International Journal of Environmental Research and Public Health

Research

18 pages, 2437 KiB

Open AccessArticle

Fuzzy K-Nearest Neighbor Based Dental Fluorosis Classification Using Multi-Prototype Unsupervised Possibilistic Fuzzy Clustering via Cuckoo Search Algorithm

by Ritipong Wongkhuenkaew, Sansanee Auephanwiriyakul, Nipon Theera-Umpon, Kasemsit Teeyapan and Uklid Yeesarapat

Int. J. Environ. Res. Public Health 2023, 20(4), 3394; https://doi.org/10.3390/ijerph20043394 - 15 Feb 2023

Cited by 4 | Viewed by 2160

Abstract

Dental fluorosis in children is a prevalent disease in many regions of the world. One of its root causes is excessive exposure to high concentrations of fluoride in contaminated drinking water during tooth formation. Typically, the disease causes undesirable chalky white or even [...] Read more.

Dental fluorosis in children is a prevalent disease in many regions of the world. One of its root causes is excessive exposure to high concentrations of fluoride in contaminated drinking water during tooth formation. Typically, the disease causes undesirable chalky white or even dark brown stains on the tooth enamel. To help dentists screen the severity of fluorosis, this paper proposes an automatic image-based dental fluorosis segmentation and classification system. Six features from red, green, and blue (RGB) and hue, saturation, and intensity (HIS) color spaces are clustered using unsupervised possibilistic fuzzy clustering (UPFC) into five categories: white, yellow, opaque, brown, and background. The fuzzy k-nearest neighbor method is used for feature classification, and the number of clusters is optimized using the cuckoo search algorithm. The resulting multi-prototypes are further utilized to create a binary mask of teeth and used to segment the tooth region into three groups: white–yellow, opaque, and brown pixels. Finally, a fluorosis classification rule is created based on the proportions of opaque and brown pixels to classify fluorosis into four classes: Normal, Stage 1, Stage 2, and Stage 3. The experimental results on 128 blind test images showed that the average pixel accuracy of the segmented binary tooth mask was 92.24% over the four fluorosis classes, and the average pixel accuracy of segmented teeth into white–yellow, opaque, and brown pixels was 79.46%. The proposed method correctly classified four classes of fluorosis in 86 images from a total of 128 blind test images. When compared with a previous work, this result also indicates 10 out of 15 correct classifications on the blind test images, which is equivalent to a 13.33% improvement over the previous work. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

15 pages, 6562 KiB

Open AccessArticle

Signal Acquisition-Independent Lossless Electrocardiogram Compression Using Adaptive Linear Prediction

by Krittapat Bannajak, Nipon Theera-Umpon and Sansanee Auephanwiriyakul

Int. J. Environ. Res. Public Health 2023, 20(3), 2753; https://doi.org/10.3390/ijerph20032753 - 3 Feb 2023

Cited by 2 | Viewed by 1964

Abstract

In this paper, we propose a lossless electrocardiogram (ECG) compression method using a prediction error-based adaptive linear prediction technique. This method combines the adaptive linear prediction, which minimizes the prediction error in the ECG signal prediction, and the modified Golomb–Rice coding, which encodes [...] Read more.

In this paper, we propose a lossless electrocardiogram (ECG) compression method using a prediction error-based adaptive linear prediction technique. This method combines the adaptive linear prediction, which minimizes the prediction error in the ECG signal prediction, and the modified Golomb–Rice coding, which encodes the prediction error to the binary code as the compressed data. We used the PTB Diagnostic ECG database, the European ST-T database, and the MIT-BIH Arrhythmia database for the evaluation and achieved the average compression ratios for single-lead ECG signals of 3.16, 3.75, and 3.52, respectively, despite different signal acquisition setup in each database. As the prediction order is very crucial for this particular problem, we also investigate the validity of the popular linear prediction coefficients that are generally used in ECG compression by determining the prediction coefficients from the three databases using the autocorrelation method. The findings are in agreement with the previous works in that the second-order linear prediction is suitable for the ECG compression application. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

15 pages, 2560 KiB

Open AccessArticle

Integrated Machine Learning Decision Tree Model for Risk Evaluation in Patients with Non-Valvular Atrial Fibrillation When Taking Different Doses of Dabigatran

by Yung-Chuan Huang, Yu-Chen Cheng, Mao-Jhen Jhou, Mingchih Chen and Chi-Jie Lu

Int. J. Environ. Res. Public Health 2023, 20(3), 2359; https://doi.org/10.3390/ijerph20032359 - 29 Jan 2023

Viewed by 2411

Abstract

The new generation of nonvitamin K antagonists are broadly applied for stroke prevention due to their notable efficacy and safety. Our study aimed to develop a suggestive utilization of dabigatran through an integrated machine learning (ML) decision-tree model. Participants taking different doses of [...] Read more.

