measure both LAS and RAS. In this paper, we provide a general review of the ARTSENS and compare it with other state-of-the-art AS measurement systems. cfPWV measurement using ARTSENS was cross-validated against SphygmoCor by successive measurements with both devices on 41 human subjects and excellent agreement between both devices was demonstrated (Coefficient of determination and, limits of agreement m/s). The inter-device correlation between ARTSENS and SphygmoCor was found to be better than other similar studies reported in the literature.

Over past few years our group has been working on the development of a low-cost device, ARTSENS, for measurement of local arterial stiffness (AS) of the common carotid artery (CCA). This uses a single element ultrasound transducer to obtain A-mode frames from the CCA. It is designed to be fully automatic in its operation such that, a general medical practitioner can use the device without any prior knowledge of ultrasound modality. Placement of the probe over CCA and identification of echo positions corresponding to its two walls are critical steps in the process of measurement of AS. We had reported an algorithm to locate the CCA walls based on their characteristic motion. Unfortunately, in supine position, the internal jugular vein (IJV) expands in the carotid triangle and pulsates in a manner that confounds the existing algorithm and leads to wrong measurements of the AS. Jugular venous pulse (JVP), on its own right, is a very important physiological signal for diagnosis of morbidities of the right side of the heart and there is a lack of noninvasive methods for its accurate estimation. We integrated an ECG device to the existing hardware of ARTSENS and developed a method based on physiology of the vessels, which now enable us to segregate the CCA pulse (CCP) and the JVP. False identification rate is less than 4%. To retain the capabilities of ARTSENS to operate without ECG, we designed another method where the classification can be achieved without an ECG, albeit errors are a bit higher. These improvements enable ARTSENS to perform automatic measurement of AS even in the supine position and make it a unique and handy tool to perform JVP analysis.

Evaluation of arterial compliance is very significant in early detection of coronary heart disease. Here we present an imageless portable system for automated estimation of local arterial compliance, designed to be operated by a general medical practitioner with no prior knowledge of ultrasonography. An algorithm for automatic detection and tracking of the arterial wall locations has been developed to minimize the operator expertise required for measurement. The performance of the automated algorithm was thoroughly characterized using a simulation platform developed for the purpose. Measurements performed on a few human volunteers by untrained personnel clearly illustrated the practical utility of the automatic algorithm during in-vivo tests. The proposed system could be used for developing an inexpensive cardiovascular screening device for large scale deployment in primary health care centers.

Over past few years, we are developing a system for facilitating large scale screening of patients for cardiovascular risk—arterial stiffness evaluation for noninvasive screening (ARTSENS). ARTSENS is an image-free device that uses a single element ultrasound transducer to obtain noninvasive measurements of arterial stiffness (AS) in a fully automated manner. AS is directly proportional to end-diastolic lumen diameter (Dd). Multilayered structure of the arterial walls and indistinct characteristics of intima-lumen interface (ILI) makes it quite difficult to accurately estimate Dd in A-mode radio-frequency (RF) frames obtained from ARTSENS. In this paper, we propose a few methods based on fitting simple mathematical models to the echoes from arterial walls, followed by a novel method to fuse the information from curve fitting error and distension curve to arrive at an accurate measure of Dd. To bring down the curve fitting time and facilitate processing on low-end processors, a novel approach using the autocorrelation of echoes from opposite walls of the artery has been discussed. The methods were analyzed for their comparative accuracy against reference Dd obtained from 85 human volunteers using Hitachi-Aloka eTRACKING system. Dd from all reported methods show strong and statistically significant positive correlation with eTRACKING and mean error of less than 7% could be achieved. As expected, Dd from all methods show significant positive correlation with age.