Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.42557
Noor Fadhil Baqir, Rasha Sabeeh Ahmed, Khaleel Ibraheem Mohsen
Fat suppression magnetic resonance imaging (MRI) sequences are routinely included in the MRI protocol for patients with perianal fistula to improve the visibility of the abnormal tracts and abscesses against the background of hypo-signal intensity on the image. The objective of this study is to compare the turbo inversion recovery magnitude (TIRM) and frequency selective fat saturation turbo spin echo (FSTSE) MRI sequences in detecting perianal fistulas in terms of time and clarity. The MRI protocol included a coronal T2 turbo inversion recovery magnitude sequence, a T2 fat saturation turbo spin echo, and T2 turbo inversion recovery magnitude sequences in the axial plane. The evaluation of sequence image quality involved calculating the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). Additionally three radiologists assessed the best image using a questionnaire designed to align with the study’s objectives. The T2 TIRM sequence was found to have the highest number of ticked images. The inter-rater kappa agreement showed fair agreement (k = 0.370) between the raters. However, the SNR and CNR values for the T2 FSTSE were higher than those of the T2 TIRM sequence, with a p-value less than 0.001. There is a significant difference in the meantime in that the T2 TIRM sequence has less time than the T2 FSTSE with a p-value < 0.001. Due to its uniform fat suppression in the MR image and shorter acquisition time, the turbo inversion recovery magnitude sequence exhibited superior performance compared to the T2 frequency-selective turbo spin echo sequence.
{"title":"Comparison between T2 Turbo Inversion Recovery Magnitude and T2 Frequency Selective Fat Saturation Turbo Spin Echo MRI Sequences in Detection of Perianal Fistula","authors":"Noor Fadhil Baqir, Rasha Sabeeh Ahmed, Khaleel Ibraheem Mohsen","doi":"10.3991/ijoe.v19i14.42557","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.42557","url":null,"abstract":"Fat suppression magnetic resonance imaging (MRI) sequences are routinely included in the MRI protocol for patients with perianal fistula to improve the visibility of the abnormal tracts and abscesses against the background of hypo-signal intensity on the image. The objective of this study is to compare the turbo inversion recovery magnitude (TIRM) and frequency selective fat saturation turbo spin echo (FSTSE) MRI sequences in detecting perianal fistulas in terms of time and clarity. The MRI protocol included a coronal T2 turbo inversion recovery magnitude sequence, a T2 fat saturation turbo spin echo, and T2 turbo inversion recovery magnitude sequences in the axial plane. The evaluation of sequence image quality involved calculating the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). Additionally three radiologists assessed the best image using a questionnaire designed to align with the study’s objectives. The T2 TIRM sequence was found to have the highest number of ticked images. The inter-rater kappa agreement showed fair agreement (k = 0.370) between the raters. However, the SNR and CNR values for the T2 FSTSE were higher than those of the T2 TIRM sequence, with a p-value less than 0.001. There is a significant difference in the meantime in that the T2 TIRM sequence has less time than the T2 FSTSE with a p-value < 0.001. Due to its uniform fat suppression in the MR image and shorter acquisition time, the turbo inversion recovery magnitude sequence exhibited superior performance compared to the T2 frequency-selective turbo spin echo sequence.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136210483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.43163
Cornel Samoila, Doru Ursutiu, Florin Munteanu
The interference of technology in education requires the development of new theories of learning. The paper analyzes connectivism as the most important representative of the theories related to the “digital age.” From the point of view of the environment, called a remote experiment, learning occurs initially at the individual level, encompassing all three classic theories of learning: behaviorism, cognitivism, and constructivism. It shows that the virtual environment has introduced a powerful lever of imbalance for the real environment. This is how we arrived at the explanation of learning theories in real-virtual environments through the theory of chaos or complex environments. Like any knowledge storage network with nodes between which connections can be made, even the remote experiment is subject to random laws. The addition of knowledge is not simply the sum of the effects produced by each individual node (the system is not linear). A distinction is made between information and knowledge. Even if the information in the nodes can be read, this aspect does not represent learning. The remote experiment not only expanded the realm of knowledge but also emphasized the critical role of time. The time remained constant, while the amount of information increased. The teacher, as a knowledge synthesizer, can help orient the student to this vast amount of information, especially when time is limited. Additionally, the student can also play an active role in organizing and systematizing the information. Two examples of experiments are given, which, being inter- and transdisciplinary, can contribute to the introduction of the elements of non-linearity and unpredictability as a method of designing the educational environment, precisely to be able to transform it into a thinking system suitable for the mixture between real and virtual environments in which we live more and more intensely.