The new generation of nonvitamin K antagonists are broadly applied for stroke prevention due to their notable efficacy and safety. Our study aimed to develop a suggestive utilization of dabigatran through an integrated machine learning (ML) decision-tree model. Participants taking different doses of dabigatran in the Randomized Evaluation of Long-Term Anticoagulant Therapy trial were included in our analysis and defined as the 110 mg and 150 mg groups. The proposed scheme integrated ML methods, namely naive Bayes, random forest (RF), classification and regression tree (CART), and extreme gradient boosting (XGBoost), which were used to identify the essential variables for predicting vascular events in the 110 mg group and bleeding in the 150 mg group. RF (0.764 for 110 mg; 0.747 for 150 mg) and XGBoost (0.708 for 110 mg; 0.761 for 150 mg) had better area under the receiver operating characteristic curve (AUC) values than logistic regression (benchmark model; 0.683 for 110 mg; 0.739 for 150 mg). We then selected the top ten important variables as internal nodes of the CART decision tree. The two best CART models with ten important variables output tree-shaped rules for predicting vascular events in the 110 mg group and bleeding in the 150 mg group. Our model can be used to provide more visualized and interpretable suggestive rules to clinicians managing NVAF patients who are taking dabigatran. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

20 pages, 3432 KiB

Open AccessArticle

EEG Emotion Recognition Applied to the Effect Analysis of Music on Emotion Changes in Psychological Healthcare

by Tie Hua Zhou, Wenlong Liang, Hangyu Liu, Ling Wang, Keun Ho Ryu and Kwang Woo Nam

Int. J. Environ. Res. Public Health 2023, 20(1), 378; https://doi.org/10.3390/ijerph20010378 - 26 Dec 2022

Cited by 6 | Viewed by 3228

Abstract

Music therapy is increasingly being used to promote physical health. Emotion semantic recognition is more objective and provides direct awareness of the real emotional state based on electroencephalogram (EEG) signals. Therefore, we proposed a music therapy method to carry out emotion semantic matching [...] Read more.

Music therapy is increasingly being used to promote physical health. Emotion semantic recognition is more objective and provides direct awareness of the real emotional state based on electroencephalogram (EEG) signals. Therefore, we proposed a music therapy method to carry out emotion semantic matching between the EEG signal and music audio signal, which can improve the reliability of emotional judgments, and, furthermore, deeply mine the potential influence correlations between music and emotions. Our proposed EER model (EEG-based Emotion Recognition Model) could identify 20 types of emotions based on 32 EEG channels, and the average recognition accuracy was above 90% and 80%, respectively. Our proposed music-based emotion classification model (MEC model) could classify eight typical emotion types of music based on nine music feature combinations, and the average classification accuracy was above 90%. In addition, the semantic mapping was analyzed according to the influence of different music types on emotional changes from different perspectives based on the two models, and the results showed that the joy type of music video could improve fear, disgust, mania, and trust emotions into surprise or intimacy emotions, while the sad type of music video could reduce intimacy to the fear emotion. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

23 pages, 754 KiB

Open AccessArticle

Effects of the Developing and Using a Model to Predict Dengue Risk Villages Based on Subdistrict Administrative Organization in Southern Thailand

by Orratai Nontapet, Jiraporn Jaroenpool, Sarunya Maneerattanasa, Supaporn Thongchan, Chumpron Ponprasert, Patthanasak Khammaneechan, Cua Ngoc Le, Nirachon Chutipattana and Charuai Suwanbamrung

Int. J. Environ. Res. Public Health 2022, 19(19), 11989; https://doi.org/10.3390/ijerph191911989 - 22 Sep 2022

Cited by 2 | Viewed by 2220

Abstract

The purpose of this study was to evaluate the effects of developing and using a model to predict dengue risk in villages and of a larval indices surveillance system for 2372 households in 10 Thai villages. A community participatory action research method was [...] Read more.

The purpose of this study was to evaluate the effects of developing and using a model to predict dengue risk in villages and of a larval indices surveillance system for 2372 households in 10 Thai villages. A community participatory action research method was used in five steps: (1) community preparation covering all stakeholders, (2) assessment of the understanding of a dengue solution and a larval indices surveillance system, (3) development of a prediction and intervention model for dengue risk villages, (4) implementation of the model that responds to all stakeholders, and (5) evaluation of the effects of using the model. The questionnaires to assess and evaluate were validated and reliability tested. The chi-square test and Fisher’s exact test were used to analyze the quantitative data collected by means of questionnaires. Thematic analysis was applied to the qualitative data collected through interviews. The results found that the model consisted of six main activities, including (1) setting team leader responsibility, (2) situation assessment, (3) prediction of the dengue risk in villages, (4) the six steps of the larval indices surveillance system, (5) the understanding of the dengue solution and the understanding of the larval indices surveillance system training program, and (6) local wisdom innovation. The effects of using the model showed a statistically significant increase in correct understanding among 932 family leaders, 109 village health volunteers, and 59 student leaders regarding dengue prevention and control (p < 0.05). The larval indices and dengue morbidity were diminished and related to the nine themes present in the community leaders’ reflections and to the satisfaction of the community members. Hence, local administrative organizations should use community-based approaches as the subdistrict dengue solution innovation to reduce the dengue problem. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

21 pages, 4066 KiB

Open AccessArticle

Discovering Thematically Coherent Biomedical Documents Using Contextualized Bidirectional Encoder Representations from Transformers-Based Clustering

by Khishigsuren Davagdorj, Ling Wang, Meijing Li, Van-Huy Pham, Keun Ho Ryu and Nipon Theera-Umpon

Int. J. Environ. Res. Public Health 2022, 19(10), 5893; https://doi.org/10.3390/ijerph19105893 - 12 May 2022

Cited by 9 | Viewed by 2582

Abstract

The increasing expansion of biomedical documents has increased the number of natural language textual resources related to the current applications. Meanwhile, there has been a great interest in extracting useful information from meaningful coherent groupings of textual content documents in the last decade. [...] Read more.

