Exploring machine learning for audio-based respiratory condition screening: A concise review of databases, methods, and open issues.
Ontology highlight
ABSTRACT:
SUBMITTER: Xia T
PROVIDER: S-EPMC9791302 | biostudies-literature | 2022 Nov
REPOSITORIES: biostudies-literature
Publications
Exploring machine learning for audio-based respiratory condition screening: A concise review of databases, methods, and open issues.
Xia Tong T Han Jing J Mascolo Cecilia C
Experimental biology and medicine (Maywood, N.J.) 20220816 22
Auscultation plays an important role in the clinic, and the research community has been exploring machine learning (ML) to enable remote and automatic auscultation for respiratory condition screening via sounds. To give the big picture of what is going on in this field, in this narrative review, we describe publicly available audio databases that can be used for experiments, illustrate the developed ML methods proposed to date, and flag some under-considered issues which still need attention. Co ...[more]
Similar Datasets
Project description:The use of computer-based automated approaches and improvements in lung sound recording techniques have made lung sound-based diagnostics even better and devoid of subjectivity errors. Using a computer to evaluate lung sound features more thoroughly with the use of analyzing changes in lung sound behavior, recording measurements, suppressing the presence of noise contaminations, and graphical representations are all made possible by computer-based lung sound analysis. This paper starts with a discussion of the need for this research area, providing an overview of the field and the motivations behind it. Following that, it details the survey methodology used in this work. It presents a discussion on the elements of sound-based lung disease classification using machine learning algorithms. This includes commonly prior considered datasets, feature extraction techniques, pre-processing methods, artifact removal methods, lung-heart sound separation, deep learning algorithms, and wavelet transform of lung audio signals. The study introduces studies that review lung screening including a summary table of these references and discusses the literature gaps in the existing studies. It is concluded that the use of sound-based machine learning in the classification of respiratory diseases has promising results. While we believe this material will prove valuable to physicians and researchers exploring sound-signal-based machine learning, large-scale investigations remain essential to solidify the findings and foster wider adoption within the medical community.
| S-EPMC10900411 | biostudies-literature
Project description:Combination therapy has shown an obvious efficacy on complex diseases and can greatly reduce the development of drug resistance. However, even with high-throughput screens, experimental methods are insufficient to explore novel drug combinations. In order to reduce the search space of drug combinations, there is an urgent need to develop more efficient computational methods to predict novel drug combinations. In recent decades, more and more machine learning (ML) algorithms have been applied to improve the predictive performance. The object of this study is to introduce and discuss the recent applications of ML methods and the widely used databases in drug combination prediction. In this study, we first describe the concept and controversy of synergism between drug combinations. Then, we investigate various publicly available data resources and tools for prediction tasks. Next, ML methods including classic ML and deep learning methods applied in drug combination prediction are introduced. Finally, we summarize the challenges to ML methods in prediction tasks and provide a discussion on future work.
| S-EPMC8769702 | biostudies-literature
Project description:BackgroundDrug toxicity does not affect patients equally; the toxicity may only exert in patients who possess certain attributes of susceptibility to specific drug properties (i.e., drug-host interaction). This concept is crucial for personalized drug safety but remains under-studied, primarily due to methodological challenges and limited data availability. By monitoring a large volume of adverse event reports in the postmarket stage, spontaneous adverse event reporting systems provide an unparalleled resource of information for adverse events and could be utilized to explore risk disparities of specific adverse events by age, sex, and other host factors. However, well-formulated statistical methods to formally address such risk disparities are currently lacking.MethodsIn this paper, we present a statistical framework to explore spontaneous adverse event reporting databases for drug-host interactions and detect risk disparities in adverse drug events by various host factors, adapting methods for safety signal detection. We proposed four different methods, including likelihood ratio test, normal approximation test, and two tests using subgroup ratios. We applied our proposed methods to simulated data and Food and Drug Administration (FDA) Adverse Event Reporting Systems (FAERS) and explored sex-/age-disparities in reported liver events associated with specific drug classes.ResultsThe simulation result demonstrates that two tests (likelihood ratio, normal approximation) can detect disparities in adverse drug events associated with host factors while controlling the family wise error rate. Application to real data on drug liver toxicity shows that the proposed method can be used to detect drugs with unusually high level of disparity regarding a host factor (sex or age) for liver toxicity or to determine whether an adverse event demonstrates a significant unbalance regarding the host factor relative to other events for the drug.ConclusionThough spontaneous adverse event reporting databases require careful data processing and inference, the sheer size of the databases with diverse data from different countries provides unique resources for exploring various questions for drug safety that are otherwise impossible to address. Our proposed methods can be used to facilitate future investigation on drug-host interactions in drug toxicity using a large number of reported adverse events.
