Alarm Recognition Research Articles

Improving auditory alarms: Reducing perceived annoyance with musical timbre (a randomized trial)

BackgroundAuditory alarms in medical environments are widely criticized due to problematic design issues affecting learnability, recognition, and annoyance. To overcome these issues, new alarm sets vary properties like pitch, timing, and timbre—the “quality” of a sound. Past work identifies timbre as a promising candidate for improving alarm recognition, however most studies manipulate multiple acoustic parameters, obfuscating the independent effect of timbre on melodic perception. MethodsTo better understand timbre's potential for improving alarm recognizability and reducing annoyance, we recruited forty-four participants for a melody recognition experiment. In a learning phase, participants learned short alarm melodies performed with either a complex musical timbre based on a xylophone or one following the specifications of a universal alarm standard. During a recall phase, participants heard the melodies again presented in either the xylophone or standard timbre and attempted to identify the corresponding label for each alarm. Afterward, participants heard alarm pairs presented in both timbres and chose which one they perceived to be more annoying. ResultsParticipants recalled xylophone alarm melodies with comparable accuracy to the standard timbre, however they found the musical timbre substantially less annoying. Regardless of the timbre at exposure, recognition of familiar melodies decreased if the timbre changed between learning and recall. ConclusionsWe compared the recognizability and annoyance of alarm melodies presented in either a standard alarm timbre or a synthesized xylophone timbre. Our findings reveal musical sounding alarms are comparably recognizable, yet significantly less annoying than alarm signals common in medical environments. These outcomes provide a promising first step to improving patient care through musically informed alarm design.

A fire alarm judgment method using multiple smoke alarms based on Bayesian estimation

Smoke alarms are the most widely used fire sensors; however, their false-alarm rate is high. Owing to the difficulties in replacing all installed smoke alarms in a short period, improving the fire detection accuracy of the installed alarms is a significant challenge for the fire emergency response team. Therefore, this study proposes a fire alarm judgment method with multiple smoke alarms based on Bayesian estimation. To this end, we propose a novel criterion called fire alarm authenticity, wherein the alarm time intervals of multiple smoke alarms are used to calculate the posterior probability distribution of the location and intensity of the fire source without installing any other type of sensors. Subsequently, an alarm sequence classification method based on the proposed criterion is developed to judge a real fire and is validated through simulations of both real fires and false alarms. The results show that the proposed method can identify 77.5% of false alarms based on no real fires to be misjudged. Applications in real fire alarm systems are also discussed, which indicate that the false alarm recognition rate reaches 80%. The proposed method provides a theoretical basis for reducing the existing false-alarm rate with multiple same-type fire alarms or detectors.

Nurses’ clinical alarm-related behaviors and influencing factors in China†

Abstract Objective To explore the nurses’ behaviors regarding clinical alarms, analyze the related influencing factors, and provide rationales for alarm management. Methods A cross-sectional survey was conducted in China. The self-made questionnaire of nurses’ clinical alarm-related knowledge, attitude, and behavior (NCAKAB) was used. Results The valid response rate was 98.66% (n = 2368). The average nurses’ clinical alarm-related behaviors (NCAB) score was 65.14 ± 7.95 (out of 85). The dimension scores of NCAB from high to low were alarm learning (4.02 ± 0.85, out of 5), alarm response (27.99 ± 3.64, out of 35), alarm setting (19.24 ± 3.88, out of 25), alarm recognition (7.63 ± 1.68, out of 10) and alarm notification (6.25 ± 1.84, out of 10). There were significant differences in alarm behavior scores between nurses of different ages (F = 4.619, P = 0.000), nursing stints (F = 9.564, P = 0.000), professional titles (F = 4.425, P = 0.004), departments (F = 9.166, P = 0.000), and hospital levels (t = 2.705, P = 0.007). The study showed that nurses’ total alarm behavior scores were positively correlated with the total alarm knowledge score (r = 0.267; P < 0.001) and the total alarm attitude score (r = 438; P < 0.001). Conclusions Nurses scored highest in alarm learning, followed by alarm response, alarm setting, alarm recognition, and alarm notification behavior. The factors that influenced alarm behavior included age, title, department, nursing stint, hospital level, professional title, alarm-related training, willingness to participate in alarm-related training, whether or not departments have improved alarm management over the last 3 years, and whether or not departments have formulated norms for alarm management. Nurses with higher scores for clinical alarm knowledge had higher correlating scores for alarm behavior; similarly, nurses with higher scores for clinical alarm attitude had higher scores for alarm-related behavior.