{"title":"The Remote Experiment in the Light of the Learning Theories","authors":"Cornel Samoila, Doru Ursutiu, Florin Munteanu","doi":"10.3991/ijoe.v19i14.43163","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.43163","url":null,"abstract":"The interference of technology in education requires the development of new theories of learning. The paper analyzes connectivism as the most important representative of the theories related to the “digital age.” From the point of view of the environment, called a remote experiment, learning occurs initially at the individual level, encompassing all three classic theories of learning: behaviorism, cognitivism, and constructivism. It shows that the virtual environment has introduced a powerful lever of imbalance for the real environment. This is how we arrived at the explanation of learning theories in real-virtual environments through the theory of chaos or complex environments. Like any knowledge storage network with nodes between which connections can be made, even the remote experiment is subject to random laws. The addition of knowledge is not simply the sum of the effects produced by each individual node (the system is not linear). A distinction is made between information and knowledge. Even if the information in the nodes can be read, this aspect does not represent learning. The remote experiment not only expanded the realm of knowledge but also emphasized the critical role of time. The time remained constant, while the amount of information increased. The teacher, as a knowledge synthesizer, can help orient the student to this vast amount of information, especially when time is limited. Additionally, the student can also play an active role in organizing and systematizing the information. Two examples of experiments are given, which, being inter- and transdisciplinary, can contribute to the introduction of the elements of non-linearity and unpredictability as a method of designing the educational environment, precisely to be able to transform it into a thinking system suitable for the mixture between real and virtual environments in which we live more and more intensely.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136210484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.41949
Vielka Mita, Liliana Castillo, José Luis Castillo-Sequera, Lenis Wong
The health sector around the world faces the continuous challenge of improving the services provided to patients. Therefore, digital transformation in health services plays a key role in integrating new technologies such as artificial intelligence. However, the health system in Peru has not yet taken the big step towards digitising its services, currently ranking 71st according to the World Health Organisation (WHO). This article proposes a learning health system for the management and monitoring of private health services in Peru based on the three key components of intelligent health care: (1) a health data platform (HDP); (2) intelligent technologies (IT); and (3) an intelligent health care suite (HIS). The solution consists of four layers: (1) data source, (2) data warehousing, (3) data analytics, and (4) visualization. In layer 1, all data sources are selected to create a database. The proposed learning health system is built, and the data storage is executed through the extract, transform and load (ETL) process in layer 2. In layer 3, the Kaggle dataset and the decision tree (DT) and random forest (RF) algorithms are used to predict the diagnosis of disease, resulting in the RF algorithm having the best performance. Finally, in layer 4, the intelligent health-care suite dashboards and interfaces are designed. The proposed system was applied in a clinic focused on preventing chronic kidney disease. A total of 100 patients and six kidney health experts participated. The results proved that the diagnosis of chronic kidney disease by the learning health system had a low error rate in positive diagnoses (err = 1.12%). Additionally, it was demonstrated that experts were “satisfied” with the dashboards and interfaces of the intelligent health-care suite as well as the quality of the learning health system.
