Mehrdad Nourani

A Patient-Adaptive Profiling Scheme for ECG Beat Classification

Recent trends in clinical and telemedicine applications highly demand automation in electrocardiogram (ECG) signal processing and heart beat classification. A patient-adaptive cardiac profiling scheme using repetition-detection concept is proposed in this paper. We first employ an efficient wavelet-based beat-detection mechanism to extract precise fiducial ECG points. Then, we implement a novel local ECG beat classifier to profile each patients normal cardiac behavior. ECG morphologies vary from person to person and even for each person, it can vary over time depending on the persons physical condition and/or environment. Having such profile is essential for various diagnosis (e.g., arrhythmia) purposes. One application of such profiling scheme is to automatically raise an early warning flag for the abnormal cardiac behavior of any individual. Our extensive experimental results on the MIT-BIH arrhythmia database show that our technique can detect the beats with 99.59% accuracy and can identify abnormalities with a high classification accuracy of 97.42%.

Bed posture classification for pressure ulcer prevention

Pressure ulcer is an age-old problem imposing a huge cost to our health care system. Detecting and keeping record of the patients posture on bed, help care givers reposition patient more efficiently and reduce the risk of developing pressure ulcer. In this paper, a commercial pressure mapping system is used to create a time-stamped, whole-body pressure map of the patient. An image-based processing algorithm is developed to keep an unobtrusive and informative record of patients bed posture over time. The experimental results show that proposed algorithm can predict patients bed posture with up to 97.7% average accuracy. This algorithm could ultimately be used with current support surface technologies to reduce the risk of ulcer development.

Adaptive cancellation of motion artifact in wearable biosensors

The performance of wearable biosensors is highly influenced by motion artifact. In this paper, a model is proposed for analysis of motion artifact in wearable photoplethysmography (PPG) sensors. Using this model, we proposed a robust real-time technique to estimate fundamental frequency and generate a noise reference signal. A Least Mean Square (LMS) adaptive noise canceler is then designed and validated using our synthetic noise generator. The analysis and results on proposed technique for noise cancellation shows promising performance.

Wavelet-based denoising and beat detection of ECG signal

This paper presents the design and implementation of an automatic ECG beat detection system. We proposed modifications to the existing Pan-Tompkins algorithm by introducing only one set of adaptive threshold computations to reduce the amount of data processing significantly. LabVIEW signal processing tools were used to test the performance of wavelet based analysis for denoising and feature extraction of the ECG signal. Our design achieved an overall accuracy of 99.51% when applied on the MIT/BIH Arrhythmia Database, which is far better than the old method of digital filtering.

Real-time sleep quality assessment using single-lead ECG and multi-stage SVM classifier

Sleep efficiency measures provide an objective assessment to gauge the quality of individuals sleep. In this study we present a home-based, automated and non-intrusive system that is based on Electrocardiogram (ECG) measurements and uses a multi-stage Support Vector Machines (SVM) classifier to measure three indices for sleep quality assessment per 30 s epoch segment: Sleep Efficiency Index, Delta-Sleep Efficiency Index and Sleep Onset Latency. This method provides an alternative to the intrusive and expensive Polysomnography (PSG) and scoring by Rechtschaffen and Kales visual method.

A smart bed platform for monitoring & Ulcer prevention

The focus of this paper is to develop a software-hardware platform that addresses one of the most costly, acute health conditions, pressure ulcers — or bed sores. Caring for pressure ulcers is extremely costly, increases the length of hospital stays and is very labor intensive. The proposed platform collects information from various sensors incorporated into the bed, analyzes the data to create a time-stamped, whole-body pressure distribution map, and commands the beds actuators to periodically adjust its surface profile to redistribute pressure over the entire body. These capabilities are combined to form a cognitive support system, that augments the ability of a care giver, allowing them to provide better care to more patients in less time. For proof of concept, we have implemented algorithms and architectures that cover four key aspects of this platform: 1) data collection, 2) modeling & profiling, 3) machine learning, and 4) acting.

A Scalable Wireless Body Area Network for Bio-Telemetry

In this paper, we propose a framework for the real-time monitoring of wireless biosensors. This is a scalable platform that requires minimum human interaction during set-up and monitoring. Its main components include a biosensor, a smart gateway to automatically set up the body area network, a mechanism for delivering data to an Internet monitoring server, and automatic data collection, profiling and feature extraction from bio-potentials. Such a system could increase the quality of life and significantly lower healthcare costs for everyone in general, and for the elderly and those with disabilities in particular.

Transient ST-segment episode detection for ECG beat classification

Sudden Cardiac Death (SCD) is an unexpected death caused by loss of heart function when the electrical impulses fired from the ventricles become irregular. Most common SCDs are caused by cardiac arrhythmias and coronary heart disease. They are mainly due to Acute Myocardial Infarction (AMI), myocardial ischaemia and cardiac arrhythmia. This paper aims at automating the recognition of ST-segment deviations and transient ST episodes which helps in the diagnosis of myocardial ischaemia and also classifying major cardiac arrhythmia. Our approach is based on the application of signal processing and artificial intelligence to the heart signal known as the ECG (Electrocardiogram). We propose an improved morphological feature vector including ST-segment information for heart beat classification by supervised learning using the support vector machine approach. Our system has been tested and yielded an accuracy of 93.33% for the ST episode detection on the European ST-T Database and 96.35% on MIT-BIH Arrhythmia Database for classifying six major groups, i.e. Normal, Ventricular, Atrial, Fusion, Right Bundle and Left Bundle Branch Block beats.

A Resource-Efficient Planning for Pressure Ulcer Prevention

Pressure ulcer is a critical problem for bed-ridden and wheelchair-bound patients, diabetics, and the elderly. Patients need to be regularly repositioned to prevent excessive pressure on a single area of body, which can lead to ulcers. Pressure ulcers are extremely costly to treat and may lead to several other health problems, including death. The current standard for prevention is to reposition at-risk patients every 2 h. Even if it is done properly, a fixed schedule is not sufficient to prevent all ulcers. Moreover, it may result in nurses being overworked by turning some patients too frequently. In this paper, we present an algorithm for finding a nurse-effort optimal repositioning schedule that prevents pressure ulcer formation for a finite planning horizon. Our proposed algorithm uses data from a commercial pressure mat assembled on the beds surface and provides a sequence of next positions and the time of repositioning for each patient.

Pressure ulcer prevention: An efficient turning schedule for bed-bound patients

Pressure ulcer is a critical problem for bed-ridden and wheelchair-bound patients, diabetics, and the elderly. Patients need to be regularly repositioned to prevent excessive pressure on a single area of body, which can lead to ulcers. Pressure ulcers are costly to treat and cause many other health problems, including death. The current standard for prevention is to reposition at-risk patients every two hours. This level of attention is becoming increasingly unrealistic for already overworked nursing staff. In this paper, we present a scheduling algorithm that uses data from a pressure mat on the hospital bed to compute a repositioning schedule that minimizes nursing staff interaction while still preventing pressure ulcer formation. Our experimental results show a 30% increase in the average time between repositioning over the standard schedule. Furthermore, some postures were found to be unsafe if not changed for more than one hour.