Evaluating the Alarm Fatigue and its Associated Factors among Clinicians in Critical Care Units

The Alarm Fatigue (AF) occurs when clinicians are exposed to a large number of false alarms which can cause alarm desensitization. AF is a well-recognized patient safety concern in intensive care settings. This study aims to evaluate the alarm fatigue in clinicians along with its associated factors in critical care settings using a valid and reliable instrument. During 2020, a cross-sectional study was conducted among clinicians working in Critical care units in Lebanon and enrolled 337 participants via self-administered electronic questionnaire. The survey included questions on socio-demographic characteristics, health status, staff recognition of alarms and alarm management, and Alarm Fatigue questionnaire. The translation and cultural adaptation of the Alarm Fatigue questionnaire in Arabic followed a standardized protocol. The mean age of participants was 30.79 (SD=7.26; median=29) years, 61.4% were women. 77.2% were nursing staff and 22.8% were physicians. 78.8% worked in private hospitals. 41.3% of staff were from neonatal and pediatric intensive care units. 58.8% of clinicians find that more than 50% of alarms were irrelevant, and 69.4% deactivate and silence the alarm when annoyed. The Alarm Fatigue questionnaire has good reliability and validity. The Cronbach alpha was 0.69 which establishes good reliability for group comparison. AF in physicians (OR=2.25; 95%CI=-0.01-4.50, p=0.05) and Registered Nurses (OR=2.64; 95%CI=-0.52-4.75, p=0.02) was higher than in nurses and head nurses; AF was more frequent in ICUs units (OR=2.1; 95%CI=0.05-4.18, p=0.04) compared to other critical care units. The AF was higher for the clinicians that deactivate and silence the alarm when annoyed compared to other subgroup (OR=4.14, 95%CI=2.75–5.53, p<0.001). The AF was 3.14 times frequent for the staff who reported stress (p<0.001). Non-actionable alarms occurred frequently and associated to alarm fatigue. Physicians score slightly higher alarm fatigue than nurses. The Arabic Alarm Fatigue questionnaire has a good psychometric properties and suitable to evaluate the AF in Lebanon. Multidisciplinary approaches can be established to minimize the incidence of AF using a valid and reliable instrument.

Development and Validation of an Algorithm for the Identification of Audible Medical Alarms.

Audible medical alarms are ubiquitous in acute healthcare environments, but caregivers cannot reliably identify them. Furthermore, background noise and psychoacoustic factors can interfere with alarm recognition and contribute to alarm fatigue. We developed and validated an acoustic digital signal processing algorithm for the automatic identification of audible medical alarms. The algorithm uses the short-time Fourier transform to decompose audio signals and extract the alarm sounds' fundamental frequencies, harmonics, and periodicity. This information is then used to classify and recognize these sounds. The identification algorithm demonstrates robust performance (F1 score of 93% to 100%) and 100% negative predictive value in identifying single or multiple medical audible alarms under both quiet and noisy conditions. The algorithm we developed represents a robust approach for the identification of audible medical alarms that perform with high accuracy in noisy environments. It can be used to identify and classify alarms in medical settings for research and clinical purposes.

A Method of False Alarm Recognition in Built-in Test Considering Its Time Series Characteristics

Built-in test (BIT) technology provides an effective way of state monitoring, and is widely used in many fields. However, the false alarm prevents its wider usage and reduces its reliability. The existing BIT methods typically assume that the probability of generating a false alarm is a stationary process, which is typically a far too simplifying assumption. In this article, we propose a new method of false alarm recognition, considering the time series characteristics of BIT. The presented method divides the evolution process of false alarm into three phases, and the false alarm recognition in the intermediate phase is discussed and studied. Multiclassification support-vector machine is introduced to identify the intermediate phase, and a new feature, the irrational entropy, is proposed as its input. Based on the phase identification result, we construct a hidden Markov model for every state-normal, false alarm, and faulty state. The state with the highest probability is taken as the true state. Finally, a dc-dc converter is taken as an example, and the proposed method is applied and compared with the repeated test method, artificial neural network and recurrent neural network. The results show that the presented method outperforms the other algorithms.