{"title":"A Learning Health-Care System for Improving Renal Health Services in Peru Using Data Analytics","authors":"Vielka Mita, Liliana Castillo, José Luis Castillo-Sequera, Lenis Wong","doi":"10.3991/ijoe.v19i14.41949","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.41949","url":null,"abstract":"The health sector around the world faces the continuous challenge of improving the services provided to patients. Therefore, digital transformation in health services plays a key role in integrating new technologies such as artificial intelligence. However, the health system in Peru has not yet taken the big step towards digitising its services, currently ranking 71st according to the World Health Organisation (WHO). This article proposes a learning health system for the management and monitoring of private health services in Peru based on the three key components of intelligent health care: (1) a health data platform (HDP); (2) intelligent technologies (IT); and (3) an intelligent health care suite (HIS). The solution consists of four layers: (1) data source, (2) data warehousing, (3) data analytics, and (4) visualization. In layer 1, all data sources are selected to create a database. The proposed learning health system is built, and the data storage is executed through the extract, transform and load (ETL) process in layer 2. In layer 3, the Kaggle dataset and the decision tree (DT) and random forest (RF) algorithms are used to predict the diagnosis of disease, resulting in the RF algorithm having the best performance. Finally, in layer 4, the intelligent health-care suite dashboards and interfaces are designed. The proposed system was applied in a clinic focused on preventing chronic kidney disease. A total of 100 patients and six kidney health experts participated. The results proved that the diagnosis of chronic kidney disease by the learning health system had a low error rate in positive diagnoses (err = 1.12%). Additionally, it was demonstrated that experts were “satisfied” with the dashboards and interfaces of the intelligent health-care suite as well as the quality of the learning health system.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136210635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
At present, there is diverse scientific evidence of the contributions of smart environments (SE) that have positively impacted various urban problems. However, the concept of SE is very broad, so it is relevant to investigate how these technological trends have been integrated into the university educational environment. Therefore, the objective of this study is to explore and describe the state of the art on the impact of intelligent environments implemented through the Internet of Things (IoT) in university education. Therefore, a systematic review of the literature was developed. The research was developed with a mixed approach and descriptive scope. From this study, it was determined that the purpose of implementing SE in university education is focused on contributing to the teaching and learning process and managing and optimizing the use of resources provided by the educational environment. In addition, smart classrooms are the type of environments that have been implemented to a greater extent and whose results show a positive impact on indicators such as motivation, participation, interaction, satisfaction, and student attitude. With which it is concluded that universities should reflect on the implementation of institutional policies that lead to the progressive implementation of SE, seeking to transcend from being just simple learning classrooms to sustainable environments that contribute to student health and environmental conservation.
{"title":"Smart Environments through the Internet of Things and Its Impact on University Education: A Systematic Review","authors":"Omar Chamorro-Atalaya, Guillermo Morales-Romero, Adrián Quispe-Andía, Beatriz Caycho-Salas, Primitiva Ramos-Salazar, Elvira Cáceres-Cayllahua, Maritza Arones, Renan Auqui-Ramos","doi":"10.3991/ijoe.v19i14.41531","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.41531","url":null,"abstract":"At present, there is diverse scientific evidence of the contributions of smart environments (SE) that have positively impacted various urban problems. However, the concept of SE is very broad, so it is relevant to investigate how these technological trends have been integrated into the university educational environment. Therefore, the objective of this study is to explore and describe the state of the art on the impact of intelligent environments implemented through the Internet of Things (IoT) in university education. Therefore, a systematic review of the literature was developed. The research was developed with a mixed approach and descriptive scope. From this study, it was determined that the purpose of implementing SE in university education is focused on contributing to the teaching and learning process and managing and optimizing the use of resources provided by the educational environment. In addition, smart classrooms are the type of environments that have been implemented to a greater extent and whose results show a positive impact on indicators such as motivation, participation, interaction, satisfaction, and student attitude. With which it is concluded that universities should reflect on the implementation of institutional policies that lead to the progressive implementation of SE, seeking to transcend from being just simple learning classrooms to sustainable environments that contribute to student health and environmental conservation.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"237 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136063050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.43059
None Omar Hashim Yahya, None Vladimir Vitalievich Alekseev, None Denis Vyacheslavovich Lakomov, None Olga Vladimirovna Fomina, None Irina Sergeevna Iskevich, None Elena Alexandrovna Frolova, None Elena Yurievna Kutimova
Cardiac arrhythmias are abnormalities caused by irregularities in the heart’s electrical conduction system. Cardiovascular diseases (CVD) have been identified as the leading cause of death worldwide. Premature ventricular contraction (PVC) is one of these diseases. It is an arrhythmia that can be linked to a several heart diseases that affect between 40% and 75% of the population. Ventricular bigeminy occurs when one or two premature beats are detected on an electrocardiogram when there is ventricular contraction between two normal heartbeats or trigeminy. The appearance of ventricular bigeminy or trigeminy rhythms is related to angina. Myocardial infarction, hypertension, and congestive heart failure are also possible conditions. Based on deep learning, this paper proposes creating a robust approach for automatically detecting and classifying cardiovascular arrhythmias in long-term electrocardiogram (ECG) recordings from halters based on deep learning (DL). We present a convolutional neural network (CNN) and long-short-time memory (LSTM) model that identifies cardiovascular arrhythmias. We have designed and implemented the proposed model using Python. The model was trained and validated on a database that includes a total of 17 long-recorded ECG signals (24 h) from 17 subjects, which were obtained from Yfa Hospital. The signals were recorded with seven leads holter. The CNN classifier achieved an accuracy of 91.14% as a final result, validated through a 10-fold cross-validation. Moreover, the proposed model was found to be capable of analyzing ECG recordings to classify multiple cardiovascular arrhythmias in the ECG record signals efficiently.