The increasing expansion of biomedical documents has increased the number of natural language textual resources related to the current applications. Meanwhile, there has been a great interest in extracting useful information from meaningful coherent groupings of textual content documents in the last decade. However, it is challenging to discover informative representations and define relevant articles from the rapidly growing biomedical literature due to the unsupervised nature of document clustering. Moreover, empirical investigations demonstrated that traditional text clustering methods produce unsatisfactory results in terms of non-contextualized vector space representations because that neglect the semantic relationship between biomedical texts. Recently, pre-trained language models have emerged as successful in a wide range of natural language processing applications. In this paper, we propose the Gaussian Mixture Model-based efficient clustering framework that incorporates substantially pre-trained (Bidirectional Encoder Representations from Transformers for Biomedical Text Mining) BioBERT domain-specific language representations to enhance the clustering accuracy. Our proposed framework consists of main three phases. First, classic text pre-processing techniques are used biomedical document data, which crawled from the PubMed repository. Second, representative vectors are extracted from a pre-trained BioBERT language model for biomedical text mining. Third, we employ the Gaussian Mixture Model as a clustering algorithm, which allows us to assign labels for each biomedical document. In order to prove the efficiency of our proposed model, we conducted a comprehensive experimental analysis utilizing several clustering algorithms while combining diverse embedding techniques. Consequently, the experimental results show that the proposed model outperforms the benchmark models by reaching performance measures of Fowlkes mallows score, silhouette coefficient, adjusted rand index, Davies-Bouldin score of 0.7817, 0.3765, 0.4478, 1.6849, respectively. We expect the outcomes of this study will assist domain specialists in comprehending thematically cohesive documents in the healthcare field. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

11 pages, 2120 KiB

Open AccessArticle

Associations between Sex and Risk Factors for Predicting Chronic Kidney Disease

by Hao-Yun Kao, Chi-Chang Chang, Chin-Fang Chang, Ying-Chen Chen, Chalong Cheewakriangkrai and Ya-Ling Tu

Int. J. Environ. Res. Public Health 2022, 19(3), 1219; https://doi.org/10.3390/ijerph19031219 - 22 Jan 2022

Cited by 11 | Viewed by 3033

Abstract

Gender is an important risk factor in predicting chronic kidney disease (CKD); however, it is under-researched. The purpose of this study was to examine whether gender differences affect the risk factors of early CKD prediction. This study used data from 19,270 adult health [...] Read more.

Gender is an important risk factor in predicting chronic kidney disease (CKD); however, it is under-researched. The purpose of this study was to examine whether gender differences affect the risk factors of early CKD prediction. This study used data from 19,270 adult health screenings, including 5101 with CKD, to screen for 11 independent variables selected as risk factors and to test for the significant effects of statistical Chi-square test variables, using seven machine learning techniques to train the predictive models. Performance indicators included classification accuracy, sensitivity, specificity, and precision. Unbalanced category issues were addressed using three extraction methods: manual sampling, the synthetic minority oversampling technique, and SpreadSubsample. The Chi-square test revealed statistically significant results (p < 0.001) for gender, age, red blood cell count in urine, urine protein (PRO) content, and the PRO-to-urinary creatinine ratio. In terms of classifier prediction performance, the manual extraction method, logistic regression, exhibited the highest average prediction accuracy rate (0.8053) for men, whereas the manual extraction method, linear discriminant analysis, demonstrated the highest average prediction accuracy rate (0.8485) for women. The clinical features of a normal or abnormal PRO-to-urinary creatinine ratio indicated that PRO ratio, age, and urine red blood cell count are the most important risk factors with which to predict CKD in both genders. As a result, this study proposes a prediction model with acceptable prediction accuracy. The model supports doctors in diagnosis and treatment and achieves the goal of early detection and treatment. Based on the evidence-based medicine, machine learning methods are used to develop predictive model in this study. The model has proven to support the prediction of early clinical risk of CKD as much as possible to improve the efficacy and quality of clinical decision making. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

16 pages, 10371 KiB

Open AccessArticle

Medical Prognosis of Infectious Diseases in Nursing Homes by Applying Machine Learning on Clinical Data Collected in Cloud Microservices

by Alberto Garcés-Jiménez, Huriviades Calderón-Gómez, José M. Gómez-Pulido, Juan A. Gómez-Pulido, Miguel Vargas-Lombardo, José L. Castillo-Sequera, Miguel Pablo Aguirre, José Sanz-Moreno, María-Luz Polo-Luque and Diego Rodríguez-Puyol

Int. J. Environ. Res. Public Health 2021, 18(24), 13278; https://doi.org/10.3390/ijerph182413278 - 16 Dec 2021

Cited by 4 | Viewed by 3799

Abstract

Background: treating infectious diseases in elderly individuals is difficult; patient referral to emergency services often occurs, since the elderly tend to arrive at consultations with advanced, serious symptoms. Aim: it was hypothesized that anticipating an infectious disease diagnosis by a few days could [...] Read more.

Background: treating infectious diseases in elderly individuals is difficult; patient referral to emergency services often occurs, since the elderly tend to arrive at consultations with advanced, serious symptoms. Aim: it was hypothesized that anticipating an infectious disease diagnosis by a few days could significantly improve a patient’s well-being and reduce the burden on emergency health system services. Methods: vital signs from residents were taken daily and transferred to a database in the cloud. Classifiers were used to recognize patterns in the spatial domain process of the collected data. Doctors reported their diagnoses when any disease presented. A flexible microservice architecture provided access and functionality to the system. Results: combining two different domains, health and technology, is not easy, but the results are encouraging. The classifiers reported good results; the system has been well accepted by medical personnel and is proving to be cost-effective and a good solution to service disadvantaged areas. In this context, this research found the importance of certain clinical variables in the identification of infectious diseases. Conclusions: this work explores how to apply mobile communications, cloud services, and machine learning technology, in order to provide efficient tools for medical staff in nursing homes. The scalable architecture can be extended to big data applications that may extract valuable knowledge patterns for medical research. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

