Dimitra Emmanouilidou

Adaptive threshold QRS detector with best channel selection based on a noise rating system

QRS detection performance can depend on the type of noise present in each lead involved in the overall processing. A common approach to QRS detection is based on a QRS enhanced signal obtained from the derivatives of the pre-filtered leads. However, the signal pre-filtering cannot be able to perform a complete noise rejection and the use of derivatives can enhance the noise as well. In many cases the noise occurs only on one lead and the addition of a noisy lead to the QRS enhanced signal decreases the overall detection performances of the QRS detector. For this reason the noise estimation on each channel, providing information for the channel inclusion or rejection in building the QRS enhanced signal, can improve the overall performances of the QRS detector. The results have been evaluated on the 48 records of the MIT-BIH Arrhythmia Database where each ECG record is composed by 2 leads sampled at 360 Hz for a total duration of about 30 minutes. The annotated QRSs are 109494 in total. The results have been very satisfying on all the annotated QRSs and, with the inclusion of an automatic criterion for ventricular flutter detection, a sensitivity=99.76% and a positive predictive value=99.81% have been obtained.

A multiresolution analysis for detection of abnormal lung sounds

Automated analysis and detection of abnormal lung sound patterns has great potential for improving access to standardized diagnosis of pulmonary diseases, especially in low-resource settings. In the current study, we develop signal processing tools for analysis of paediatric auscultations recorded under non-ideal noisy conditions. The proposed model is based on a biomimetic multi-resolution analysis of the spectro-temporal modulation details in lung sounds. The methodology provides a detailed description of joint spectral and temporal variations in the signal and proves to be more robust than frequency-based techniques in distinguishing crackles and wheezes from normal breathing sounds.

Morphological classification of heartbeats using similarity features and a two-phase decision tree

Significant clinical information can be obtained from the analysis of the dominant beat morphology. In such respect, the identification of the dominant beats and their averaging can be very helpful, allowing clinicians to perform the measurement of amplitudes and intervals on a beat much cleaner from noise than a generic beat selected from the entire ECG recording. In this paper an algorithm for the morphological classification of heartbeats based on a two-phase decision tree is described. Similarity features extracted from every beat are used in the decision trees for the identification of different morphological classes for the beats of the ECG signal. The results, in terms of dominant beat discrimination, have been evaluated on all annotated beats of the MIT-BIH arrhythmia database with sensitivity = 99.05%, specificity = 93.94%, positive predictive value (PPV) = 99.32% and negative predictive value (NPV) = 91.69%.. Satisfactory results have been also obtained on all the detected beats of the same database using an already published QRS detector developed by the same authors and obtaining sensitivity = 98.71%, specificity = 93.81%, PPV = 99.30% and NPV = 89.11%.

Adaptive Noise Suppression of Pediatric Lung Auscultations With Real Applications to Noisy Clinical Settings in Developing Countries

Goal: Chest auscultation constitutes a portable low-cost tool widely used for respiratory disease detection. Though it offers a powerful means of pulmonary examination, it remains riddled with a number of issues that limit its diagnostic capability. Particularly, patient agitation (especially in children), background chatter, and other environmental noises often contaminate the auscultation, hence affecting the clarity of the lung sound itself. This paper proposes an automated multiband denoising scheme for improving the quality of auscultation signals against heavy background contaminations. Methods: The algorithm works on a simple two-microphone setup, dynamically adapts to the background noise and suppresses contaminations while successfully preserving the lung sound content. The proposed scheme is refined to offset maximal noise suppression against maintaining the integrity of the lung signal, particularly its unknown adventitious components that provide the most informative diagnostic value during lung pathology. Results: The algorithm is applied to digital recordings obtained in the field in a busy clinic in West Africa and evaluated using objective signal fidelity measures and perceptual listening tests performed by a panel of licensed physicians. A strong preference of the enhanced sounds is revealed. Significance: The strengths and benefits of the proposed method lie in the simple automated setup and its adaptive nature, both fundamental conditions for everyday clinical applicability. It can be simply extended to a real-time implementation, and integrated with lung sound acquisition protocols.

Characterization of noise contaminations in lung sound recordings

Lung sound auscultation in non-ideal or busy clinical settings is challenged by contaminations of environmental noise. Digital pulmonary measurements are inevitably degraded, impeding the physicians work or any further processing of the acquired signals. The task is even harder when the patient population includes young children. Agitation and/or crying are captured into the recordings, additionally to any existing ambient noise. This study focuses on characterizing the different types of signal contaminations, expected to be encountered during lung sound measurements in non-ideal environments. Different noise types were considered, including background talk, radio playing, subjects crying, electronic interference sounds and stethoscope displacement artifacts. The individual characteristics were extracted, discussed and further compared to characteristics of clean segments. Additional exploration of discriminatory features led to a spectro-temporal signal representation followed by a standard SVM classifier. Although pulmonary and ambient sounds were both dominant in most sound clips, such a complex representation was deemed to be adequate, capturing most of the signals distinguishing characteristics.

