Prediction of hearing loss among the noise-exposed workers in a steel factory using artificial intelligence approach

Machine learning-based longitudinal prediction for GJB2-related sensorineural hearing loss

BACKGROUND

Recessive GJB2 variants, the most common genetic cause of hearing loss, may contribute to progressive sensorineural hearing loss (SNHL). The aim of this study is to build a realistic predictive model for GJB2-related SNHL using machine learning to enable personalized medical planning for timely intervention.

METHOD

Patients with SNHL with confirmed biallelic GJB2 variants in a nationwide cohort between 2005 and 2022 were included. Different data preprocessing protocols and computational algorithms were combined to construct a prediction model. We randomly divided the dataset into training, validation, and test sets at a ratio of 72:8:20, and repeated this process ten times to obtain an average result. The performance of the models was evaluated using the mean absolute error (MAE), which refers to the discrepancy between the predicted and actual hearing thresholds.

RESULTS

We enrolled 449 patients with 2184 audiograms available for deep learning analysis. SNHL progression was identified in all models and was independent of age, sex, and genotype. The average hearing progression rate was 0.61 dB HL per year. The best MAE for linear regression, multilayer perceptron, long short-term memory, and attention model were 4.42, 4.38, 4.34, and 4.76 dB HL, respectively. The long short-term memory model performed best with an average MAE of 4.34 dB HL and acceptable accuracy for up to 4 years.

CONCLUSIONS

We have developed a prognostic model that uses machine learning to approximate realistic hearing progression in GJB2-related SNHL, allowing for the design of individualized medical plans, such as recommending the optimal follow-up interval for this population.

Utilizing Artificial Neural Networks for Establishing Hearing-Loss Predicting Models Based on a Longitudinal Dataset and Their Implications for Managing the Hearing Conservation Program

Background

Though the artificial neural network (ANN) technique has been used to predict noise-induced hearing loss (NIHL), the established prediction models have primarily relied on cross-sectional datasets, and hence, they may not comprehensively capture the chronic nature of NIHL as a disease linked to long-term noise exposure among workers.

Methods

A comprehensive dataset was utilized, encompassing eight-year longitudinal personal hearing threshold levels (HTLs) as well as information on seven personal variables and two environmental variables to establish NIHL predicting models through the ANN technique. Three subdatasets were extracted from the afirementioned comprehensive dataset to assess the advantages of the present study in NIHL predictions.

Results

The dataset was gathered from 170 workers employed in a steel-making industry, with a median cumulative noise exposure and HTL of 88.40 dBA-year and 19.58 dB, respectively. Utilizing the longitudinal dataset demonstrated superior prediction capabilities compared to cross-sectional datasets. Incorporating the more comprehensive dataset led to improved NIHL predictions, particularly when considering variables such as noise pattern and use of personal protective equipment. Despite fluctuations observed in the measured HTLs, the ANN predicting models consistently revealed a discernible trend.

Conclusions

A consistent correlation was observed between the measured HTLs and the results obtained from the predicting models. However, it is essential to exercise caution when utilizing the model-predicted NIHLs for individual workers due to inherent personal fluctuations in HTLs. Nonetheless, these ANN models can serve as a valuable reference for the industry in effectively managing its hearing conservation program.

Risk analysis of noise-induced hearing loss of workers in the automobile manufacturing industries based on back-propagation neural network model: a cross-sectional study in Han Chinese population

OBJECTIVES

This study aims to predict the risk of noise-induced hearing loss (NIHL) through a back-propagation neural network (BPNN) model. It provides an early, simple and accurate prediction method for NIHL.

DESIGN

Population based, a cross sectional study.

SETTING

Han, China.

PARTICIPANTS

This study selected 3266 Han male workers from three automobile manufacturing industries.

PRIMARY OUTCOME MEASURES

Information including personal life habits, occupational health test information and occupational exposure history were collected and predictive factors of NIHL were screened from these workers. BPNN and logistic regression models were constructed using these predictors.

RESULTS

The input variables of BPNN model were 20, 16 and 21 important factors screened by univariate, stepwise and lasso-logistic regression. When the BPNN model was applied to the test set, it was found to have a sensitivity (TPR) of 83.33%, a specificity (TNR) of 85.92%, an accuracy (ACC) of 85.51%, a positive predictive value (PPV) of 52.85%, a negative predictive value of 96.46% and area under the receiver operating curve (AUC) is: 0.926 (95% CI: 0.891 to 0.961), which demonstrated the better overall properties than univariate-logistic regression modelling (AUC: 0.715) (95% CI: 0.652 to 0.777). The BPNN model has better predictive performance against NIHL than the stepwise-logistic and lasso-logistic regression model in terms of TPR, TNR, ACC, PPV and NPV (p<0.05); the area under the receiver operating characteristics curve of NIHL is also higher than that of the stepwise and lasso-logistic regression model (p<0.05). It was a relatively important factor in NIHL to find cumulative noise exposure, auditory system symptoms, age, listening to music or watching video with headphones, exposure to high temperature and noise exposure time in the trained BPNN model.

CONCLUSIONS

The BPNN model was a valuable tool in dealing with the occupational risk prediction problem of NIHL. It can be used to predict the risk of an individual NIHL.

A risk model for occupational noise-induced hearing loss in workers

BACKGROUND

Occupational hearing loss is one of the most common work-related diseases with various risk factors and considerable negative impacts on both physical and mental well-being of affected workers. Occupational noise-induced hearing loss (ONIHL) has a complex interaction with personal, environmental and occupational factors.

OBJECTIVE

This study aimed to develop a risk model for ONIHL in workers by identifying risk factors and their interactions.

METHODS

The subjects were 605 males in an industrial factory in Arak, Iran. The study took place between 2022 and 2023. The sociodemographic and occupational characteristics were collected by a health technician using questionnaires and medical records. Hearing status was assessed using audiometry by a qualified audiologist. Hearing loss was analyzed by univariate logistic analysis including age, smoking, medical history, type of occupation, and some workplace hazards. The risk model was generated by logistic regression.

RESULTS

Hearing loss in the participants was 44.13% (n = 267). In univariate logistic analysis, age (OR: 2.93,95% CI: 1.848-4.656), smoking (OR: 1.80, 95% CI: 1.224-2.655), work experience (OR: 1.06, 95% CI: 1.016-1.107), previous exposure to noise (OR: 1.60, 95% CI: 1.112-2.312) or vibration (OR: 1.68, 95% CI: 1.150-2.475) and type of occupation (OR: 2.126, 95% CI: 1.055-4.285) were associated with an increased risk of ONIHL (P < 0.05).

CONCLUSION

It was found that vibration exposure, work experience, previous noise exposure, type of occupation as well as age and smoking significantly affected the likelihood of developing ONIHL. This risk model could help management to prevent ONIHL and enhance application-oriented research on the condition.

Explainable machine learning reveals the relationship between hearing thresholds and speech-in-noise recognition in listeners with normal audiograms

Some individuals complain of listening-in-noise difficulty despite having a normal audiogram. In this study, machine learning is applied to examine the extent to which hearing thresholds can predict speech-in-noise recognition among normal-hearing individuals. The specific goals were to (1) compare the performance of one standard (GAM, generalized additive model) and four machine learning models (ANN, artificial neural network; DNN, deep neural network; RF, random forest; XGBoost; eXtreme gradient boosting), and (2) examine the relative contribution of individual audiometric frequencies and demographic variables in predicting speech-in-noise recognition. Archival data included thresholds (0.25-16 kHz) and speech recognition thresholds (SRTs) from listeners with clinically normal audiograms (n = 764 participants or 1528 ears; age, 4-38 years old). Among the machine learning models, XGBoost performed significantly better than other methods (mean absolute error; MAE = 1.62 dB). ANN and RF yielded similar performances (MAE = 1.68 and 1.67 dB, respectively), whereas, surprisingly, DNN showed relatively poorer performance (MAE = 1.94 dB). The MAE for GAM was 1.61 dB. SHapley Additive exPlanations revealed that age, thresholds at 16 kHz, 12.5 kHz, etc., on the order of importance, contributed to SRT. These results suggest the importance of hearing in the extended high frequencies for predicting speech-in-noise recognition in listeners with normal audiograms.

The use of artificial intelligence for delivery of essential health services across WHO regions: a scoping review

Background

Artificial intelligence (AI) is a broad outlet of computer science aimed at constructing machines capable of simulating and performing tasks usually done by human beings. The aim of this scoping review is to map existing evidence on the use of AI in the delivery of medical care.

Methods

We searched PubMed and Scopus in March 2022, screened identified records for eligibility, assessed full texts of potentially eligible publications, and extracted data from included studies in duplicate, resolving differences through discussion, arbitration, and consensus. We then conducted a narrative synthesis of extracted data.

Results

Several AI methods have been used to detect, diagnose, classify, manage, treat, and monitor the prognosis of various health issues. These AI models have been used in various health conditions, including communicable diseases, non-communicable diseases, and mental health.

Conclusions

Presently available evidence shows that AI models, predominantly deep learning, and machine learning, can significantly advance medical care delivery regarding the detection, diagnosis, management, and monitoring the prognosis of different illnesses.

Proposal of Combined Noise and Hand-Arm Vibration Index for Occupational Exposure: Application to a Study Case in the Olive Sector

In many production and industrial sectors, workers are exposed to noise and hand-arm vibrations (HAV). European directives have established the maximum limit values or exposure action values for noise and vibration independently. However, in many cases, workers who endure hand-arm vibration also receive high noise levels. This research suggests a procedure to aid the establishment of precautionary measures for workers with simultaneous exposure to both physical agents. This procedure defines a combined index based on the energy doses for both noise and HAV. From this combined index, the suggested methodology allows a recommended exposure time for workers with simultaneous noise and HAV exposure to be calculated. This methodology can be adapted to tackle the relative importance assigned to both agents according to the safety manager and new knowledge on combined health effects. To test this method, a measurement campaign under real working conditions was conducted with workers from the olive fruit-harvesting sector, where a variety of hand-held machinery is used. The results of the study case show that the suggested procedure can obtain reliable exposure time recommendations for simultaneous noise and HAV exposures and is therefore a useful tool for establishing prevention measures.

Prediction of hearing recovery in unilateral sudden sensorineural hearing loss using artificial intelligence

Despite the significance of predicting the prognosis of idiopathic sudden sensorineural hearing loss (ISSNHL), no predictive models have been established. This study used artificial intelligence to develop prognosis models to predict recovery from ISSNHL. We retrospectively reviewed the medical data of 453 patients with ISSNHL (men, 220; women, 233; mean age, 50.3 years) who underwent treatment at a tertiary hospital between January 2021 and December 2019 and were followed up after 1 month. According to Siegel's criteria, 203 patients recovered in 1 month. Demographic characteristics, clinical and laboratory data, and pure-tone audiometry were analyzed. Logistic regression (baseline), a support vector machine, extreme gradient boosting, a light gradient boosting machine, and multilayer perceptron were used. The outcomes were the area under the receiver operating characteristic curve (AUROC) primarily, area under the precision-recall curve, Brier score, balanced accuracy, and F1 score. The light gradient boosting machine model had the best AUROC and balanced accuracy. Together with multilayer perceptron, it was also significantly superior to logistic regression in terms of AUROC. Using the SHapley Additive exPlanation method, we found that the initial audiogram shape is the most important prognostic factor. Machine/deep learning methods were successfully established to predict the prognosis of ISSNHL.

Prediction of absenteeism in public schools teachers with machine learning

OBJECTIVE

To predict the risk of absence from work due to morbidities of teachers working in early childhood education in the municipal public schools, using machine learning algorithms.

METHODS

This is a cross-sectional study using secondary, public and anonymous data from the Relação Anual de Informações Sociais, selecting early childhood education teachers who worked in the municipal public schools of the state of São Paulo between 2014 and 2018 (n = 174,294). Data on the average number of students per class and number of inhabitants in the municipality were also linked. The data were separated into training and testing, using records from 2014 to 2016 (n = 103,357) to train five predictive models, and data from 2017 to 2018 (n = 70,937) to test their performance in new data. The predictive performance of the algorithms was evaluated using the value of the area under the ROC curve (AUROC).

RESULTS

All five algorithms tested showed an area under the curve above 0.76. The algorithm with the best predictive performance (artificial neural networks) achieved 0.79 of area under the curve, with accuracy of 71.52%, sensitivity of 72.86%, specificity of 70.52%, and kappa of 0.427 in the test data.

CONCLUSION

It is possible to predict cases of sickness absence in teachers of public schools with machine learning using public data. The best algorithm showed a better result of the area under the curve when compared with the reference model (logistic regression). The algorithms can contribute to more assertive predictions in the public health and worker health areas, allowing to monitor and help prevent the absence of these workers due to morbidity.

Improving classification based on physical surface tension-neural net for the prediction of psychosocial-risk level in public school teachers

BACKGROUND

Psychosocial risks, also present in educational processes, are stress factors particularly critical in state-schools, affecting the efficacy, stress, and job satisfaction of the teachers. This study proposes an intelligent algorithm to improve the prediction of psychosocial risk, as a tool for the generation of health and risk prevention assistance programs.

METHODS

The proposed approach, Physical Surface Tension-Neural Net (PST-NN), applied the theory of superficial tension in liquids to an artificial neural network (ANN), in order to model four risk levels (low, medium, high and very high psychosocial risk). The model was trained and tested using the results of tests for measurement of the psychosocial risk levels of 5,443 teachers. Psychosocial, and also physiological and musculoskeletal symptoms, factors were included as inputs of the model. The classification efficiency of the PST-NN approach was evaluated by using the sensitivity, specificity, accuracy and ROC curve metrics, and compared against other techniques as the Decision Tree model, Naïve Bayes, ANN, Support Vector Machines, Robust Linear Regression and the Logistic Regression Model.

RESULTS

The modification of the ANN model, by the adaptation of a layer that includes concepts related to the theory of physical surface tension, improved the separation of the subjects according to the risk level group, as a function of the mass and perimeter outputs. Indeed, the PST-NN model showed better performance to classify psychosocial risk level on state-school teachers than the linear, probabilistic and logistic models included in this study, obtaining an average accuracy value of 97.31%.

CONCLUSIONS

The introduction of physical models, such as the physical surface tension, can improve the classification performance of ANN. Particularly, the PST-NN model can be used to predict and classify psychosocial risk levels among state-school teachers at work. This model could help to early identification of psychosocial risk and to the development of programs to prevent it.

A risk model and nomogram for high-frequency hearing loss in noise-exposed workers

Background

High-frequency hearing loss is a significant occupational health concern in many countries, and early identification can be effective for preventing hearing loss. The study aims to construct and validate a risk model for HFHL, and develop a nomogram for predicting the individual risk in noise-exposed workers.

Methods

The current research used archival data from the National Key Occupational Diseases Survey-Sichuan conducted in China from 2014 to 2017. A total of 32,121 noise-exposed workers completed the survey, of whom 80% workers (n = 25,732) comprised the training cohort for risk model development and 20% workers (n = 6389) constituted the validation cohort for model validation. The risk model and nomogram were constructed using binary logistic models. The effectiveness and calibration of the model were evaluated with the receiver operating characteristic curve and calibration plots, respectively.

Results

A total of 10.06% of noise-exposed workers had HFHL. Age (OR = 1.09, 95% CI: 1.083–1.104), male sex (OR = 3.25, 95% CI: 2.85–3.702), noise exposure duration (NED) (OR = 1.15, 95% CI: 1.093–1.201), and a history of working in manufacturing (OR = 1.50, 95% CI: 1.314–1.713), construction (OR = 2.29, 95% CI: 1.531–3.421), mining (OR = 2.63, 95% CI: 2.238–3.081), or for a private-owned enterprise (POE) (OR = 1.33, 95% CI: 1.202–1.476) were associated with an increased risk of HFHL (P < 0.05).

Conclusions

The risk model and nomogram for HFHL can be used in application-oriented research on the prevention and management of HFHL in workplaces with high levels of noise exposure.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-021-10730-y.

Contributions and limitations of using machine learning to predict noise-induced hearing loss

Purpose

Noise-induced hearing loss (NIHL) is a global issue that impacts people’s life and health. The current review aims to clarify the contributions and limitations of applying machine learning (ML) to predict NIHL by analyzing the performance of different ML techniques and the procedure of model construction.

Methods

The authors searched PubMed, EMBASE and Scopus on November 26, 2020.

Results

Eight studies were recruited in the current review following defined inclusion and exclusion criteria. Sample size in the selected studies ranged between 150 and 10,567. The most popular models were artificial neural networks (n = 4), random forests (n = 3) and support vector machines (n = 3). Features mostly correlated with NIHL and used in the models were: age (n = 6), duration of noise exposure (n = 5) and noise exposure level (n = 4). Five included studies used either split-sample validation (n = 3) or ten-fold cross-validation (n = 2). Assessment of accuracy ranged in value from 75.3% to 99% with a low prediction error/root-mean-square error in 3 studies. Only 2 studies measured discrimination risk using the receiver operating characteristic (ROC) curve and/or the area under ROC curve.

Conclusion

In spite of high accuracy and low prediction error of machine learning models, some improvement can be expected from larger sample sizes, multiple algorithm use, completed reports of model construction and the sufficient evaluation of calibration and discrimination risk.

Comparison of data mining algorithms for sex determination based on mastoid process measurements using cone-beam computed tomography

Purpose

The mastoid region is ideal for studying sexual dimorphism due to its anatomical position at the base of the skull. This study aimed to determine sex in the Iranian population based on measurements of the mastoid process using different data mining algorithms.

Materials and Methods

This retrospective study was conducted on 190 3-dimensional cone-beam computed tomographic (CBCT) images of 105 women and 85 men between the ages of 18 and 70 years. On each CBCT scan, the following 9 landmarks were measured: the distance between the porion and the mastoidale; the mastoid length, height, and width; the distance between the mastoidale and the mastoid incision; the intermastoid distance (IMD); the distance between the lowest point of the mastoid triangle and the most prominent convex surface of the mastoid (MF); the distance between the most prominent convex mastoid point (IMSLD); and the intersecting angle drawn from the most prominent right and left mastoid point (MMCA). Several predictive models were constructed and their accuracy was compared using cross-validation.

Results

The results of the t-test revealed a statistically significant difference between the sexes in all variables except MF and MMCA. The random forest model, with an accuracy of 97.0%, had the best performance in predicting sex. The IMSLD and IMD made the largest contributions to predicting sex, while the MMCA variable had the least significant role.

Conclusion

These results show the possibility of developing an accurate tool using data mining algorithms for sex determination in the forensic framework.

Explanation and prediction of accidents using the path analysis approach in industrial units: The effect of safety performance and climate

BACKGROUND

The safety climate in an organization depends on people's understanding of the safety policies and procedures, as well as the value, importance, and priority of safety in the workplace.

OBJECTIVE

This study aimed to describe and predict accidents using the path analysis model (PAM) in industrial units though the analysis of the effect of safety performance and climate.

METHODS

This cross-sectional study was conducted on 294 workers in industrial units in Hamadan, a province in the western part of Iran. The data on safety performance and climate was collected using a questionnaire. The first part of the questionnaire was a short version inventory (with 25 items on safety climate) that was used to assess five variables of management commitment, supportive environment, training, personal safety prioritization, and perceived work pressure. Moreover, the safety performance was measured using 10 items on safety rules and participation. The PAM was used to describe the effects of safety climate and performance on accidents.

RESULTS

The results showed that the safety climate had the strongest negative impact on work pressure and safety compliance toward accident, followed by safety participation, and quality of training. Moreover, the negative influence of safety climate on accident was mainly mediated by two variables: work pressure and safety participation toward accident. The work pressure had the strongest indirect and total influence on accidents. However, none of the variables had a direct effect on accidents. Training was the most important direct cause of promoting personal safety priority. The safety compliance was more effective than safety participation in reducing accidents rates.

CONCLUSIONS

Therefore, it seems that perceived work pressure has an indirect effect on accidents which is mediated by other variables, mainly personal safety priority and safety performance.

Artificial Intelligence Applications in Otology: A State of the Art Review

Objective

Recent advances in artificial intelligence (AI) are driving innovative new health care solutions. We aim to review the state of the art of AI in otology and provide a discussion of work underway, current limitations, and future directions.

Data Sources

Two comprehensive databases, MEDLINE and EMBASE, were mined using a directed search strategy to identify all articles that applied AI to otology.

Review Methods

An initial abstract and title screening was completed. Exclusion criteria included nonavailable abstract and full text, language, and nonrelevance. References of included studies and relevant review articles were cross-checked to identify additional studies.

Conclusion

The database search identified 1374 articles. Abstract and title screening resulted in full-text retrieval of 96 articles. A total of N = 38 articles were retained. Applications of AI technologies involved the optimization of hearing aid technology (n = 5; 13% of all articles), speech enhancement technologies (n = 4; 11%), diagnosis and management of vestibular disorders (n = 11; 29%), prediction of sensorineural hearing loss outcomes (n = 9; 24%), interpretation of automatic brainstem responses (n = 5; 13%), and imaging modalities and image-processing techniques (n = 4; 10%). Publication counts of the included articles from each decade demonstrated a marked increase in interest in AI in recent years.

Implications for Practice

This review highlights several applications of AI that otologists and otolaryngologists alike should be aware of given the possibility of implementation in mainstream clinical practice. Although there remain significant ethical and regulatory challenges, AI powered systems offer great potential to shape how healthcare systems of the future operate and clinicians are key stakeholders in this process.

Recent advances in hearing conservation programmes: A systematic review

Background

Current evidence from low- and middle-income (LAMI) countries, such as South Africa, indicates that occupational noise-induced hearing loss (ONIHL) continues to be a health and safety challenge for the mining industry. There is also evidence of hearing conservation programmes (HCPs) being implemented with limited success.

Objectives

The aim of this study was to explore and document current evidence reflecting recent advances in HCPs in order to identify gaps within the South African HCPs.

Method

A systematic literature review was conducted in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis. Electronic databases including Sage, Science Direct, PubMed, Scopus MEDLINE, ProQuest and Google Scholar were searched for potential studies published in English between 2010 and 2019 reporting on recent advances in HCPs within the mining industry.

Results

The study findings revealed a number of important recent advances internationally, which require deliberation for possible implementation within the South African HCPs context. These advances have been presented under seven themes: (1) the use of metrics, (2) pharmacological interventions and hair cell regeneration, (3) artificial neural network, (4) audiology assessment measures, (5) noise monitoring advances, (6) conceptual approaches to HCPs and (7) buying quiet.

Conclusion

The study findings raise important advances that may have significant implications for HCPs in LAMI countries where ONIHL remains a highly prevalent occupational health challenge. Establishing feasibility and efficacy of these advances in these contexts to ensure contextual relevance and responsiveness is one of the recommendations to facilitate the success of HCPs targets.

Diagnosing thyroid disorders: Comparison of logistic regression and neural network models

BACKGROUND

The main goal of this study was to diagnose the two most common thyroid disorders, namely, hyperthyroidism and hypothyroidism, based on multinomial logistic regression and neural network models. In addition, the study evaluated the predictive ability of laboratory tests against the individual clinical symptoms score.

MATERIALS AND METHODS

In this study, the data from patients with thyroid dysfunction who referred to Imam Khomeini Clinic and Shahid Beheshti Hospital in Hamadan were collected. The data contained 310 subjects in one of three classes-euthyroid, hyperthyroidism, and hypothyroidism. Collected variables included demographics and symptoms of hypothyroidism and hyperthyroidism, as well as laboratory tests. To compare the predictive ability of the clinical signs and laboratory tests, different multinomial logistic regression and neural network models were fitted to the data. These models were compared in terms of the mean of the accuracy and area under the curve (AUC).

RESULTS

The results showed better performance of neural network model than multinomial logistic regression in all cases. The best predictive performance for logistic regression (with a mean accuracy of 91.4%) and neural network models (with a mean accuracy of 96.3%) was when all variables were included in the model. In addition, the predictive performance of two models based on symptomatic variables was superior to laboratory variables.

CONCLUSIONS

Both neural network and logistic regression models have a high predictive ability to diagnose thyroid disorder, although neural network performance is better than logistic regression. In addition, as achieving less error prediction model has always been a matter of concern for researchers in the field of disease diagnosis, predictive nonparametric techniques, such as neural networks, provide new opportunities to obtain more accurate predictions in the field of medical research.

Machine Learning Models for the Hearing Impairment Prediction in Workers Exposed to Complex Industrial Noise: A Pilot Study

Objectives:

To demonstrate the feasibility of developing machine learning models for the prediction of hearing impairment in humans exposed to complex non-Gaussian industrial noise.

Design:

Audiometric and noise exposure data were collected on a population of screened workers (N = 1,113) from 17 factories located in Zhejiang province, China. All the subjects were exposed to complex noise. Each subject was given an otologic examination to determine their pure-tone hearing threshold levels and had their personal full-shift noise recorded. For each subject, the hearing loss was evaluated according to the hearing impairment definition of the National Institute for Occupational Safety and Health. Age, exposure duration, equivalent A-weighted SPL (L_Aeq), and median kurtosis were used as the input for four machine learning algorithms, that is, support vector machine, neural network multilayer perceptron, random forest, and adaptive boosting. Both classification and regression models were developed to predict noise-induced hearing loss applying these four machine learning algorithms. Two indexes, area under the curve and prediction accuracy, were used to assess the performances of the classification models for predicting hearing impairment of workers. Root mean square error was used to quantify the prediction performance of the regression models.

Results:

A prediction accuracy between 78.6 and 80.1% indicated that the four classification models could be useful tools to assess noise-induced hearing impairment of workers exposed to various complex occupational noises. A comprehensive evaluation using both the area under the curve and prediction accuracy showed that the support vector machine model achieved the best score and thus should be selected as the tool with the highest potential for predicting hearing impairment from the occupational noise exposures in this study. The root mean square error performance indicated that the four regression models could be used to predict noise-induced hearing loss quantitatively and the multilayer perceptron regression model had the best performance.

Conclusions:

This pilot study demonstrated that machine learning algorithms are potential tools for the evaluation and prediction of noise-induced hearing impairment in workers exposed to diverse complex industrial noises.

Use of Neural Networks to Identify Safety Prevention Priorities in Agro-Manufacturing Operations within Commercial Grain Elevators

The grain handling industry plays a significant role in U.S. agriculture by storing, distributing, and processing a variety of agricultural commodities. Commercial grain elevators are hazardous agro-manufacturing work environments where workers are prone to severe injuries, due to the nature of the activities and workplace. Safety incidents in agro-manufacturing operations generally arise from a combination of factors, rather than a single cause, therefore, research on occupational incidents must look deeper into identifying the underlying causes, through the application of advanced analyses methods. In occupational safety, it is possible to estimate and predict probability of safety risks through developing artificial neural network predictive models. Due to the significance of safety risk assessment in the design and prioritization of effective prevention measures, this study aimed at classifying and predicting causes of occupational incidents in grain elevator agro-manufacturing operations in the Midwest region of the United States. Workers’ compensation claims data, from 2008 to 2016, were utilized for training multilayer perceptron (MLP) and radial basis function (RBF) neural networks. Both MLP and RBF models could predict the probability of safety risks with a high overall accuracy of 60%, 61%. Based on values of AUC (area under the curve) from the ROC (receiving operating charts), both models predicted the probability of individual safety risks with a high accuracy rate of between 71.5% and 99.2%. In addition, sensitivity analysis showed that nature of injury is the most significant determinant of safety risks probability, along with type of injury. The novelty of this study is the use of the artificial neural network methodology to analyze multi-level causes of occupational incidents as the sources of safety risks in bulk storage facilities. The results confirm that artificial neural networks are useful in safety risk estimation, and identifying the incidents’ risk factors. The implementation of safety measures in grain elevators can help in preventing occupational injuries, saving lives, and reducing the occurrence and severity of such incidents in industrial work environments.