{"title":"Deep Learning Approach for Detecting Cardiovascular Arrhythmias in Seven Lead ECG Signal from Holter","authors":"None Omar Hashim Yahya, None Vladimir Vitalievich Alekseev, None Denis Vyacheslavovich Lakomov, None Olga Vladimirovna Fomina, None Irina Sergeevna Iskevich, None Elena Alexandrovna Frolova, None Elena Yurievna Kutimova","doi":"10.3991/ijoe.v19i14.43059","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.43059","url":null,"abstract":"Cardiac arrhythmias are abnormalities caused by irregularities in the heart’s electrical conduction system. Cardiovascular diseases (CVD) have been identified as the leading cause of death worldwide. Premature ventricular contraction (PVC) is one of these diseases. It is an arrhythmia that can be linked to a several heart diseases that affect between 40% and 75% of the population. Ventricular bigeminy occurs when one or two premature beats are detected on an electrocardiogram when there is ventricular contraction between two normal heartbeats or trigeminy. The appearance of ventricular bigeminy or trigeminy rhythms is related to angina. Myocardial infarction, hypertension, and congestive heart failure are also possible conditions. Based on deep learning, this paper proposes creating a robust approach for automatically detecting and classifying cardiovascular arrhythmias in long-term electrocardiogram (ECG) recordings from halters based on deep learning (DL). We present a convolutional neural network (CNN) and long-short-time memory (LSTM) model that identifies cardiovascular arrhythmias. We have designed and implemented the proposed model using Python. The model was trained and validated on a database that includes a total of 17 long-recorded ECG signals (24 h) from 17 subjects, which were obtained from Yfa Hospital. The signals were recorded with seven leads holter. The CNN classifier achieved an accuracy of 91.14% as a final result, validated through a 10-fold cross-validation. Moreover, the proposed model was found to be capable of analyzing ECG recordings to classify multiple cardiovascular arrhythmias in the ECG record signals efficiently.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136063054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.41065
Yanda Sailaja, Velumurugan Pattani
Ischemic stroke is a life-threatening disorder that significantly reduces a person’s lifespan. The timely diagnosis of stroke heavily relies on medical imaging techniques such as magnetic resonance imaging (MRI), computerized tomography (CT), and x-ray imaging. However, the manual localization and analysis of these images can be time-consuming and yield less accurate results. To address this challenge, we propose the implementation of deep-learning object detection techniques for computerized lesion identification in medical images. In this study, we employ three categories of deep learning object identification networks: deep convolutional neural network (DCNN), you only look once (YOLO) 5, and single-shot detector (SSD). By leveraging these advanced deep learning models, we aim to reduce the effort and time required for screening and analyzing a significant number of daily medical images, including MRI, CT, and x-ray images. With the addition of YOLO5 and SSD among these networks, the accuracy achieved was 96.43%, demonstrating their effectiveness in accurately identifying lesions associated with ischemic stroke.
{"title":"Effective Brain Stroke Prediction with Deep Learning Model by Incorporating YOLO_5 and SSD","authors":"Yanda Sailaja, Velumurugan Pattani","doi":"10.3991/ijoe.v19i14.41065","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.41065","url":null,"abstract":"Ischemic stroke is a life-threatening disorder that significantly reduces a person’s lifespan. The timely diagnosis of stroke heavily relies on medical imaging techniques such as magnetic resonance imaging (MRI), computerized tomography (CT), and x-ray imaging. However, the manual localization and analysis of these images can be time-consuming and yield less accurate results. To address this challenge, we propose the implementation of deep-learning object detection techniques for computerized lesion identification in medical images. In this study, we employ three categories of deep learning object identification networks: deep convolutional neural network (DCNN), you only look once (YOLO) 5, and single-shot detector (SSD). By leveraging these advanced deep learning models, we aim to reduce the effort and time required for screening and analyzing a significant number of daily medical images, including MRI, CT, and x-ray images. With the addition of YOLO5 and SSD among these networks, the accuracy achieved was 96.43%, demonstrating their effectiveness in accurately identifying lesions associated with ischemic stroke.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136209705","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.43761
None Alaa Aljazara, None Nadine Abu Tuhaimer, None Ahmed Alawwad, None Khalid Bani Hani, None Abdallah D. Qusef, None Najeh Rajeh Alsalhi, None Aras Al-Dawoodi
3D printers are known for providing parts with relatively good accuracy. However, the level of accuracy in the dimensions of printed objects may not matter if they do not have a mechanical purpose. When multiple 3D-printed parts are intended to be integrated with each other to create a larger system, even a fraction of a millimeter can have a significant impact on the entire system. This study aims to investigate the variation in dimension when a single print file is replicated using the same slicing settings. The findings are then analyzed using quality control tools and compared to the designed measurements. Fused deposition modeling (FDM) technology or fused filament fabrication (FFF) technology was chosen for this study due to its availability to the common user, its relatively low cost, and its increasing popularity in different applications and industries. The material used in this study is polylactic acid (PLA) which is a thermoplastic and the most widely used plastic filament in 3D printing. It has a low melting point, high strength, low thermal expansion, and is relatively cheap. The dimensional accuracy of FDM-produced parts was evaluated by comparing the dimensions of the fabricated specimens with their computer-aided design (CAD) models. Statistical analysis revealed that the mean dimensional deviations were within the specified tolerance limits for most of the tested parts. This suggests that FDM technology is reliable in terms of achieving dimensional accuracy.