28 pages, 1835 KiB

Open AccessArticle

Assessing Patient-Perceived Hospital Service Quality and Sentiment in Malaysian Public Hospitals Using Machine Learning and Facebook Reviews

by Afiq Izzudin A. Rahim, Mohd Ismail Ibrahim, Kamarul Imran Musa, Sook-Ling Chua and Najib Majdi Yaacob

Int. J. Environ. Res. Public Health 2021, 18(18), 9912; https://doi.org/10.3390/ijerph18189912 - 21 Sep 2021

Cited by 18 | Viewed by 6150

Abstract

Social media is emerging as a new avenue for hospitals and patients to solicit input on the quality of care. However, social media data is unstructured and enormous in volume. Moreover, no empirical research on the use of social media data and perceived [...] Read more.

Social media is emerging as a new avenue for hospitals and patients to solicit input on the quality of care. However, social media data is unstructured and enormous in volume. Moreover, no empirical research on the use of social media data and perceived hospital quality of care based on patient online reviews has been performed in Malaysia. The purpose of this study was to investigate the determinants of positive sentiment expressed in hospital Facebook reviews in Malaysia, as well as the association between hospital accreditation and sentiments expressed in Facebook reviews. From 2017 to 2019, we retrieved comments from 48 official public hospitals’ Facebook pages. We used machine learning to build a sentiment analyzer and service quality (SERVQUAL) classifier that automatically classifies the sentiment and SERVQUAL dimensions. We utilized logistic regression analysis to determine our goals. We evaluated a total of 1852 reviews and our machine learning sentiment analyzer detected 72.1% of positive reviews and 27.9% of negative reviews. We classified 240 reviews as tangible, 1257 reviews as trustworthy, 125 reviews as responsive, 356 reviews as assurance, and 1174 reviews as empathy using our machine learning SERVQUAL classifier. After adjusting for hospital characteristics, all SERVQUAL dimensions except Tangible were associated with positive sentiment. However, no significant relationship between hospital accreditation and online sentiment was discovered. Facebook reviews powered by machine learning algorithms provide valuable, real-time data that may be missed by traditional hospital quality assessments. Additionally, online patient reviews offer a hitherto untapped indication of quality that may benefit all healthcare stakeholders. Our results confirm prior studies and support the use of Facebook reviews as an adjunct method for assessing the quality of hospital services in Malaysia. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

18 pages, 1258 KiB

Open AccessArticle

Infectious Disease Relational Data Analysis Using String Grammar Non-Euclidean Relational Fuzzy C-Means

by Apiwat Budwong, Sansanee Auephanwiriyakul and Nipon Theera-Umpon

Int. J. Environ. Res. Public Health 2021, 18(15), 8153; https://doi.org/10.3390/ijerph18158153 - 1 Aug 2021

Cited by 1 | Viewed by 2519

Abstract

Statistical analysis in infectious diseases is becoming more important, especially in prevention policy development. To achieve that, the epidemiology, a study of the relationship between the occurrence and who/when/where, is needed. In this paper, we develop the string grammar non-Euclidean relational fuzzy C-means [...] Read more.

Statistical analysis in infectious diseases is becoming more important, especially in prevention policy development. To achieve that, the epidemiology, a study of the relationship between the occurrence and who/when/where, is needed. In this paper, we develop the string grammar non-Euclidean relational fuzzy C-means (sgNERF-CM) algorithm to determine a relationship inside the data from the age, career, and month viewpoint for all provinces in Thailand for the dengue fever, influenza, and Hepatitis B virus (HBV) infection. The Dunn’s index is used to select the best models because of its ability to identify the compact and well-separated clusters. We compare the results of the sgNERF-CM algorithm with the string grammar relational hard C-means (sgRHCM) algorithm. In addition, their numerical counterparts, i.e., relational hard C-means (RHCM) and non-Euclidean relational fuzzy C-means (NERF-CM) algorithms are also applied in the comparison. We found that the sgNERF-CM algorithm is far better than the numerical counterparts and better than the sgRHCM algorithm in most cases. From the results, we found that the month-based dataset does not help in relationship-finding since the diseases tend to happen all year round. People from different age ranges in different regions in Thailand have different numbers of dengue fever infections. The occupations that have a higher chance to have dengue fever are student and teacher groups from the central, north-east, north, and south regions. Additionally, students in all regions, except the central region, have a high risk of dengue infection. For the influenza dataset, we found that a group of people with the age of more than 1 year to 64 years old has higher number of influenza infections in every province. Most occupations in all regions have a higher risk of infecting the influenza. For the HBV dataset, people in all regions with an age between 10 to 65 years old have a high risk in infecting the disease. In addition, only farmer and general contractor groups in all regions have high chance of infecting HBV as well. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

16 pages, 707 KiB

Open AccessArticle

Facebook Reviews as a Supplemental Tool for Hospital Patient Satisfaction and Its Relationship with Hospital Accreditation in Malaysia

by Afiq Izzudin A. Rahim, Mohd Ismail Ibrahim, Kamarul Imran Musa and Sook-Ling Chua

Int. J. Environ. Res. Public Health 2021, 18(14), 7454; https://doi.org/10.3390/ijerph18147454 - 13 Jul 2021

Cited by 15 | Viewed by 4235

Abstract

Patient satisfaction is one indicator used to assess the impact of accreditation on patient care. However, traditional patient satisfaction surveys have a few disadvantages, and some researchers have suggested that social media be used in their place. Social media usage is gaining popularity [...] Read more.

