Accurate assessment of pacemaker function or malfunction is essential to make clinical interpretations on pacemaker therapy and patient symptoms. This article presents an innovative approach for detecting pacemaker pulses at sampling frequency as low as 125Hz. The proposed method is validated in wide range of simulated clinical ECG conditions such as arrhythmia (sinus rhythms, supraventricular rhythms, and AV blocks), pulse amplitudes (~100µV to ~3mV), pulse durations (~100µs to ~2ms), pacemaker modes and types (fixed– rate or on–demand single chamber, dual chamber, and bi– ventricular pacing), and physiological noise (tremor). The proposed algorithm demonstrates clinically acceptable detection accuracies with sensitivity and PPV of 98.1 ± 4.4 % and 100 %, respectively. In conclusion, the approach is well suited for integration in long–term wearable ECG sensor devices operating at a low sample frequency to monitor pacemaker function.
With rapid advancement in wearable biosensor technology, systems capable of real time, continuous and\ ambulatory monitoring of vital signs are increasingly emerging and their use can potentially help improve patient outcome. Monitoring continuous body temperature offers insights into its trend, allows early detection of fever and is critical in several diseases and clinical conditions including septicemia, infectious disease and others. There is a complex interaction between physiological and ambient parameters including heart rate, respiratory rate, muscle rigors and shivers, diaphoresis, local humidity, clothing, body, skin and ambient temperatures among others. This article presents feasibility analysis of a wireless biosensor patch device called as VitalPatch in capturing this physio-ambient-thermodynamic interaction to determine core body temperature, and details comparative performance assessments using oral thermometer and ingestible pill as reference devices. Based on a study on a cohort of 30 subjects with reference oral temperature, the proposed method showed a bias of 0.1 ± 0.37 ºC, mean absolute error (MAE) of 0.29 ± 0.25 ºC. Another cohort of 22 subjects with continuous core body temperature pill as reference showed a bias of 0.16 ± 0.38 ºC and MAE of 0.42 ± 0.22 ºC.
Objective classification of heart sound signals can provide significant advancements in the screening of structural heart abnormalities. An algorithm, based on auditory filter models and probabilistic segmentation of heart cycle sequences, is proposed for classifying phonocardiogram (PCG) signals as normal or abnormal. It first involves segmentation of cardiac recordings into a sequence of four heart stages, namely S1, systole, S2 and diastole, using a hidden Markov model approach. Secondly, gammatone frequency cepstral coefficient (GFCC) features are extracted by applying the gammatone filterbank model to two independent binary classification problems: one with PCG segmentation, and one without segmentation. Weighted support vector machine models are trained to classify the PCG signals as normal vs. abnormal records using the GFCC features. The algorithm is trained and cross-validated using the 2016 PhysioNet Computing in Cardiology Challenge database of 3,240 PCG recordings. Based on 10-fold stratified cross-validation, the performance of the proposed “with segmentation” approach is determined to have a sensitivity of 90.3% and a specificity of 89.9%, while the “without segmentation” approach shows a comparable performance of 87.1% sensitivity and 88.5% specificity. Thus, the proposed algorithm demonstrates strong, clinically acceptable performance for automated screening of heart sound signals, and it can be a useful diagnostic tool in clinical practice to screen patients for structural and functional heart diseases.