Widget-based Alarm Recognition and a Management Support Interface System in Nuclear Power Plants

As small modular reactors (SMRs) and various computerized operating support systems have been developed, visual display unit(VDU)-based multifunctional consoles have been proposed. The more a console depends on VDUs, the more human system interface (HSI) designers need to consider the decrease in the human performance caused by the effect of navigation tasks. This paper proposes a widget-based alarm recognition and management support interface system to increase the operator’s ability to recognize alarms and to manage alarm-related tasks by reducing additional and excessive navigation tasks. With a small display screen, the system provides operators with quick and accurate access to alarm-related information and allows them to make corrective responses.

Game Theoretic Approach for Systematic Feature Selection; Application in False Alarm Detection in Intensive Care Units.

Intensive Care Units (ICUs) are equipped with many sophisticated sensors and monitoring devices to provide the highest quality of care for critically ill patients. However, these devices might generate false alarms that reduce standard of care and result in desensitization of caregivers to alarms. Therefore, reducing the number of false alarms is of great importance. Many approaches such as signal processing and machine learning, and designing more accurate sensors have been developed for this purpose. However, the significant intrinsic correlation among the extracted features from different sensors has been mostly overlooked. A majority of current data mining techniques fail to capture such correlation among the collected signals from different sensors that limits their alarm recognition capabilities. Here, we propose a novel information-theoretic predictive modeling technique based on the idea of coalition game theory to enhance the accuracy of false alarm detection in ICUs by accounting for the synergistic power of signal attributes in the feature selection stage. This approach brings together techniques from information theory and game theory to account for inter-features mutual information in determining the most correlated predictors with respect to false alarm by calculating Banzhaf power of each feature. The numerical results show that the proposed method can enhance classification accuracy and improve the area under the ROC (receiver operating characteristic) curve compared to other feature selection techniques, when integrated in classifiers such as Bayes-Net that consider inter-features dependencies.

Raising Students' Clinical Alarm Awareness: Evaluating a Learning Activity

The American College of Clinical Engineering (ACCE) has found that lack of attention to alarms applies to both experienced and inexperi- enced practicing nurses (ACCE Healthcare Technology Foundation, 2006). The Joint Commission also notes safety concerns that arise from a lack of attention to clinical alarms (HCPro, 2005). A literature review on attentiveness in the workplace indicates that concentrating on another task, fatigue, distractions, and numerous alarms (some of which are false) can interfere with an individual's capacity to pay attention (Curry, Meyer, & McKinney, 2006; Institute for Safe Medication Practices, 2009; Mack & Roche, 1998, as cited in Karns & Rivardo, 2010).In response to concerns by the Quality and Safety Education for Nurses (QSEN) Committee regarding the gap between the learning envi- ronment and practice (Thornlow & McGuinn, 2010), Ardoin and Wilson attempted to connect the classroom to the practice setting by using Spencer and Foss' (2009) Clinical Alarm Safety learning activity with a group of eight junior baccalaureate nursing (BSN) students. Comments and actions from this group of students validated that the activity met the QSEN 2011 proposed targets for knowledge, skills, and attitudes. Although students were found to begin to internalize new values for alarm recognition and safety (Ardoin & Wilson, 2014), it was uncertain whether their heightened awareness of clinical alarms was due to the learning activity or simply from the exposure to alarms on a clinical unit.The intent of this study was to determine if first-semester junior BSN students showed heightened awareness of clinical alarms in their student practice after using the Spencer and Foss (2009) learning activity. It was hoped that positive findings would support the integra- tion of the activity in the curriculum and a reevaluation of the students' clinical alarm awareness upon graduation and one year thereafter.methodStudy DesignA quasi-experimental descriptive study, with a nonequivalent con- trol group design, allowed for subjecting half the class to the learn- ing activity and for following the same group of students over time (Polit & Beck, 2010). Students completed the Students' Clinical Alarm Awareness Survey at the beginning and end of the semester. To ensure anonymity, no signatures were required.Population and SampleFollowing approval from the university's institutional review board, students learned of the research study's purpose and voluntary nature and that participation would not affect their course grade. A nonprob- ability convenience selection tapered a population of 78 first-semester junior BSN students to 58. Four experimental groups (EG) contained a total of 24 students, and four control groups (CG) had 34 students; no group had more than 10 students. Students with previous and/or current hospital experience were excluded from the study.InstrumentThe Students' Clinical Alarm Awareness Survey was a modification (with permission) of the 2006 Clinical Alarm Awareness Survey devel- oped by the ACCE (ACCE Healthcare Technology Foundation, 2006). No reliability or validity testing of previous versions of the survey has been reported in the literature. The modified survey centers on stu- dents in a learning situation versus practicing nurses in the workplace. The original survey consisted of 26 items; to focus on student aware- ness of clinical alarms, a panel of five nursing professors reestablished face and content validity for nine items in the modified version. The panelists felt the instrument looked appropriate and contained an ade- quate representation of the construct (Polit & Beck, 2010).Data AnalysisStatistical Package for the Social Sciences (SPSS) predictive software, version 18.0.3, was used to analyze data. RM-ANOVA was used to examine group differences over time. Applying p values

