Early Detection Of Respiratory Disease Research Articles

Responding to challenge call for machine learning model development in diagnosing respiratory disease sounds

The normal and abnormal sounds from the respiratory system shed great light on medical science by revealing the quality, diseases and changes in people’s lungs. In medicine, this easy and old method, realized by the stethoscope, facilitates the diagnosis of diseases by specialists. This manual method can sometimes lead to wrong decisions in terms of sound detection due to different audibility. Detailed sound analysis is crucial for accurately detecting lung diseases with high mortality rates. As technology advances, the development of automated approaches based on machine learning is of great interest as they provide modern and highly accurate analysis. In today’s most popular topic, i.e., the COVID-19 disaster, the conflict between early detection of respiratory disease and machine learning for sound signal processing is extremely important. In this study, a machine learning model was developed to detect respiratory system sounds such as sneezing/coughing in disease diagnosis. The automatic model and approach development of breath sounds, which carry valuable information, results in early diagnosis and treatment. A successful machine learning model was developed in this study, which was a strong response to the challenge called the “Pfizer digital medicine challenge” on the “OSFHOME” open access platform. In the database provided in this challenge, which consists of 3 parts, features that effectively showed coughing/sneezing sound analysis were extracted from training, testing and validating samples. Based on the Mel frequency cepstral coefficients (MFCC) feature extraction method, mathematical and statistical features were prepared. The sequential forward selection (SFS) feature selection method was used to select the relevant and dominant variables among the obtained features to represent the model fully and accurately. Three different classification techniques were considered for successful respiratory sound classification in the dataset containing more than 3800 sounds. Support vector machine (SVM) with radial basis function (RBF) kernels, decision tree and ensemble aggregation classification methods were used as classification techniques. In an attempt to classify coughing/sneezing sounds from other sounds, SVM with RBF kernels was achieved with 83% success.

Monitoring respiratory state from near-infrared face video images

AbstractIn this paper, we propose an algorithm for estimating respiratory state using near-infrared facial video images. Estimation of respiratory state is an important indicator for early detection of respiratory diseases. In particular, there is a demand for monitoring respiratory state during the night. One method of monitoring respiratory state is to use contact-type sensors. However, this method requires the installation of many sensors and a visit to a hospital, which place a burden on patients. Therefore, we propose to acquire respiratory-induced features from near-infrared face video images and investigate their similarity to measurements obtained with a respirometer for non-contact monitoring of respiratory state in the dark. Respiratory-induced features were obtained from pulse wave signals extracted from the face video images. The results showed correlations in several respiratory states. This study opens some perspectives in non-contact monitoring of respiratory states.

Cochleogram-based adventitious sounds classification using convolutional neural networks

Background:The World Health Organization (WHO) establishes as a top priority the early detection of respiratory diseases. This detection could be performed by means of recognizing the presence of acoustic bio-markers (adventitious sounds) from auscultation because it is still the main technique applied in any health center to assess the status of the respiratory system due to its non-invasive, low-cost, easy to apply, fast to diagnose and safe nature. Method:Despite the novel deep learning approaches applied in this biomedical field, there is a notable lack of research that rigorously focuses on different time–frequency representations to determine the most suitable transformation to feed data into Convolutional Neural Network (CNN) architectures. In this paper, we propose the use of the cochleogram, based on modeling the frequency selectivity of the human cochlea, as an improved time–frequency representation to optimize the learning process of a CNN model in the classification of respiratory adventitious sounds. Our proposal is evaluated using the largest and most challenging public database of respiratory sounds. Results:The cochleogram obtains the best binary classification results among the compared methods with an average accuracy of 85.1% in wheezes and 73.8% in crackles, and a competitive performance evaluating a multiclass classification scenario in comparison with other well-known state-of-the-art deep learning models. Conclusion:The cochleogram provides a suitable time–frequency representation since it is able to model respiratory adventitious content more accurately by means of non-uniform spectral resolution and due to its increased robustness to noise and acoustic changes. This fact implies a significant improvement in the learning process of CNN models applied in the classification of respiratory adventitious sounds.

Extracellular Vesicle-Encapsulated microRNAs as Novel Biomarkers of Lung Health.