{"title":"Quality of 3D Printed Objects Using Fused Deposition Modeling (FDM) Technology in Terms of Dimensional Accuracy","authors":"None Alaa Aljazara, None Nadine Abu Tuhaimer, None Ahmed Alawwad, None Khalid Bani Hani, None Abdallah D. Qusef, None Najeh Rajeh Alsalhi, None Aras Al-Dawoodi","doi":"10.3991/ijoe.v19i14.43761","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.43761","url":null,"abstract":"3D printers are known for providing parts with relatively good accuracy. However, the level of accuracy in the dimensions of printed objects may not matter if they do not have a mechanical purpose. When multiple 3D-printed parts are intended to be integrated with each other to create a larger system, even a fraction of a millimeter can have a significant impact on the entire system. This study aims to investigate the variation in dimension when a single print file is replicated using the same slicing settings. The findings are then analyzed using quality control tools and compared to the designed measurements. Fused deposition modeling (FDM) technology or fused filament fabrication (FFF) technology was chosen for this study due to its availability to the common user, its relatively low cost, and its increasing popularity in different applications and industries. The material used in this study is polylactic acid (PLA) which is a thermoplastic and the most widely used plastic filament in 3D printing. It has a low melting point, high strength, low thermal expansion, and is relatively cheap. The dimensional accuracy of FDM-produced parts was evaluated by comparing the dimensions of the fabricated specimens with their computer-aided design (CAD) models. Statistical analysis revealed that the mean dimensional deviations were within the specified tolerance limits for most of the tested parts. This suggests that FDM technology is reliable in terms of achieving dimensional accuracy.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136063051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.41291
Solomon Oluwole Akinola, Wang Qingguo, Peter Olukanmi, Marwala Tshilidzi
In recent years, there has been an increase in studies on time-series forecasting for the future occurrence of disease incidents. Improvements in deep learning approaches offer techniques for modelling long-term temporal relationships. Nonetheless, this design practice is rigorously painstaking, prone to errors, and requires human expertise. The advent of feature enrichment with automatic architecture search typically optimises the discovery of new neural architectures applicable in domains such as time-series modelling. The main methodological contribution of this study is an approach for time-series forecasting using feature-enriched filters and an evolutionary neural architecture search with sequence-to-sequence gated recurrent units (GRU-Seq2Seq). This is applied to the prediction of daily cases of coronavirus disease in South Africa. The highly pathogenic coronavirus pandemic incident data was modelled with filters, optimised hyper-parameter search trials and an evolutional neural algorithm. The proposed model was benchmarked against ARIMA and SARIMA. The model predicted trends for 30, 60 and 90-day horizons and evaluated them for 7, 14 and 31 days. Simulation results demonstrate that observed daily case counts with added filters and evolutionary search optimisation for forecasting improve performance accuracy. Generally, the proposed bFilter+GRU-Seq2Seq with optimal search configuration outperformed ARIMA and SARIMA with lower error scores and higher performance metrics, with an R2 score of 7.48E-01 for a 30-day forecast horizon.