Semi-supervised detection of intracranial pressure alarms using waveform dynamics

Patient monitoring systems in intensive care units (ICU) are usually set to trigger alarms when abnormal values are detected. Alarms are generated by threshold-crossing rules that lead to high false alarm rates. This is a recognized issue that causes alarm fatigue, waste of human resources, and increased patient risks. Recently developed smart alarm models require alarms to be validated by experts during the training phase. The manual annotation process involved is time-consuming and virtually impossible to achieve for the thousands of alarms recorded in the ICU every week. To tackle this problem, we investigate in this study if the use of semi-supervised learning methods, that can naturally integrate unlabeled data samples in the model, can be used to improve the accuracy of the alarm detection. As a proof of concept, the detection system is evaluated on intracranial pressure (ICP) signal alarms. Specific morphological and trending features are extracted from the ICP signal waveform to capture the dynamic of the signal prior to alarms. This study is based on a comprehensive dataset of 4791 manually labeled alarms recorded from 108 neurosurgical patients. A comparative analysis is provided between kernel spectral regression (SR-KDA) and support vector machine (SVM) both modified for the semi-supervised setting. Results obtained during the experimental evaluations indicate that the two models can significantly reduce false alarms using unlabeled samples; especially in the presence of a restrained number of labeled examples. At a true alarm recognition rate of 99%, the false alarm reduction rates improved from 9% (supervised) to 27% (semi-supervised) for SR-KDA, and from 3% (supervised) to 16% (semi-supervised) for SVM.

Alarming features: birds use specific acoustic properties to identify heterospecific alarm calls.

Vertebrates that eavesdrop on heterospecific alarm calls must distinguish alarms from sounds that can safely be ignored, but the mechanisms for identifying heterospecific alarm calls are poorly understood. While vertebrates learn to identify heterospecific alarms through experience, some can also respond to unfamiliar alarm calls that are acoustically similar to conspecific alarm calls. We used synthetic calls to test the role of specific acoustic properties in alarm call identification by superb fairy-wrens, Malurus cyaneus. Individuals fled more often in response to synthetic calls with peak frequencies closer to those of conspecific calls, even if other acoustic features were dissimilar to that of fairy-wren calls. Further, they then spent more time in cover following calls that had both peak frequencies and frequency modulation rates closer to natural fairy-wren means. Thus, fairy-wrens use similarity in specific acoustic properties to identify alarms and adjust a two-stage antipredator response. Our study reveals how birds respond to heterospecific alarm calls without experience, and, together with previous work using playback of natural calls, shows that both acoustic similarity and learning are important for interspecific eavesdropping. More generally, this study reconciles contrasting views on the importance of alarm signal structure and learning in recognition of heterospecific alarms.

Fault Diagnosis with Multi-State Alarms in a Nuclear Power Control Simulator

This research addresses how alarm systems can increase operator performance within nuclear power plant operations. The experiment examined the effect of two types of alarm systems (two-state and three-state alarms) on alarm compliance and diagnosis for two types of faults differing in complexity. We hypothesized three-state alarms would improve performance in alarm recognition and fault diagnoses over that of two-state alarms. We used sensitivity and criterion based on Signal Detection Theory to measure performance. We further hypothesized that operator trust would be highest when using three-state alarms. The findings from this research showed participants performed better and had more trust in three-state alarms compared to two-state alarms. Furthermore, these findings have significant theoretical implications and practical applications as they apply to improving the efficiency and effectiveness of nuclear power plant operations.

System-level BIT false alarm reducing technology based on sensor network time stress analysis and the fault propagation correlation model

Equipment fault propagation and sensor faults are the main causes of false alarms in built-in test (BIT) systems. Hence, the current paper analyses the mechanism of system-level BIT false alarms. The technical process of system-level BIT false alarm recognition is presented. First, sensor damage states are estimated using sensor network time stress data generated during its lifetime. In different damage states, a dual support vector machine (SVM) relevance model is built, it can recognise system-level BIT type II false alarms induced by sensor faults. Subsequently, a fault propagation correlation model of the sensor network is set up. Based on this model, a false alarm recognition method is proposed based on probability reasoning. This method can effectively reduce system-level BIT type I false alarms caused by sensor network fault propagation. Lastly, a case about the helicopter horizon system is studied to validate the proposed technology.

Contextual effect in people with Williams syndrome

This study was aimed at investigating the semantic integration ability of people with WS in building up a coherent and gist theme from the context of presented sentences. Previous studies have indicated rich lexical semantic knowledge and typical semantic priming in this clinical group, but atypical brainwave patterns have been reported in studies of semantic comprehension. An integration difficulty hypothesis of merging meanings into sentences was proposed to explain the discrepancy (Tyler et al., 1997). In this study, this hypothesis was examined by means of proposition integration. Participants were presented sentences embedded with different numbers of propositions under various scenarios. Successful integration of semantically related propositions under same scenario was demonstrated by high false alarm recognition to sentences with more propositions (a maximum of four in this study). The results revealed that healthy adults showed integration ability when the number of propositions in a sentence were more than three, with increased false alarm rates and confidence ratings, whereas people with WS showed no differences in sentences with increased proposition numbers. Typically developing children controls confined recognition phenomenon to sentences carrying more than two propositions. Hence, it was concluded that contextual effects in terms of proposition integration was a gradual development from childhood into adulthood and a deviant development in people with WS.

Evaluation of the user-friendliness of seven new generation intensive care ventilators

To explore the user-friendliness and ergonomics of seven new generation intensive care ventilators. Prospective task-performing study. Intensive care research laboratory, university hospital. Ten physicians experienced in mechanical ventilation, but without prior knowledge of the ventilators, were asked to perform eight specific tasks [turning the ventilator on; recognizing mode and parameters; recognizing and setting alarms; mode change; finding and activating the pre-oxygenation function; pressure support setting; stand-by; finding and activating non-invasive ventilation (NIV) mode]. The time needed for each task was compared to a reference time (by trained physiotherapist familiar with the devices). A time >180 s was considered a task failure. For each of the tests on the ventilators, all physicians' times were significantly higher than the reference time (P < 0.001). A mean of 13 +/- 8 task failures (16%) was observed by the ventilator. The most frequently failed tasks were mode and parameter recognition, starting pressure support and finding the NIV mode. Least often failed tasks were turning on the pre-oxygenation function and alarm recognition and management. Overall, there was substantial heterogeneity between machines, some exhibiting better user-friendliness than others for certain tasks, but no ventilator was clearly better that the others on all points tested. The present study adds to the available literature outlining the ergonomic shortcomings of mechanical ventilators. These results suggest that closer ties between end-users and manufacturers should be promoted, at an early development phase of these machines, based on the scientific evaluation of the cognitive processes involved by users in the clinical setting.

Increased activation of the left hippocampus region in Complex PTSD during encoding and recognition of emotional words: A pilot study

To gain insight into memory disturbances in Complex Posttraumatic Stress Disorder (Complex PTSD), we investigated declarative memory function and medial temporal lobe activity in patients and healthy non-traumatized controls. A case-control study was performed in nine patients with Complex PTSD and nine controls. All respondents performed a declarative memory task with neutral and emotional, negative words during functional magnetic resonance imaging. Memory performance of neutral words was impaired in Complex PTSD with a relative conservation of recall of negative words. Deep encoding of later remembered negative words, as well as correct recognition of negative words and false alarms, was associated with an enhanced Blood Oxygenation Level Dependent (BOLD) response in the left hippocampus extending into the parahippocampal gyrus of Complex PTSD patients compared with controls. Post-hoc volumetric comparisons did not reveal significant anatomical differences in the medial temporal lobe between Complex PTSD patients and controls. We conclude that in Complex PTSD preferential recall of negative words is associated with increased activation in the left hippocampus and parahippocampal gyrus during both successful and false recall. These findings support a model of an abnormally functioning hippocampus in Complex PTSD.

Implicit contributions of context to recognition

Studies of the impact of context on remembering have not focused on the influence of contextual contingency on subsequent recognition in the condition in which the contingency cannot be verbalized. In two experiments, we analyzed the effect of an implicitly encoded position contingency involving location and semantic category on both hit and false alarm recognition judgments after 1 day and 1 week delays. We vigorously probed for what participants could say about the contingency. We found context effects for both hits and false alarms, whether or not participants could verbalize any knowledge they might have of contingency. These results suggest that people may use contextual information when making a recognition judgment even if they are not aware of this information.

Alarms- Localization, Orientation, and Recognition

This paper documents a study at three dialysis departments on the use of different alarm systems. The design of the alarm systems is to some extent in line with the visions of augmented reality and ubiquitous computing. Our study has raised a range of questions which we think are relevant for the research on how to embed technology in users' environment. We will address how the properties of the specific alarm systems, concerted in the local physical environment of the dialysis department, affords the localisation, orientation and recognition of alarms. In this paper, localisation is about the ways in which the nurses exactly locate where the alarm comes from and orientation is how the nurses orientate themselves towards the alarm in a more imprecise way. Recognition is about how the nurses notice that there is an alarm in the first place and if there are multiple alarms.

Automatic classification of wideband acoustic signals

Until now, very few contributions were published in the field of wideband acoustic signal recognition, especially for handling impulsive noise signals such as glass breaking, detonations, or door slams, as encountered in security applications, where the signals are highly nonstationary and composed of higher frequency components. This paper shows how the audio alarm recognition problem can efficiently be tackled using either pattern recognition methods relying on Bayes classifiers, or on artificial neural networks (ANN). After extraction of filterbank coefficients in the acoustic analysis module, typical feature vectors are achieved from the concatenation of k consecutive signal frames, in order to exploit dynamic temporal information. The redundancy induced by this dynamic modeling requires a reduction of the feature space dimension, performed with a conventional principal component analysis. The performance of both systems is evaluated experimentally (7000 tests) for the classification of three types of noise events (glass breaking, door slams, and stationary noises). The score of correct classification reaches 99% for the statistical approach, and 98% for the ANN-based system. It is further noticed that errors do not occur on the same signal files for both methods. Thus robustness enhancement can be reasonably expected, when using hybrid methods or fusion strategies.

Application of a passive polarimetric infrared sensor in improved detection and classification of targets

Automatic detection and recognition of targets by means of passive IR sensors suffer from limitations due to lack of sufficient contrast between the targets and their background, and among the facets of a target. In this paper the results of a suite of polarization-sensitive automatic target detection and recognition algorithms on sets of simulated and real polarimetric IR imagery are presented. A custom designed Polarimetric IR (PIR) imaging sensor is used for collecting real polarimetric target data-three of the four Stokes parameters under a variety of conditions. Then a set of novel algorithms are designed and tested that uses the target and background Stokes parameters for detection, segmentation and classification of targets. The empirical performance results are obtained in terms of the probabilities of detection, false alarm rate, segmentation accuracy, and recognition probabilities as functions of number of pixels on target, aspect and depression angles and under several background conditions (clutter densities) on the polarimetric and non-polarimetrirc data. These results show that a noticeable improvement over the non-polarimetric ATR can be achieved.