Rationale: Early detection of respiratory diseases is critical to facilitate delivery of disease-modifying interventions. Extracellular vesicle-enriched microRNAs (EV-miRNAs) may represent reliable markers of early lung injury. Objectives: Evaluate associations of plasma EV-miRNAs with lung function. Methods: The prospective NAS (Normative Aging Study) collected plasma EV-miRNA measurements from 1996-2015 and spirometry every 3-5 years through 2019. Associations of EV-miRNAs with baseline lung function were modeled using linear regression. To complement the individual miRNA approach, unsupervised machine learning was used to identify clusters of participants with distinct EV-miRNA profiles. Associations of EV-miRNA profiles with multivariate latent longitudinal lung function trajectories were modeled using log binomial regression. Biological functions of significant EV-miRNAs were explored using pathway analyses. Results were replicated in an independent sample of NAS participants and in the HEALS (Health Effects of Arsenic Longitudinal Study). Measurements and Main Results: In the main cohort of 656 participants, 51 plasma EV-miRNAs were associated with baseline lung function (false discovery rate-adjusted P value < 0.05), 28 of which were replicated in the independent NAS sample and/or in the HEALS cohort. A subset of participants with distinct EV-miRNA expression patterns had increased risk of declining lung function over time, which was replicated in the independent NAS sample. Significant EV-miRNAs were shown in pathway analyses to target biological pathways that regulate respiratory cellular immunity, the lung inflammatory response, and airway structural integrity. Conclusions: Plasma EV-miRNAs may represent a robust biomarker of subclinical lung injury and may facilitate early identification and treatment of patients at risk of developing overt lung disease.

Early Detection of Respiratory Diseases in Calves by Use of an Ear-Attached Accelerometer.

Simple SummaryBovine respiratory disease is one of the most important diseases in group-housed calves worldwide, with impacts on calf welfare and farm economics. Early detection of the disease is important for the well-being of the animals and a targeted treatment. Therefore, tools for an automated monitoring of individual calves would be a breakthrough in health management. In this study, we used an ear-attached accelerometer to evaluate its potential for the early detection of behavioral changes related to respiratory disease in calves. Our result showed that accelerometers are able to detect changes in activity and lying times that can be used to predict respiratory disease before clinical diagnosis.Accelerometers (ACL) can identify behavioral and activity changes in calves. In the present study, we examined the association between bovine respiratory disease (BRD) and behavioral changes detected by an ear-tag based ACL system in weaned dairy calves. Accelerometer data were analyzed from 7 d before to 1 d after clinical diagnosis of BRD. All calves in the study (n = 508) were checked daily by an adapted University of Wisconsin Calf Scoring System. Calves with a score ≥ 4 and fever for at least two consecutive days were categorized as diseased (DIS). The day of clinical diagnosis of BRD was defined as d 0. The data analysis showed a significant difference in high active times between DIS and healthy control calves (CON), with CON showing more high active times on every day, except d −3. Diseased calves showed significantly more inactive times on d −4, −2, and 0, as well as longer lying times on d −5, −2, and +1. These results indicate the potential of the ACL to detect BRD prior to a clinical diagnosis in group-housed calves. Furthermore, in this study, we described the ‘normal’ behavior in 428 clinically healthy weaned dairy calves obtained by the ACL system.

X-ray dark-field chest imaging for detection and quantification of emphysema in patients with chronic obstructive pulmonary disease: a diagnostic accuracy study

SummaryBackgroundAlthough advanced medical imaging technologies give detailed diagnostic information, a low-dose, fast, and inexpensive option for early detection of respiratory diseases and follow-ups is still lacking. The novel method of x-ray dark-field chest imaging might fill this gap but has not yet been studied in living humans. Enabling the assessment of microstructural changes in lung parenchyma, this technique presents a more sensitive alternative to conventional chest x-rays, and yet requires only a fraction of the dose applied in CT. We studied the application of this technique to assess pulmonary emphysema in patients with chronic obstructive pulmonary disease (COPD).MethodsIn this diagnostic accuracy study, we designed and built a novel dark-field chest x-ray system (Technical University of Munich, Munich, Germany)—which is also capable of simultaneously acquiring a conventional thorax radiograph (7 s, 0·035 mSv effective dose). Patients who had undergone a medically indicated chest CT were recruited from the department of Radiology and Pneumology of our site (Klinikum rechts der Isar, Technical University of Munich, Munich, Germany). Patients with pulmonary pathologies, or conditions other than COPD, that might influence lung parenchyma were excluded. For patients with different disease stages of pulmonary emphysema, x-ray dark-field images and CT images were acquired and visually assessed by five readers. Pulmonary function tests (spirometry and body plethysmography) were performed for every patient and for a subgroup of patients the measurement of diffusion capacity was performed. Individual patient datasets were statistically evaluated using correlation testing, rank-based analysis of variance, and pair-wise post-hoc comparison.FindingsBetween October, 2018 and December, 2019 we enrolled 77 patients. Compared with CT-based parameters (quantitative emphysema ρ=–0·27, p=0·089 and visual emphysema ρ=–0·45, p=0·0028), the dark-field signal (ρ=0·62, p<0·0001) yields a stronger correlation with lung diffusion capacity in the evaluated cohort. Emphysema assessment based on dark-field chest x-ray features yields consistent conclusions with findings from visual CT image interpretation and shows improved diagnostic performance than conventional clinical tests characterising emphysema. Pair-wise comparison of corresponding test parameters between adjacent visual emphysema severity groups (CT-based, reference standard) showed higher effect sizes. The mean effect size over the group comparisons (absent–trace, trace–mild, mild–moderate, and moderate–confluent or advanced destructive visual emphysema grades) for the COPD assessment test score is 0·21, for forced expiratory volume in 1 s (FEV1)/functional vital capacity is 0·25, for FEV1% of predicted is 0·23, for residual volume % of predicted is 0·24, for CT emphysema index is 0·35, for dark-field signal homogeneity within lungs is 0·38, for dark-field signal texture within lungs is 0·38, and for dark-field-based emphysema severity is 0·42.InterpretationX-ray dark-field chest imaging allows the diagnosis of pulmonary emphysema in patients with COPD because this technique provides relevant information representing the structural condition of lung parenchyma. This technique might offer a low radiation dose alternative to CT in COPD and potentially other lung disorders.FundingEuropean Research Council, Deutsche Forschungsgemeinschaft, Royal Philips, and Karlsruhe Nano Micro Facility.