{"title":"A Boosted Evolutionary Neural Architecture Search for Timeseries Forecasting with Application to South African COVID-19 Cases","authors":"Solomon Oluwole Akinola, Wang Qingguo, Peter Olukanmi, Marwala Tshilidzi","doi":"10.3991/ijoe.v19i14.41291","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.41291","url":null,"abstract":"In recent years, there has been an increase in studies on time-series forecasting for the future occurrence of disease incidents. Improvements in deep learning approaches offer techniques for modelling long-term temporal relationships. Nonetheless, this design practice is rigorously painstaking, prone to errors, and requires human expertise. The advent of feature enrichment with automatic architecture search typically optimises the discovery of new neural architectures applicable in domains such as time-series modelling. The main methodological contribution of this study is an approach for time-series forecasting using feature-enriched filters and an evolutionary neural architecture search with sequence-to-sequence gated recurrent units (GRU-Seq2Seq). This is applied to the prediction of daily cases of coronavirus disease in South Africa. The highly pathogenic coronavirus pandemic incident data was modelled with filters, optimised hyper-parameter search trials and an evolutional neural algorithm. The proposed model was benchmarked against ARIMA and SARIMA. The model predicted trends for 30, 60 and 90-day horizons and evaluated them for 7, 14 and 31 days. Simulation results demonstrate that observed daily case counts with added filters and evolutionary search optimisation for forecasting improve performance accuracy. Generally, the proposed bFilter+GRU-Seq2Seq with optimal search configuration outperformed ARIMA and SARIMA with lower error scores and higher performance metrics, with an R2 score of 7.48E-01 for a 30-day forecast horizon.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136210796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.42163
Abdulkareem Z. Mohammed, None Loay E. George
This paper presents a system for diagnosing osteoporosis using x-rays by leveraging transfer learning in the same domain. The proposed system consists of phase 1 and phase 2; each phase includes several stages, as the pre-processing stage appropriately prepares the source image via noise reduction by the average filter, contrast enhancement using histogram equalization, and obtaining the region of interest by employing K-mean and edge detection, followed by the smudging stage through a mean filter with a large window size, which subsequently contributed to facilitating the diagnosis. The stages mentioned in both phases are similar. In phase 1, the model is trained on a large unlabeled x-ray dataset collected from different orthopedic centers to identify the general features of the image. In phase 2, fine-tune the trained model with the target dataset; this approach is beneficial when the target task has limited labeled data or when training a model from scratch is computationally expensive. It is worth noting that two datasets were used as target datasets. The accuracy of diagnosing osteoporosis using the proposed deep convolutional neural network (DCNN) model was 94.5 with the osteoporosis knee x-ray database (Dataset A). The accuracy of diagnosing osteoporosis using transfer learning in the same field was 98.91 when training the proposed DCNN model with a large unlabeled dataset and fine-tuning with the target database, osteoporosis knee x-ray database (Dataset A). The accuracy of diagnosing osteoporosis using the proposed DCNN model was 91.5 with the knee x-ray osteoporosis database (Dataset B). The accuracy of diagnosing osteoporosis using transfer learning in the same field was 96.61 when training the proposed DCNN model with a large unlabeled dataset and fine-tuning with the target knee x-ray osteoporosis database (Dataset B).