| S-EPMC10041785 | biostudies-literature
Project description:Passive acoustic monitoring is used widely in ecology, biodiversity, and conservation studies. Data sets collected via acoustic monitoring are often extremely large and built to be processed automatically using artificial intelligence and machine learning models, which aim to replicate the work of domain experts. These models, being supervised learning algorithms, need to be trained on high quality annotations produced by experts. Since the experts are often resource-limited, a cost-effective process for annotating audio is needed to get maximal use out of the data. We present an open-source interactive audio data annotation tool, NEAL (Nature+Energy Audio Labeller). Built using R and the associated Shiny framework, the tool provides a reactive environment where users can quickly annotate audio files and adjust settings that automatically change the corresponding elements of the user interface. The app has been designed with the goal of having both expert birders and citizen scientists contribute to acoustic annotation projects. The popularity and flexibility of R programming in bioacoustics means that the Shiny app can be modified for other bird labelling data sets, or even to generic audio labelling tasks. We demonstrate the app by labelling data collected from wind farm sites across Ireland.
| S-EPMC10461540 | biostudies-literature
Project description:Systematic condition and analytical surveys were carried out on Zoe Leonard’s (b. 1961) Mouth Open, Teeth Showing 2000, an installation artwork in Tate’s collection consisting of 162 children’s dolls. The dolls were manufactured at various points within the 20th century and encompass several potentially problematic synthetic polymers found in modern and contemporary museum collections. To explore the doll materials and conservation condition, a multi-analytical approach was used to identify key synthetic polymer types and additives present, including portable and bench analytical techniques. Challenging degradation phenomena associated with different types of doll have been discussed and related to their material composition, which has helped our understanding of the conservation challenges inherent to this contemporary artwork.
| S-EPMC9824361 | biostudies-literature
Project description:The identification of interactions between drugs/compounds and their targets is crucial for the development of new drugs. In vitro screening experiments (i.e. bioassays) are frequently used for this purpose; however, experimental approaches are insufficient to explore novel drug-target interactions, mainly because of feasibility problems, as they are labour intensive, costly and time consuming. A computational field known as 'virtual screening' (VS) has emerged in the past decades to aid experimental drug discovery studies by statistically estimating unknown bio-interactions between compounds and biological targets. These methods use the physico-chemical and structural properties of compounds and/or target proteins along with the experimentally verified bio-interaction information to generate predictive models. Lately, sophisticated machine learning techniques are applied in VS to elevate the predictive performance. The objective of this study is to examine and discuss the recent applications of machine learning techniques in VS, including deep learning, which became highly popular after giving rise to epochal developments in the fields of computer vision and natural language processing. The past 3 years have witnessed an unprecedented amount of research studies considering the application of deep learning in biomedicine, including computational drug discovery. In this review, we first describe the main instruments of VS methods, including compound and protein features (i.e. representations and descriptors), frequently used libraries and toolkits for VS, bioactivity databases and gold-standard data sets for system training and benchmarking. We subsequently review recent VS studies with a strong emphasis on deep learning applications. Finally, we discuss the present state of the field, including the current challenges and suggest future directions. We believe that this survey will provide insight to the researchers working in the field of computational drug discovery in terms of comprehending and developing novel bio-prediction methods.
| S-EPMC6917215 | biostudies-literature
Project description:Metagenomic sequencing has revolutionized our understanding of microbiology. While metagenomic tools and approaches have been extensively evaluated and benchmarked, far less attention has been given to the reference sequence database used in metagenomic classification. Issues with reference sequence databases are pervasive. Database contamination is the most recognized issue in the literature; however, it remains relatively unmitigated in most analyses. Other common issues with reference sequence databases include taxonomic errors, inappropriate inclusion and exclusion criteria, and sequence content errors. This review covers ten common issues with reference sequence databases and the potential downstream consequences of these issues. Mitigation measures are discussed for each issue, including bioinformatic tools and database curation strategies. Together, these strategies present a path towards more accurate, reproducible and translatable metagenomic sequencing.
| S-EPMC10978663 | biostudies-literature
Project description:Mucopolysaccharidosis I-Hurler (MPS I-H) is the most severe form of a metabolic genetic disease caused by mutations of IDUA gene encoding the lysosomal α-L-iduronidase enzyme. MPS I-H is a rare, life-threatening disease, evolving in multisystem morbidity including progressive neurological disease, upper airway obstruction, skeletal deformity and cardiomyopathy. Allogeneic hematopoietic stem cell transplantation (HSCT) is currently the gold standard for the treatment of MPS I-H in patients diagnosed and treated before 2-2.5 years of age, having a high rate of success. Beyond the child's age, other factors influence the probability of treatment success, including the selection of patients, of graft source and the donor type employed. Enzyme replacement therapy (ERT) with human recombinant laronidase has also been demonstrated to be effective in ameliorating the clinical conditions of pre-transplant MPS I-H patients and in improving HSCT outcome, by peri-transplant co-administration. Nevertheless the long-term clinical outcome even after successful HSCT varies considerably, with a persisting residual disease burden. Other strategies must then be considered to improve the outcome of these patients: one is to pursue early pre-symptomatic diagnosis through newborn screening and another one is the identification of novel treatments. In this perspective, even though newborn screening can be envisaged as a future attractive perspective, presently the best path to be pursued embraces an improved awareness of signs and symptoms of the disorder by primary care providers and pediatricians, in order for the patients' timely referral to a qualified reference center. Furthermore, sensitive new biochemical markers must be identified to better define the clinical severity of the disease at birth, to support clinical judgement during the follow-up and to compare the effects of the different therapies. A prolonged neuropsychological follow-up of post-transplant cognitive development of children and residual disease burden is needed. In this perspective, the reference center must guarantee a multidisciplinary follow-up with an expert team. Diagnostic and interventional protocols of reference centers should be standardized whenever possible to allow comparison of clinical data and evaluation of results. This review will focus on all these critical issues related to the management of MPS I-H.
| S-EPMC5472858 | biostudies-literature
Project description:IntroductionThe main goal of this work is to investigate the feasibility of estimating an anatomical index that can be used as an Alzheimer's disease (AD) risk factor in the Women's Health Initiative Magnetic Resonance Imaging Study (WHIMS-MRI) using MRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), a well-characterized imaging database of AD patients and cognitively normal subjects. We called this index AD Pattern Similarity (AD-PS) scores. To demonstrate the construct validity of the scores, we investigated their associations with several AD risk factors. The ADNI and WHIMS imaging databases were collected with different goals, populations and data acquisition protocols: it is important to demonstrate that the approach to estimating AD-PS scores can bridge these differences.MethodsMRI data from both studies were processed using high-dimensional warping methods. High-dimensional classifiers were then estimated using the ADNI MRI data. Next, the classifiers were applied to baseline and follow-up WHIMS-MRI GM data to generate the GM AD-PS scores. To study the validity of the scores we investigated associations between GM AD-PS scores at baseline (Scan 1) and their longitudinal changes (Scan 2 -Scan 1) with: 1) age, cognitive scores, white matter small vessel ischemic disease (WM SVID) volume at baseline and 2) age, cognitive scores, WM SVID volume longitudinal changes respectively. In addition, we investigated their associations with time until classification of independently adjudicated status in WHIMS-MRI.ResultsHigher GM AD-PS scores from WHIMS-MRI baseline data were associated with older age, lower cognitive scores, and higher WM SVID volume. Longitudinal changes in GM AD-PS scores (Scan 2 - Scan 1) were also associated with age and changes in WM SVID volumes and cognitive test scores. Increases in the GM AD-PS scores predicted decreases in cognitive scores and increases in WM SVID volume. GM AD-PS scores and their longitudinal changes also were associated with time until classification of cognitive impairment. Finally, receiver operating characteristic curves showed that baseline GM AD-PS scores of cognitively normal participants carried information about future cognitive status determined during follow-up.DiscussionWe applied a high-dimensional machine learning approach to estimate a novel AD risk factor for WHIMS-MRI study participants using ADNI data. The GM AD-PS scores showed strong associations with incident cognitive impairment and cross-sectional and longitudinal associations with age, cognitive function, cognitive status and WM SVID volume lending support to the ongoing validation of the GM AD-PS score.
| S-EPMC6457113 | biostudies-literature
Project description:BackgroundPhysician-to-physician teleconsultation has increasingly played an essential role in delivering optimum health care services, particularly in orthopedic practice. In this study, the usability of a smartphone app for teleconsultation among orthopedic specialists was investigated to explore issues informing further recommendations for improvement in the following iterations.ObjectiveThis study aimed to explore usability issues emerging from users' interactions with MEDIC app, a smartphone-based patient-centered physician-to-physician teleconsultation system.MethodsFive attending physicians in the Department of Orthopedics in a large medical school in Bangkok, Thailand, were recruited and asked to perform 5 evaluation tasks, namely, group formation, patient registration, clinical data capturing, case record form creation, and teleconsultation. In addition, one expert user was recruited as the control participant. Think aloud was adopted while performing the tasks. Semistructured interviews were conducted after each task and prior to the exit. Quantitative and qualitative measures were used to identify usability issues in 7 domains based on the People At the Centre of Mobile Application Development model: effectiveness, efficiency, satisfaction, learnability, memorability, error, and cognitive load.ResultsSeveral measures indicate various aspects of usability of the app, including completion rates, time to completion, number of clicks, number of screens, errors, incidents where participants were unable to perform tasks, which had previously been completed, and perceived task difficulty. Major and critical usability issues based on participant feedback were rooted from the limitation of screen size and resolution. Errors in data input (eg, typing errors, miscalculation), action failures, and misinterpretation of data (ie, radiography) were the most critical and common issues found in this study. A few participants did not complete the assigned tasks mostly owing to the navigation design and misreading/misunderstanding icons. However, the novice users were quite positive that they would be able to become familiar with the app in a short period of time.ConclusionsThe usability issues in physician-to-physician teleconsultation systems in smartphones, in general, are derived from the limitations of smartphones and their operating systems. Although some recommendations were devised to handle these usability issues, usability evaluation for additional development should still be further investigated.
| S-EPMC8726029 | biostudies-literature