Respiratory Disease Occupational Biomonitoring Collaborative Project (ROBoCoP): A longitudinal pilot study and implementation research in the Parisian transport company

The ROBoCoP project is launched within the EU COST Action CA16113 “CliniMARK” aiming to increase the number of clinically validated biomarkers and focused on chronic obstructive pulmonary disease (COPD) biomarker development and validation. ROBoCoP encompasses two consecutive studies consisting of a pilot study followed by a field study. The pilot study is a longitudinal exposure assessment and biomarker study aiming at: 1-understanding the suitability of the candidate biomarkers in surveying populations at risk such as workers exposed to COPD causing agents; 2-determining the best sampling plan with respect to the half-life of the candidate biomarkers; 3-implementing and validating the sampling procedures and analytical methods; 4-selecting the best suitable biomarkers to be measured in the field. Each study participant is surveyed every day during the 6–8 h work-shifts for two consecutive weeks. The field study has an implementation research designe that enabled us to demonstrate the applicability of the standardized protocol for biomarker measurements in occupational settings while also assessing the biomarkers’ validity. ROBoCoP will focus on particulate matter (PM) exposure measurements, exposure biomarkers and a series of effect biomarkers, including markers of lipoperoxidation: 8-isoprostane, malondialdehyd in exhaled breath condensate (EBC) and urine, potential markers of nitrosative stress: NO2−, NO3− and formate anion in EBC; markers of DNA oxidation: 8-hydroxy-2’deoxyguanosine in EBC and urine, marker of genotoxicity: micronuclei in buccal cells, and oxidative potential in exhaled air (OPEA). OPEA appears particularly promising as a clinical biomarker for detecting COPD, and will be tested independently and as part of a biomarker panel. COPD diagnosis will be performed by an experienced occupational physician according to international diagnostic standards and confirmed by a pulmonologist.This research will include approximatively 300 underground subway workers randomly selected from the personnel registry of a large Parisian transport company. Underground subways are suggested as the most PM polluted urban transport environment. We believe this occupational exposure is relevant for biomonitoring of workers and early detection of respiratory diseases.

Using imagery and computer vision as remote monitoring methods for early detection of respiratory disease in pigs

Respiratory diseases in pigs impact the wellbeing of animals and increase the cost of production. One of the most appropriate approaches to minimizing these negative effects is the early detection of ill animals. The use of cameras coupled with computer-based techniques could assist the early detection of physiological changes in pigs when they are beginning to become ill and prior to exhibiting clinical signs. This study consisted of two experiments that aimed to (a) evaluate the use of computer-based techniques over RGB (red, green, and blue) and thermal infrared imagery to measure heart rate and respiration rate of pigs, and (b) to investigate whether eye-temperature, heart rate and respiration rate assessed remotely could be used to identify early signs of respiratory diseases in free-moving, and group-housed growing pigs in a commercial piggery. In the first experiment, the remotely-obtained heart rate and respiration rate were compared with the measures obtained with standard methods, showing positive correlations (r = 0.61 – 0.66; p < 0.05). In the second experiment, pigs were recorded by overhead cameras and the remotely-obtained physiological measures were analysed to identify whether physiological changes could be detected in sick pigs before clinical signs were observed. The changes in eye-temperature and heart rate remotely obtained showed clear differences between sick and healthy pigs two days before clinical signs were detected. While significant changes in respiration rate occurred the day before clinical signs of illness were identified. The results of the present study indicate the possible use of computer vision technique for constant animal monitoring and rapid detection of physiological changes related to illness in commercial pigs. Further research is recommended to continue the development, automatization, and commercial practicality of this novel technology.

Remotely Sensed Imagery for Early Detection of Respiratory Disease in Pigs: A Pilot Study.

Simple SummaryRespiratory disease in pigs causes suffering in infected animals and economic losses to producers. One of the most appropriate approaches to minimising these negative effects is by the early detection of infected animals. This pilot study aimed to use computer-based techniques to measure changes in temperature (eye and ear-base temperature), heart rate and respiration rate of pigs from thermal-infrared and conventional images. These measures, together with clinical observations, were obtained from pigs that were infected with Actinobacillus pleuropneumoniae (APP) and from pigs that were healthy. Infected pigs showed higher temperature and heart rate than healthy pigs across the period analysed. Respiration rate showed less difference between infected and healthy pigs. In addition, the biggest changes in these measures were recorded from six hours before the clinical observations identified sick animals. Results have highlighted that computer vision techniques can provide important and useable data regarding physiological changes that can indicate early signs of respiratory infection in pigs. This could aid the management of the disease, increasing the success of the treatment and decreasing the rate of severe cases and death.Respiratory diseases are a major problem in the pig industry worldwide. Due to the impact of these diseases, the early identification of infected herds is essential. Computer vision technology, using RGB (red, green and blue) and thermal infrared imagery, can assist the early detection of changes in animal physiology related to these and other diseases. This pilot study aimed to identify whether these techniques are a useful tool to detect early changes of eye and ear-base temperature, heart rate and respiration rate in pigs that were challenged with Actinobacillus pleuropneumoniae. Clinical observations and imagery were analysed, comparing data obtained from animals that showed some signs of illness with data from animals that showed no signs of ill health. Highly significant differences (p < 0.05) were observed between sick and healthy pigs in heart rate, eye and ear temperature, with higher heart rate and higher temperatures in sick pigs. The largest change in temperature and heart rate remotely measured was observed around 4–6 h before signs of clinical illness were observed by the skilled technicians. These data suggest that computer vision techniques could be a useful tool to detect indicators of disease before the symptoms can be observed by stock people, assisting the early detection and control of respiratory diseases in pigs, promoting further research to study the capability and possible uses of this technology for on farm monitoring and management.

A Combined Deep Learning GRU-Autoencoder for the Early Detection of Respiratory Disease in Pigs Using Multiple Environmental Sensors.

We designed and evaluated an assumption-free, deep learning-based methodology for animal health monitoring, specifically for the early detection of respiratory disease in growing pigs based on environmental sensor data. Two recurrent neural networks (RNNs), each comprising gated recurrent units (GRUs), were used to create an autoencoder (GRU-AE) into which environmental data, collected from a variety of sensors, was processed to detect anomalies. An autoencoder is a type of network trained to reconstruct the patterns it is fed as input. By training the GRU-AE using environmental data that did not lead to an occurrence of respiratory disease, data that did not fit the pattern of “healthy environmental data” had a greater reconstruction error. All reconstruction errors were labelled as either normal or anomalous using threshold-based anomaly detection optimised with particle swarm optimisation (PSO), from which alerts are raised. The results from the GRU-AE method outperformed state-of-the-art techniques, raising alerts when such predictions deviated from the actual observations. The results show that a change in the environment can result in occurrences of pigs showing symptoms of respiratory disease within 1–7 days, meaning that there is a period of time during which their keepers can act to mitigate the negative effect of respiratory diseases, such as porcine reproductive and respiratory syndrome (PRRS), a common and destructive disease endemic in pigs.

Upper respiratory tract disease in captive orangutans (Pongo sp.): prevalence in 20 European zoos and predisposing factors

Upper respiratory tract disease (URTD) is a significant cause of morbidity in captive orangutans (Pongo abelii, Pongo pygmaeus), and the pathogenesis is often unknown. The prevalence of respiratory disease in captive European orangutans (201 animals; 20 zoos) and possible predisposing factors were investigated. Bornean orangutans (P. pygmaeus) showed chronic respiratory signs significantly more often (13.8%) than Sumatran (P. abelii; 3.6%), and males (15.8%) were more often afflicted than females (3.9%). Hand-reared animals (21%) developed air sacculitis more often than parent-reared animals (5%). Diseased animals were more often genetically related to animals with respiratory diseases (93%) than to healthy animals (54%). None of the environmental conditions investigated had a significant effect on disease prevalence. Results suggest a higher importance of individual factors for the development of URTD than environmental conditions. Bornean, male and hand-reared orangutans and animals related to diseased animals need increased medical surveillance for early detection of respiratory disease.