Patient satisfaction is one indicator used to assess the impact of accreditation on patient care. However, traditional patient satisfaction surveys have a few disadvantages, and some researchers have suggested that social media be used in their place. Social media usage is gaining popularity in healthcare organizations, but there is still a paucity of data to support it. The purpose of this study was to determine the association between online reviews and hospital patient satisfaction and the relationship between online reviews and hospital accreditation. We used a cross-sectional design with data acquired from the official Facebook pages of 48 Malaysian public hospitals, 25 of which are accredited. We collected all patient comments from Facebook reviews of those hospitals between 2018 and 2019. Spearman’s correlation and logistic regression were used to evaluate the data. There was a significant and moderate correlation between hospital patient satisfaction and online reviews. Patient satisfaction was closely connected to urban location, tertiary hospital, and previous Facebook ratings. However, hospital accreditation was not found to be significantly associated with online reports of patient satisfaction. This groundbreaking study demonstrates how Facebook reviews can assist hospital administrators in monitoring their institutions’ quality of care in real time. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

19 pages, 2508 KiB

Open AccessArticle

Predicting Survival in Veterans with Follicular Lymphoma Using Structured Electronic Health Record Information and Machine Learning

by Chunyang Li, Vikas Patil, Kelli M. Rasmussen, Christina Yong, Hsu-Chih Chien, Debbie Morreall, Jeffrey Humpherys, Brian C. Sauer, Zachary Burningham and Ahmad S. Halwani

Int. J. Environ. Res. Public Health 2021, 18(5), 2679; https://doi.org/10.3390/ijerph18052679 - 7 Mar 2021

Cited by 6 | Viewed by 2938

Abstract

The most accurate prognostic approach for follicular lymphoma (FL), progression of disease at 24 months (POD24), requires two years’ observation after initiating first-line therapy (L1) to predict outcomes. We applied machine learning to structured electronic health record (EHR) data to predict individual survival [...] Read more.

The most accurate prognostic approach for follicular lymphoma (FL), progression of disease at 24 months (POD24), requires two years’ observation after initiating first-line therapy (L1) to predict outcomes. We applied machine learning to structured electronic health record (EHR) data to predict individual survival at L1 initiation. We grouped 523 observations and 1933 variables from a nationwide cohort of FL patients diagnosed 2006–2014 in the Veterans Health Administration into traditionally used prognostic variables (“curated”), commonly measured labs (“labs”), and International Classification of Diseases diagnostic codes (“ICD”) sets. We compared performance of random survival forests (RSF) vs. traditional Cox model using four datasets: curated, curated + labs, curated + ICD, and curated + ICD + labs, also using Cox on curated + POD24. We evaluated variable importance and partial dependence plots with area under the receiver operating characteristic curve (AUC). RSF with curated + labs performed best, with mean AUC 0.73 (95% CI: 0.71–0.75). It approximated, but did not surpass, Cox with POD24 (mean AUC 0.74 [95% CI: 0.71–0.77]). RSF using EHR data achieved better performance than traditional prognostic variables, setting the foundation for the incorporation of our algorithm into the EHR. It also provides for possible future scenarios in which clinicians could be provided an EHR-based tool which approximates the predictive ability of the most accurate known indicator, using information available 24 months earlier. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

14 pages, 2045 KiB

Open AccessArticle

A Deep Learning Model for Classification of Endoscopic Gastroesophageal Reflux Disease

by Chi-Chih Wang, Yu-Ching Chiu, Wei-Liang Chen, Tzu-Wei Yang, Ming-Chang Tsai and Ming-Hseng Tseng

Int. J. Environ. Res. Public Health 2021, 18(5), 2428; https://doi.org/10.3390/ijerph18052428 - 2 Mar 2021

Cited by 24 | Viewed by 9383

Abstract

Gastroesophageal reflux disease (GERD) is a common disease with high prevalence, and its endoscopic severity can be evaluated using the Los Angeles classification (LA grade). This paper proposes a deep learning model (i.e., GERD-VGGNet) that employs convolutional neural networks for automatic classification and [...] Read more.

Gastroesophageal reflux disease (GERD) is a common disease with high prevalence, and its endoscopic severity can be evaluated using the Los Angeles classification (LA grade). This paper proposes a deep learning model (i.e., GERD-VGGNet) that employs convolutional neural networks for automatic classification and interpretation of routine GERD LA grade. The proposed model employs a data augmentation technique, a two-stage no-freezing fine-tuning policy, and an early stopping criterion. As a result, the proposed model exhibits high generalizability. A dataset of images from 464 patients was used for model training and validation. An additional 32 patients served as a test set to evaluate the accuracy of both the model and our trainees. Experimental results demonstrate that the best model for the development set exhibited an overall accuracy of 99.2% (grade A–B), 100% (grade C–D), and 100% (normal group) using narrow-band image (NBI) endoscopy. On the test set, the proposed model resulted in an accuracy of 87.9%, which was significantly higher than the results of the trainees (75.0% and 65.6%). The proposed GERD-VGGNet model can assist automatic classification of GERD in conventional and NBI environments and thereby increase the accuracy of interpretation of the results by inexperienced endoscopists. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

24 pages, 5717 KiB

Open AccessArticle

Deep Learning Feature Extraction Approach for Hematopoietic Cancer Subtype Classification

by Kwang Ho Park, Erdenebileg Batbaatar, Yongjun Piao, Nipon Theera-Umpon and Keun Ho Ryu

Int. J. Environ. Res. Public Health 2021, 18(4), 2197; https://doi.org/10.3390/ijerph18042197 - 23 Feb 2021

Cited by 20 | Viewed by 4880

Abstract

Hematopoietic cancer is a malignant transformation in immune system cells. Hematopoietic cancer is characterized by the cells that are expressed, so it is usually difficult to distinguish its heterogeneities in the hematopoiesis process. Traditional approaches for cancer subtyping use statistical techniques. Furthermore, due [...] Read more.

