Giuliana Faiella

Outliers Detection and Processing in CTG Monitoring

Cardiotocography is a technique used to assess the foetal wellbeing using the recording of foetal heart rate and uterine contractions. Generally these signals are observed to get information about the foetal development and wellbeing and it represents the only medical report that has a legal value to testify the foetal health. Unfortunately, really accurate prediction of foetal wellbeing is still a goal quite difficult to reach. Some problems are related to the visual interpretation of cardiotocographic traces, others to the acquisition system that, in some conditions, can cause the degradation or the loss of the signal. These anomalies, cardiac arrhythmia and artifacts in foetal heart rate signals do not represent physiological variations, so that they can be defined outliers. In cardiotocography literature the problem of outliers processing is underestimated even if they affect both time and frequency analysis. When faced, the outlier problem is solved with a pre-processing phase that uses algorithms to detect and remove spikes.

Analysis of Foetal Heart Rate Variability Components by Means of Empirical Mode Decomposition

Foetal heart rate variability (FHRV) is important in foetal wellbeing assessment. However, a gold standard for its evaluation is not yet available. Here, a rather new methodology, the empirical mode decomposition (EMD), is proposed to decompose FHR signal in its components. To test the reliability of this methodology, we employed simulated FHR signals, “clean” and noisy, with characteristics defined a priori and computed two indices of foetal health, the sympatho-vagal balance (SVB) and the standard deviation of FHR signal (ASD). Results obtained in comparison between values set for the simulation and those estimated after EMD demonstrated that EMD could be useful for evaluation of FHRV components directly in time domain. The error in the indices estimation was on average just over 1% for SVB and zero for ASD. In presence of noise, the error in ASD estimation was below 8% whereas that in SVB evaluation increases becoming almost 30%.

Expanding Healthcare Failure Mode and Effect Analysis: a composite proactive risk analysis approach

Healthcare Failure Mode and Effect Analysis (HFMEA) is a systematic risk assessment method derived from high risk industries to prospectively examine complex healthcare processes. Like most methods, HFMEA has strengths and weaknesses. In this paper we provide a review of HFMEAs limitations and we introduce an expanded version of traditional HFMEA, with the addition of two safety management techniques: Systematic Human Error Reduction and Prediction Analysis (SHERPA) and Systems-Theoretic Accident Model and Processes – Systems-Theoretic Process Analysis (STAMP-STPA). The combination of the three methodologies addresses significant HFMEA limitations. To test the viability of the proposed hybrid technique, we applied it to assess the potential failures in the process of administration of medication in the home setting. Our findings suggest that it is both a viable and effective tool to supplement the analysis of failures and their causes. We also found that the hybrid technique was effective in identifying corrective actions to address human errors and detecting failures of the constraints necessary to maintain safety.

Effects of Wavelets Analysis on Power Spectral Distributions in Laser Doppler Flowmetry Time Series

The evaluation of endothelium function impairments are of great clinical interest in many vascular diseases and laser Doppler flowmetry (LDF) is the gold standard technique for its evaluation. LDF signals show low-frequency oscillations related to heartbeat, respiratory, myogenic and endothelial activities. Although wavelets analysis (WLT) of LDF has been shown as a better technique than Fourier (FFT) and Short Time Fourier Transform (STFT) in the resolution of these oscillatory components, the overall spectral power modifications have not yet been described. Aim of the paper is to study the effects of WLT analysis on power spectral distributions in LDF. We studied 20 min LDF recordings of 20 obese subjects by PeriFlux LDF system. Signals were detrended by moving average algorithms and transformed by FFT, STFT and WLT for spectral analysis. The spectral power has been calculated in bands I (0.6-2 Hz), II (0.145-06), III (0.052-0.145), IV (0.021-0.052), V (0.0095-0.021) and VI (0.005-0.0095) in percent values of the total spectral power. Results of the ANOVA tests between spectral powers in the six bands by the three different spectral analysis’ methods showed that WLT exhibited the highest F-value, reflecting significant difference (p<0.001) between II vs III and IV, III vs V and VI, IV vs V and VI bands. Moreover, while WLT values in I band are significantly lower of those of FFT and STFT, WLT values, both in IV and III band, are significantly higher than those of FFT and STFT (p<0.001). Therefore, the overall effects of WLT analysis on power spectral distributions in LDF time series in the studied population seem to limit the spectral power in the I band moving the power to the III and IV bands with the interesting effect to minimize heart rate variability spectral power enhancing myogenic and neurogenic sympathetic activity spectral power in LDF signals.