Biomedical Signal and Image Processing for Decision Support in Heart Failure

Signal and imaging investigations are currently a basic step of the diagnostic, prognostic and follow-up processes of heart diseases. Besides, the need of a more efficient, cost-effective and personalized care has lead nowadays to a renaissance of clinical decision support systems (CDSS). The purpose of this paper is to present an effective way to achieve a high-level integration of signal and image processing methods in the general process of care, by means of a clinical decision support system, and to discuss the advantages of such an approach. Among several heart diseases, we treat heart failure, that for its complexity highlights best the benefits of this integration. Architectural details of the related components of the CDSS are provided with special attention to their seamless integration in the general IT infrastructure. In particular, significant and suitably designed image and signal processing algorithms are introduced to objectively and reliably evaluate important features that, in collaboration with the CDSS, can facilitate decisional problems in the heart failure domain. Furthermore, additional signal and image processing tools enrich the model baseof the CDSS.

Listening panel agreement and characteristics of lung sounds digitally recorded from children aged 1–59 months enrolled in the pneumonia etiology research for child health (PERCH) case–control study

Introduction Paediatric lung sound recordings can be systematically assessed, but methodological feasibility and validity is unknown, especially from developing countries. We examined the performance of acoustically interpreting recorded paediatric lung sounds and compared sound characteristics between cases and controls. Methods Pneumonia Etiology Research for Child Health staff in six African and Asian sites recorded lung sounds with a digital stethoscope in cases and controls. Cases aged 1–59 months had WHO severe or very severe pneumonia; age-matched community controls did not. A listening panel assigned examination results of normal, crackle, wheeze, crackle and wheeze or uninterpretable, with adjudication of discordant interpretations. Classifications were recategorised into any crackle, any wheeze or abnormal (any crackle or wheeze) and primary listener agreement (first two listeners) was analysed among interpretable examinations using the prevalence-adjusted, bias-adjusted kappa (PABAK). We examined predictors of disagreement with logistic regression and compared case and control lung sounds with descriptive statistics. Results Primary listeners considered 89.5% of 792 case and 92.4% of 301 control recordings interpretable. Among interpretable recordings, listeners agreed on the presence or absence of any abnormality in 74.9% (PABAK 0.50) of cases and 69.8% (PABAK 0.40) of controls, presence/absence of crackles in 70.6% (PABAK 0.41) of cases and 82.4% (PABAK 0.65) of controls and presence/absence of wheeze in 72.6% (PABAK 0.45) of cases and 73.8% (PABAK 0.48) of controls. Controls, tachypnoea, >3 uninterpretable chest positions, crying, upper airway noises and study site predicted listener disagreement. Among all interpretable examinations, 38.0% of cases and 84.9% of controls were normal (p<0.0001); wheezing was the most common sound (49.9%) in cases. Conclusions Listening panel and case–control data suggests our methodology is feasible, likely valid and that small airway inflammation is common in WHO pneumonia. Digital auscultation may be an important future pneumonia diagnostic in developing countries.

ECG and echocardiography processing for decision support in heart failure

This paper presents an effective way to achieve a high level integration of signal and image processing methods in the general management of heart failure, by means of a clinical decision support system (CDSS). In particular, significant and suitably designed image and signal processing algorithms are introduced to objectively and reliably evaluate important features that can facilitate decisional problems in collaboration with the CDSS. Architectural details of the components of the CDSS needed for the seamless integration of image and signal analysis work-flows are finally discussed.

Rich Representation Spaces: Benefits in Digital Auscultation Signal Analysis

When developing automated techniques for analysis of auscultation signals, the choice of a proper representational space that characterizes all attributes of interest in the signal is of paramount importance. In this paper, we investigate different feature representation methods and their benefits in distinguishing auscultation sounds. The importance of choosing an appropriate feature space is explored and validated using trained classifiers that distinguish between normal and abnormal respiratory sounds. Findings of this study are two-fold: i) an increased dimensionality in the feature space can provide a more complete and distinct representation of the delicate breath sounds and ii) dimensionality of the feature space alone is not enough to fully capture discriminative attributes: an informative feature space is even more crucial for extracting accurate, disease-specific characteristics of respiratory sounds.