Hematopoietic cancer is a malignant transformation in immune system cells. Hematopoietic cancer is characterized by the cells that are expressed, so it is usually difficult to distinguish its heterogeneities in the hematopoiesis process. Traditional approaches for cancer subtyping use statistical techniques. Furthermore, due to the overfitting problem of small samples, in case of a minor cancer, it does not have enough sample material for building a classification model. Therefore, we propose not only to build a classification model for five major subtypes using two kinds of losses, namely reconstruction loss and classification loss, but also to extract suitable features using a deep autoencoder. Furthermore, for considering the data imbalance problem, we apply an oversampling algorithm, the synthetic minority oversampling technique (SMOTE). For validation of our proposed autoencoder-based feature extraction approach for hematopoietic cancer subtype classification, we compared other traditional feature selection algorithms (principal component analysis, non-negative matrix factorization) and classification algorithms with the SMOTE oversampling approach. Additionally, we used the Shapley Additive exPlanations (SHAP) interpretation technique in our model to explain the important gene/protein for hematopoietic cancer subtype classification. Furthermore, we compared five widely used classification algorithms, including logistic regression, random forest, k-nearest neighbor, artificial neural network and support vector machine. The results of autoencoder-based feature extraction approaches showed good performance, and the best result was the SMOTE oversampling-applied support vector machine algorithm consider both focal loss and reconstruction loss as the loss function for autoencoder (AE) feature selection approach, which produced 97.01% accuracy, 92.60% recall, 99.52% specificity, 93.54% F1-measure, 97.87% G-mean and 95.46% index of balanced accuracy as subtype classification performance measures. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

10 pages, 1034 KiB

Open AccessArticle

Combinatorial K-Means Clustering as a Machine Learning Tool Applied to Diabetes Mellitus Type 2

by Miroslava Nedyalkova, Sergio Madurga and Vasil Simeonov

Int. J. Environ. Res. Public Health 2021, 18(4), 1919; https://doi.org/10.3390/ijerph18041919 - 17 Feb 2021

Cited by 30 | Viewed by 4252

Abstract

A new original procedure based on k-means clustering is designed to find the most appropriate clinical variables able to efficiently separate into groups similar patients diagnosed with diabetes mellitus type 2 (DMT2) and underlying diseases (arterial hypertonia (AH), ischemic heart disease (CHD), diabetic [...] Read more.

A new original procedure based on k-means clustering is designed to find the most appropriate clinical variables able to efficiently separate into groups similar patients diagnosed with diabetes mellitus type 2 (DMT2) and underlying diseases (arterial hypertonia (AH), ischemic heart disease (CHD), diabetic polyneuropathy (DPNP), and diabetic microangiopathy (DMA)). Clustering is a machine learning tool for discovering structures in datasets. Clustering has been proven to be efficient for pattern recognition based on clinical records. The considered combinatorial k-means procedure explores all possible k-means clustering with a determined number of descriptors and groups. The predetermined conditions for the partitioning were as follows: every single group of patients included patients with DMT2 and one of the underlying diseases; each subgroup formed in such a way was subject to partitioning into three patterns (good health status, medium health status, and degenerated health status); optimal descriptors for each disease and groups. The selection of the best clustering is obtained through the parameter called global variance, defined as the sum of all variance values of all clinical variables of all the clusters. The best clinical parameters are found by minimizing this global variance. This methodology has to identify a set of variables that are assumed to separate each underlying disease efficiently in three different subgroups of patients. The hierarchical clustering obtained for these four underlying diseases could be used to build groups of patients with correlated clinical data. The proposed methodology gives surmised results from complex data based on a relationship with the health status of the group and draws a picture of the prediction rate of the ongoing health status. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

13 pages, 965 KiB

Open AccessArticle

Predictors of Contemporary under-5 Child Mortality in Low- and Middle-Income Countries: A Machine Learning Approach

by Andrea Bizzego, Giulio Gabrieli, Marc H. Bornstein, Kirby Deater-Deckard, Jennifer E. Lansford, Robert H. Bradley, Megan Costa and Gianluca Esposito

Int. J. Environ. Res. Public Health 2021, 18(3), 1315; https://doi.org/10.3390/ijerph18031315 - 1 Feb 2021

Cited by 20 | Viewed by 5684

Abstract

Child Mortality (CM) is a worldwide concern, annually affecting as many as 6.81% children in low- and middle-income countries (LMIC). We used data of the Multiple Indicators Cluster Survey (MICS) (N = 275,160) from 27 LMIC and a machine-learning approach to rank 37 [...] Read more.