{"title":"Diagnosis of Osteoporosis Using Transfer Learning in the Same Domain","authors":"Abdulkareem Z. Mohammed, None Loay E. George","doi":"10.3991/ijoe.v19i14.42163","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.42163","url":null,"abstract":"This paper presents a system for diagnosing osteoporosis using x-rays by leveraging transfer learning in the same domain. The proposed system consists of phase 1 and phase 2; each phase includes several stages, as the pre-processing stage appropriately prepares the source image via noise reduction by the average filter, contrast enhancement using histogram equalization, and obtaining the region of interest by employing K-mean and edge detection, followed by the smudging stage through a mean filter with a large window size, which subsequently contributed to facilitating the diagnosis. The stages mentioned in both phases are similar. In phase 1, the model is trained on a large unlabeled x-ray dataset collected from different orthopedic centers to identify the general features of the image. In phase 2, fine-tune the trained model with the target dataset; this approach is beneficial when the target task has limited labeled data or when training a model from scratch is computationally expensive. It is worth noting that two datasets were used as target datasets. The accuracy of diagnosing osteoporosis using the proposed deep convolutional neural network (DCNN) model was 94.5 with the osteoporosis knee x-ray database (Dataset A). The accuracy of diagnosing osteoporosis using transfer learning in the same field was 98.91 when training the proposed DCNN model with a large unlabeled dataset and fine-tuning with the target database, osteoporosis knee x-ray database (Dataset A). The accuracy of diagnosing osteoporosis using the proposed DCNN model was 91.5 with the knee x-ray osteoporosis database (Dataset B). The accuracy of diagnosing osteoporosis using transfer learning in the same field was 96.61 when training the proposed DCNN model with a large unlabeled dataset and fine-tuning with the target knee x-ray osteoporosis database (Dataset B).","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136210349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-11DOI: 10.3991/ijoe.v19i14.42725
Hicham Benradi, Issam Bouganssa, Ahmed Chater, Abdelali Lasfar
Medical imaging treatment is one of the best-known computer science disciplines. It can be used to detect the presence of several diseases such as skin cancer and brain tumors, and since the arrival of the coronavirus (COVID-19), this technique has been used to alleviate the heavy burden placed on all health institutions and personnel, given the high rate of spread of this virus in the population. One of the problems encountered in diagnosing people suspected of having contracted COVID-19 is the difficulty of distinguishing symptoms due to this virus from those of other diseases such as influenza, as they are similar. This paper proposes a new approach to distinguishing between lung diseases and COVID-19 by analyzing chest x-ray images using a convolutional neural network (CNN) architecture. To achieve this, pre-processing was carried out on the dataset using histogram equalization, and then we trained two sub-datasets from the dataset using the Train et Test, the first to be used in the training phase and the second to be used in the model validation phase. Then a CNN architecture composed of several convolution layers and fully connected layers was deployed to train our model. Finally, we evaluated our model using two different metrics: the confusion matrix and the receiver operating characteristic. The simulation results recorded are satisfactory, with an accuracy rate of 96.27%.
医学影像治疗是最著名的计算机科学学科之一。它可用于检测皮肤癌和脑肿瘤等几种疾病的存在，自冠状病毒(COVID-19)出现以来，鉴于该病毒在人群中的高传播率，该技术已被用于减轻所有卫生机构和人员的沉重负担。在诊断疑似COVID-19患者时遇到的问题之一是，很难将这种病毒引起的症状与流感等其他疾病引起的症状区分开来，因为它们相似。本文提出了一种利用卷积神经网络(CNN)架构分析胸部x线图像，区分肺部疾病和COVID-19的新方法。为了实现这一点，使用直方图均衡化对数据集进行预处理，然后我们使用Train et Test从数据集中训练两个子数据集，第一个用于训练阶段，第二个用于模型验证阶段。然后部署由多个卷积层和全连接层组成的CNN架构来训练我们的模型。最后，我们使用两个不同的指标来评估我们的模型:混淆矩阵和接收器工作特性。所记录的仿真结果令人满意，准确率达到96.27%。
{"title":"Discriminative Approach Lung Diseases and COVID-19 from Chest X-Ray Images Using Convolutional Neural Networks: A Promising Approach for Accurate Diagnosis","authors":"Hicham Benradi, Issam Bouganssa, Ahmed Chater, Abdelali Lasfar","doi":"10.3991/ijoe.v19i14.42725","DOIUrl":"https://doi.org/10.3991/ijoe.v19i14.42725","url":null,"abstract":"Medical imaging treatment is one of the best-known computer science disciplines. It can be used to detect the presence of several diseases such as skin cancer and brain tumors, and since the arrival of the coronavirus (COVID-19), this technique has been used to alleviate the heavy burden placed on all health institutions and personnel, given the high rate of spread of this virus in the population. One of the problems encountered in diagnosing people suspected of having contracted COVID-19 is the difficulty of distinguishing symptoms due to this virus from those of other diseases such as influenza, as they are similar. This paper proposes a new approach to distinguishing between lung diseases and COVID-19 by analyzing chest x-ray images using a convolutional neural network (CNN) architecture. To achieve this, pre-processing was carried out on the dataset using histogram equalization, and then we trained two sub-datasets from the dataset using the Train et Test, the first to be used in the training phase and the second to be used in the model validation phase. Then a CNN architecture composed of several convolution layers and fully connected layers was deployed to train our model. Finally, we evaluated our model using two different metrics: the confusion matrix and the receiver operating characteristic. The simulation results recorded are satisfactory, with an accuracy rate of 96.27%.","PeriodicalId":36900,"journal":{"name":"International Journal of Online and Biomedical Engineering","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136062824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：