A prospective risk assessment of informal carers’ medication administration errors within the domiciliary setting

Abstract Increasingly, medication is being administered at home by family and friends of the care-recipient. This study aims to identify and analyse risks associated with potential drug administration errors made by informal carers at home. We mapped medication administration at home with a multidisciplinary team that included carers, health care professionals and patients. Evidence-based risk-analysis methodologies were applied: Healthcare Failure Modes and Effect Analysis (HFMEA), Systematic Human Error Reduction and Prediction Analysis (SHERPA) and Systems-Theoretic Accident Model and Processes (STAMP). The process of administration comprises seven sub-processes. Thirty-four possible failure modes were identified and six of these were rated as high risk. These highlighted that medications may be given with a wrong dose, stored incorrectly, not discontinued as instructed, not recorded, or not ordered on time, and often caused by communication and support problems. Combined risk analyses contributed unique information helpful to better understand the medication administration risks and causes within homecare. Practitioner Summary: Increasingly, medication is being administered at home by family and friends of the care-recipient. This study identifies risks associated with potential drug administration errors made by informal carers at home through consensus-based quantitative techniques. The different analyses contribute unique information helpful to better understand the administration risks and causes.

Symbolic dynamics in cardiotocographic monitoring

Foetal heart rate variability is considered an important parameter, capable to highlight foetal reactivity and wellbeing. Many studies focused their attention on foetal heart rate variability analysis, using different techniques both in time and frequency domain. Proposed parameters have shown some limitations, due to their inability to analyze nonlinear dynamics, involved in heart rate control and relevant for clinical aims. In the last decades, nonlinear analysis methods have gained a growing interest. Here, we described an application of a nonlinear analysis (symbolic dynamic analysis) and introduced a new index to quantify foetal heart rate variability, in a simple and concise way, in healthy foetuses near to term. The index was able to highlight differences in foetal heart rate variability, not always so evident on printed cardiotocographic signals, and resulted significantly correlated with the different kinds of delivery (spontaneous or caesarean); in particular, higher variability index values corresponded to spontaneous delivery.

Process modeling of devices management in home care

Home care (HC) consists of providing a high quality health services to patients in their homes through the use of caregivers and appropriate technologies. HC is a complex system that can be decomposed in simple sub-processes, activities and actors to be better designed, managed and improved. This work presents an attempt to model the activities related to test and management of technologies in the HC care service provided by Local Heath Unit ASL NA1 (Naples, Italy) covering a population of one million inhabitants. The modeling was done using a set of methods / diagrams to describe the process from multiple point of views. Each one allows the identification of resources and constraints to perform the activities. The set of different diagrams has clarified many aspects of test and management of the devices in HC and it has facilitated the fulfillment of operative procedures that will regulate activities in provide HC service.

FMECA and HFMEA of indoor air quality management in home mechanical ventilation

Risk analysis (RA) plays an important role in the development of safety-critical healthcare systems and numerous methods have been specifically designed for the risk analysis. As a matter of fact, these processes are complex, composed by many sub-processes and characterized by the presence of sophisticated technologies used in diagnoses and therapies. According to these considerations, the Clinical RA (CRA) techniques derive from those used in other complex industrial systems, adapted on concepts and settings typical of healthcare. Often the industrial risk analysis methods combine more risk analysis methods with the advantage of obtaining multi-views analysis exploiting the different focus points of each method. In this work, a particular critical aspect of home care service with high level of assistance has been analysed using the combination of Failure Mode Effects and Criticality Analysis (FMECA) and Healthcare Failure Mode Effects Analysis (HFMEATM). The results show that the use of both methods increases the efficiency of analysis because it highlights a more complete and effective set of criticalities and vulnerabilities.

Use of Modelling Simulation to Monitor the Performance of a Pediatric Emergency Department

Hospitals are complex systems in which different departments interact with each other. Systemic analytical models use systematic representations to facilitate the understanding of a complex process. The models are simulated with an imitation of the operations of a system and its internal processes, usually over time, and in appropriate detail. Simulation is used to study the performance of new systems as well as predict the effect of changes. Since Emergency Departments (ED) are unpredictable and frontline of healthcare service delivery, their detailed description and analysis are highly needed in order to design improvement actions. In this study, we will present the simulation of the ED a pediatric hospital which admits more than 100 k/year patients. The simulation tool is SIMIO®.

Design of a smart EIS measurement system

The Electrical Impedance Spectroscopy (EIS) measurements, in addition to being used for research purposes, are increasingly spreading in clinical setting. Thanks to the improvement of electronic and information technology, the necessary equipment is becoming increasingly simple and small, suitable for example for home care applications. In this work, the characteristics of a smartphone based system for impedance measurements are described. Besides, it is tested its reliability by measuring resistors of known values which fall in the range of bioimpedance. The proof demonstrator, here assessed, consists of a prototypal analogic, low-power chip capable of performing impedance measurements at different frequencies, a hardware support for the connection with a standard smartphone and an application properly developed. After its final validation, the proof demonstrator could be useful to design wearable devices for sport and consumer electronic.