Child Mortality (CM) is a worldwide concern, annually affecting as many as 6.81% children in low- and middle-income countries (LMIC). We used data of the Multiple Indicators Cluster Survey (MICS) (N = 275,160) from 27 LMIC and a machine-learning approach to rank 37 distal causes of CM and identify the top 10 causes in terms of predictive potency. Based on the top 10 causes, we identified households with improved conditions. We retrospectively validated the results by investigating the association between variations of CM and variations of the percentage of households with improved conditions at country-level, between the 2005–2007 and the 2013–2017 administrations of the MICS. A unique contribution of our approach is to identify lesser-known distal causes which likely account for better-known proximal causes: notably, the identified distal causes and preventable and treatable through social, educational, and physical interventions. We demonstrate how machine learning can be used to obtain operational information from big dataset to guide interventions and policy makers. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

22 pages, 5543 KiB

Open AccessArticle

Feasibility of Using Floor Vibration to Detect Human Falls

by Yu Shao, Xinyue Wang, Wenjie Song, Sobia Ilyas, Haibo Guo and Wen-Shao Chang

Int. J. Environ. Res. Public Health 2021, 18(1), 200; https://doi.org/10.3390/ijerph18010200 - 29 Dec 2020

Cited by 14 | Viewed by 3002

Abstract

With the increasing aging population in modern society, falls as well as fall-induced injuries in elderly people become one of the major public health problems. This study proposes a classification framework that uses floor vibrations to detect fall events as well as distinguish [...] Read more.

With the increasing aging population in modern society, falls as well as fall-induced injuries in elderly people become one of the major public health problems. This study proposes a classification framework that uses floor vibrations to detect fall events as well as distinguish different fall postures. A scaled 3D-printed model with twelve fully adjustable joints that can simulate human body movement was built to generate human fall data. The mass proportion of a human body takes was carefully studied and was reflected in the model. Object drops, human falling tests were carried out and the vibration signature generated in the floor was recorded for analyses. Machine learning algorithms including K-means algorithm and K nearest neighbor algorithm were introduced in the classification process. Three classifiers (human walking versus human fall, human fall versus object drop, human falls from different postures) were developed in this study. Results showed that the three proposed classifiers can achieve the accuracy of 100, 85, and 91%. This paper developed a framework of using floor vibration to build the pattern recognition system in detecting human falls based on a machine learning approach. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

16 pages, 1439 KiB

Open AccessArticle

Screening Model for Estimating Undiagnosed Diabetes among People with a Family History of Diabetes Mellitus: A KNHANES-Based Study

by Kwang Sun Ryu, Ha Ye Jin Kang, Sang Won Lee, Hyun Woo Park, Na Young You, Jae Ho Kim, Yul Hwangbo, Kui Son Choi and Hyo Soung Cha

Int. J. Environ. Res. Public Health 2020, 17(23), 8903; https://doi.org/10.3390/ijerph17238903 - 30 Nov 2020

Cited by 8 | Viewed by 3579

Abstract

A screening model for estimating undiagnosed diabetes mellitus (UDM) is important for early medical care. There is minimal research and a serious lack of screening models for people with a family history of diabetes (FHD), especially one which incorporates gender characteristics. Therefore, the [...] Read more.

A screening model for estimating undiagnosed diabetes mellitus (UDM) is important for early medical care. There is minimal research and a serious lack of screening models for people with a family history of diabetes (FHD), especially one which incorporates gender characteristics. Therefore, the primary objective of our study was to develop a screening model for estimating UDM among people with FHD and enable its validation. We used data from the Korean National Health and Nutrition Examination Survey (KNHANES). KNAHNES (2010–2016) was used as a developmental cohort (n = 5939) and was then evaluated in a validation cohort (n = 1047) KNHANES (2017). We developed the screening model for UDM in male (SMM), female (SMF), and male and female combined (SMP) with FHD using backward stepwise logistic regression analysis. The SMM and SMF showed an appropriate performance (area under curve (AUC) = 76.2% and 77.9%) compared with SMP (AUC = 72.9%) in the validation cohort. Consequently, simple screening models were developed and validated, for the estimation of UDM among patients in the FHD group, which is expected to reduce the burden on the national health care system. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

24 pages, 2859 KiB

Open AccessArticle

Coronary Artery Disease Diagnosis; Ranking the Significant Features Using a Random Trees Model

by Javad Hassannataj Joloudari, Edris Hassannataj Joloudari, Hamid Saadatfar, Mohammad Ghasemigol, Seyyed Mohammad Razavi, Amir Mosavi, Narjes Nabipour, Shahaboddin Shamshirband and Laszlo Nadai

Int. J. Environ. Res. Public Health 2020, 17(3), 731; https://doi.org/10.3390/ijerph17030731 - 23 Jan 2020

Cited by 118 | Viewed by 7297

Abstract

Heart disease is one of the most common diseases in middle-aged citizens. Among the vast number of heart diseases, coronary artery disease (CAD) is considered as a common cardiovascular disease with a high death rate. The most popular tool for diagnosing CAD is [...] Read more.

Heart disease is one of the most common diseases in middle-aged citizens. Among the vast number of heart diseases, coronary artery disease (CAD) is considered as a common cardiovascular disease with a high death rate. The most popular tool for diagnosing CAD is the use of medical imaging, e.g., angiography. However, angiography is known for being costly and also associated with a number of side effects. Hence, the purpose of this study is to increase the accuracy of coronary heart disease diagnosis through selecting significant predictive features in order of their ranking. In this study, we propose an integrated method using machine learning. The machine learning methods of random trees (RTs), decision tree of C5.0, support vector machine (SVM), and decision tree of Chi-squared automatic interaction detection (CHAID) are used in this study. The proposed method shows promising results and the study confirms that the RTs model outperforms other models. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

19 pages, 1841 KiB

Open AccessArticle

Ontology-Based Healthcare Named Entity Recognition from Twitter Messages Using a Recurrent Neural Network Approach

by Erdenebileg Batbaatar and Keun Ho Ryu

Int. J. Environ. Res. Public Health 2019, 16(19), 3628; https://doi.org/10.3390/ijerph16193628 - 27 Sep 2019

Cited by 47 | Viewed by 7934

Abstract

Named Entity Recognition (NER) in the healthcare domain involves identifying and categorizing disease, drugs, and symptoms for biosurveillance, extracting their related properties and activities, and identifying adverse drug events appearing in texts. These tasks are important challenges in healthcare. Analyzing user messages in [...] Read more.