A parsimonious approach for screening moderate-to-profound hearing loss in a community-dwelling geriatric population based on a decision tree analysis

BACKGROUND

Hearing loss is one of the most common modifiable factors associated with cognitive and functional decline in geriatric populations. An accurate, easy-to-apply, and inexpensive hearing screening method is needed to detect hearing loss in community-dwelling elderly people, intervene early and reduce the negative consequences and burden of untreated hearing loss on individuals, families and society. However, available hearing screening tools do not adequately meet the need for large-scale geriatric hearing detection due to several barriers, including time, personnel training and equipment costs. This study aimed to propose an efficient method that could potentially satisfy this need.

METHODS

In total, 1793 participants (≥60 years) were recruited to undertake a standard audiometric air conduction pure tone test at 4 frequencies (0.5-4 kHz). Audiometric data from one community were used to train the decision tree model and generate a pure tone screening rule to classify people with or without moderate or more serious hearing impairment. Audiometric data from another community were used to validate the tree model.

RESULTS

In the decision tree analysis, 2 kHz and 0.5 kHz were found to be the most important frequencies for hearing severity classification. The tree model suggested a simple two-step screening procedure in which a 42 dB HL tone at 2 kHz is presented first, followed by a 47 dB HL tone at 0.5 kHz, depending on the individual's response to the first tone. This approach achieved an accuracy of 91.20% (91.92%), a sensitivity of 95.35% (93.50%) and a specificity of 86.85% (90.56%) in the training dataset (testing dataset).

CONCLUSIONS

A simple two-step screening procedure using the two tones (2 kHz and 0.5 kHz) selected by the decision tree analysis can be applied to screen moderate-to-profound hearing loss in a community-based geriatric population in Shanghai. The decision tree analysis is useful in determining the optimal hearing screening criteria for local elderly populations. Implanting the pair of tones into a well-calibrated sound generator may create a simple, practical and time-efficient screening tool with high accuracy that is readily available at healthcare centers of all levels, thereby facilitating the initiation of extensive nationwide hearing screening in older adults.

The combined effects of occupational exposure to noise and other risk factors - a systematic review

PURPOSE

Noise-induced health effects exacerbate by many other risk factors. This systematic review aims at shedding light on the combined effects of co-exposure to occupational noise and other factors.

MATERIAL AND METHODS

A literature search in Web of Science, Scopus, PubMed, Science Direct, and Google Scholar, with appropriate keywords on combined effects of occupational noise, and co-exposure to noise and other factors, revealed 7928 articles which were screened by two researchers. A total of 775 articles were reviewed in full text. We found 149 articles that were relevant and had sufficient quality for analysis.

RESULTS

We identified 16 risk factors that exacerbate occupational noise-induced health effects. These factors were classified into four groups: chemical (carbon monoxide (CO), solvents, heavy metals, and other chemicals), physical (lighting, heat, vibration, and cold), personal (age, gender, genetics, smoking, medication, contextual diseases) and occupational (workload and shift work). Hearing loss, hypertension, reduced performance, and cardiovascular strains, are the most important risk factors combined effects due to concurrent exposure to noise and other risk factors.

CONCLUSION

Evidences of combined effects of solvents, vibration, heavy metals, CO, smoking, chemicals, aging, heat, and shiftwork were respectively stronger than for other factors. Most of the studies have investigated only the combined effects of risk factors on hearing, and the evidence for non-auditory effects is still limited, and more studies are warranted. Therefore, in the Hearing Conservation Programs, besides noise, aggravating factors of noise effects should also be taken into account.

Perspectivas do uso de mineração de dados e aprendizado de máquina em saúde e segurança no trabalho

Resumo Introdução: a variedade, volume e velocidade de geração de dados (big data) possibilitam novas e mais complexas análises. Objetivo: discutir e apresentar técnicas de mineração de dados (data mining) e de aprendizado de máquina (machine learning) para auxiliar pesquisadores de Saúde e Segurança no Trabalho (SST) na escolha da técnica adequada para lidar com big data. Métodos: revisão bibliográfica com foco em data mining e no uso de análises preditivas com machine learning e suas aplicações para auxiliar diagnósticos e predição de riscos em SST. Resultados: a literatura indica que aplicações de data mining com algoritmos de machine learning para análises preditivas em saúde pública e em SST apresentam melhor desempenho em comparação com análises tradicionais. São sugeridas técnicas de acordo com o tipo de pesquisa almejada. Discussão: data mining tem se tornado uma alternativa cada vez mais comum para lidar com bancos de dados de saúde pública, possibilitando analisar grandes volumes de dados de morbidade e mortalidade. Tais técnicas não visam substituir o fator humano, mas auxiliar em processos de tomada de decisão, servir de ferramenta para a análise estatística e gerar conhecimento para subsidiar ações que possam melhorar a qualidade de vida do trabalhador.

Prediction Effects of Personal, Psychosocial, and Occupational Risk Factors on Low Back Pain Severity Using Artificial Neural Networks Approach in Industrial Workers

OBJECTIVES

This study aimed to provide an empirical model of predicting low back pain (LBP) by considering the occupational, personal, and psychological risk factor interactions in workers population employed in industrial units using an artificial neural networks approach.

METHODS

A total of 92 workers with LBP as the case group and 68 healthy workers as a control group were selected in various industrial units with similar occupational conditions. The demographic information and personal, occupational, and psychosocial factors of the participants were collected via interview, related questionnaires, consultation with occupational medicine, and also the Rapid Entire Body Assessment worksheet and National Aeronautics and Space Administration Task Load Index software. Then, 16 risk factors for LBP were used as input variables to develop the prediction model. Networks with various multilayered structures were developed using MATLAB.

RESULTS

The developed neural networks with 1 hidden layer and 26 neurons had the least error of classification in both training and testing phases. The mean of classification accuracy of the developed neural networks for the testing and training phase data were about 88% and 96%, respectively. In addition, the mean of classification accuracy of both training and testing data was 92%, indicating much better results compared with other methods.

CONCLUSION

It appears that the prediction model using the neural network approach is more accurate compared with other applied methods. Because occupational LBP is usually untreatable, the results of prediction may be suitable for developing preventive strategies and corrective interventions.

Individual Fit Testing of Hearing Protection Devices Based on Microphone in Real Ear

Background

Labeled noise reduction (NR) data presented by manufacturers are considered one of the main challenging issues for occupational experts in employing hearing protection devices (HPDs). This study aimed to determine the actual NR data of typical HPDs using the objective fit testing method with a microphone in real ear (MIRE) method.

Methods

Five available commercially earmuff protectors were investigated in 30 workers exposed to reference noise source according to the standard method, ISO 11904-1. Personal attenuation rating (PAR) of the earmuffs was measured based on the MIRE method using a noise dosimeter (SVANTEK, model SV 102).

Results

The results showed that means of PAR of the earmuffs are from 49% to 86% of the nominal NR rating. The PAR values of earmuffs when a typical eyewear was worn differed statistically (p < 0.05). It is revealed that a typical safety eyewear can reduce the mean of the PAR value by approximately 2.5 dB. The results also showed that measurements based on the MIRE method resulted in low variability. The variability in NR values between individuals, within individuals, and within earmuffs was not the statistically significant (p > 0.05).

Conclusion

This study could provide local individual fit data. Ergonomic aspects of the earmuffs and different levels of users experience and awareness can be considered the main factors affecting individual fitting compared with the laboratory condition for acquiring the labeled NR data. Based on the obtained fit testing results, the field application of MIRE can be employed for complementary studies in real workstations while workers perform their regular work duties.

Status of Hearing Loss and Its Related Factors among Drivers in Zahedan, South-Eastern Iran

Objective:

This study aims to investigate loss of hearing among drivers in Zahedan, southeastern Iran.

Patients and Methods:

This study carried out on a total of 1836 drivers in Zahedan in 2013. Loss of hearing in both ears was measured at 250, 1000, 2000, 3000, 4000, 6000, and 8000 Hertz. The demographic variables, blood parameter and anthropometric data were recorded through interview and examinations. Data were analyzed in Stata.12 software using paired t-tests, McNemar test and Multiple Logistic Regression.

Results:

The mean age was 38.2±9.8 years. The highest mean hearing thresholds in the right and left ears were 25.7±9.1 and 27.7±9.1, respectively at 250 Hz. There was significant difference between left and right ears hearing threshold at all frequencies (P<0.001), and the highest difference occurred at 250 Hz. Hearing threshold in the left ear was greater than in the right ear at all frequencies. Hearing threshold was correlated to marital status, type of license, and vehicle, smoking, age, and driving history at all frequencies (P<0.01), and also significantly correlated to blood sugar and cholesterol levels at 250 and 500 Hz in both left and right ears (P<0.01).

Conclusion:

In conclusion, high levels of noise increase hearing threshold with greatest damage to the left ear. Therefore, drivers should be periodically examined for ear damage in accordance to variables affecting loss of hearing. Moreover, drivers must be educated about usage of appropriate ear-plugs during driving, especially for the left ear.