Named Entity Recognition (NER) in the healthcare domain involves identifying and categorizing disease, drugs, and symptoms for biosurveillance, extracting their related properties and activities, and identifying adverse drug events appearing in texts. These tasks are important challenges in healthcare. Analyzing user messages in social media networks such as Twitter can provide opportunities to detect and manage public health events. Twitter provides a broad range of short messages that contain interesting information for information extraction. In this paper, we present a Health-Related Named Entity Recognition (HNER) task using healthcare-domain ontology that can recognize health-related entities from large numbers of user messages from Twitter. For this task, we employ a deep learning architecture which is based on a recurrent neural network (RNN) with little feature engineering. To achieve our goal, we collected a large number of Twitter messages containing health-related information, and detected biomedical entities from the Unified Medical Language System (UMLS). A bidirectional long short-term memory (BiLSTM) model learned rich context information, and a convolutional neural network (CNN) was used to produce character-level features. The conditional random field (CRF) model predicted a sequence of labels that corresponded to a sequence of inputs, and the Viterbi algorithm was used to detect health-related entities from Twitter messages. We provide comprehensive results giving valuable insights for identifying medical entities in Twitter for various applications. The BiLSTM-CRF model achieved a precision of 93.99%, recall of 73.31%, and F1-score of 81.77% for disease or syndrome HNER; a precision of 90.83%, recall of 81.98%, and F1-score of 87.52% for sign or symptom HNER; and a precision of 94.85%, recall of 73.47%, and F1-score of 84.51% for pharmacologic substance named entities. The ontology-based manual annotation results show that it is possible to perform high-quality annotation despite the complexity of medical terminology and the lack of context in tweets. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Graphical abstract

13 pages, 610 KiB

Open AccessArticle

Taming Performance Variability of Healthcare Data Service Frameworks with Proactive and Coarse-Grained Memory Cleaning

by Eunji Lee

Int. J. Environ. Res. Public Health 2019, 16(17), 3096; https://doi.org/10.3390/ijerph16173096 - 26 Aug 2019

Cited by 1 | Viewed by 2726

Abstract

This article explores the performance optimizations of an embedded database memory management system to ensure high responsiveness of real-time healthcare data frameworks. SQLite is a popular embedded database engine extensively used in medical and healthcare data storage systems. However, SQLite is essentially built [...] Read more.

This article explores the performance optimizations of an embedded database memory management system to ensure high responsiveness of real-time healthcare data frameworks. SQLite is a popular embedded database engine extensively used in medical and healthcare data storage systems. However, SQLite is essentially built around lightweight applications in mobile devices, and it significantly deteriorates when a large transaction is issued such as high resolution medical images or massive health dataset, which is unlikely to occur in embedded systems but is quite common in other systems. Such transactions do not fit in the in-memory buffer of SQLite, and SQLite enforces memory reclamation as they are processed. The problem is that the current SQLite buffer management scheme does not effectively manage these cases, and the naïve reclamation scheme used significantly increases the user-perceived latency. Motivated by this limitation, this paper identifies the causes of high latency during processing of a large transaction, and overcomes the limitation via proactive and coarse-grained memory cleaning in SQLite.The proposed memory reclamation scheme was implemented in SQLite 3.29, and measurement studies with a prototype implementation demonstrated that the SQLite operation latency decreases by 13% on an average and up to 17.3% with our memory reclamation scheme as compared to that of the original version. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

21 pages, 8573 KiB

Open AccessArticle

Predicting Methane Concentration in Longwall Regions Using Artificial Neural Networks

by Magdalena Tutak and Jarosław Brodny

Int. J. Environ. Res. Public Health 2019, 16(8), 1406; https://doi.org/10.3390/ijerph16081406 - 18 Apr 2019

Cited by 63 | Viewed by 4443

Abstract

Methane, which is released during mining exploitation, represents a serious threat to this process. This is because the gas may ignite or cause an explosion. Both of these phenomena are extremely dangerous. High levels of methane concentration in mine headings disrupt mining operations [...] Read more.

Methane, which is released during mining exploitation, represents a serious threat to this process. This is because the gas may ignite or cause an explosion. Both of these phenomena are extremely dangerous. High levels of methane concentration in mine headings disrupt mining operations and cause the risk of fire or explosion. Therefore, it is necessary to monitor and predict its concentration in the areas of ongoing mining exploitation. The paper presents the results of tests performed to improve work safety. The article presents the methodology of using artificial neural networks for predicting methane concentration values in one mining area. The objective of the paper is to develop an effective method for forecasting methane concentration in the mining industry. The application of neural networks for this purpose represents one of the first attempts in this respect. The method developed makes use of direct methane concentration values measured by a system of sensors located in the exploitation area. The forecasting model was built on the basis of a Multilayer Perceptron (MLP) network. The corresponding calculations were performed using a three-layered network with non-linear activation functions. The results obtained in the form of methane concentration prediction demonstrated minor errors in relation to the recorded values of this concentration. This offers an opportunity for a broader application of intelligent systems for effective prediction of mining hazards. Full article

(This article belongs to the Special Issue Disease Prediction, Machine Learning, and Healthcare)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Disease Prediction, Machine Learning, and Healthcare

Share This Special Issue

Special Issue Editor

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (22 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI