For more efficient training, it is of great significance to know the position-power relationship of riders of different types and genders on different venues for guiding riders’ training. A number of previous studies have used pacing strategies to examine the impact of different venues on rider decision-making. Firstly, according to the test data of different athletes, the OmPD model is used to establish the rider’s own power profile. Through force analysis, after discretization, the relationship between power and position is numerically simulated. In addition, the limitation of anaerobic work ability to decision-making is added. In order to facilitate the calculation, the two-dimensional situation is considered first, and then the three-dimensional situation of the turning is corrected separately. For different regions and courses, after determining the local environmental parameters according to the data, the optimization goal is to take the shortest time after spline interpolation. Anaerobic working capacity and maximum power are the constraints. The optimal numerical solution is carried out by using Method of Moving Asymptotes. The 2020 Olympic Games and the 2021 UCI World Championship are simulated, and the power-position curves are obtained.
{"title":"Strategy Arrangement of Road Cycling Individual Time Trial Based on Topology Optimization Algorithm","authors":"Yuelin Xu, Yue Dai, Haonan Zhang","doi":"10.1145/3545729.3545778","DOIUrl":"https://doi.org/10.1145/3545729.3545778","url":null,"abstract":"For more efficient training, it is of great significance to know the position-power relationship of riders of different types and genders on different venues for guiding riders’ training. A number of previous studies have used pacing strategies to examine the impact of different venues on rider decision-making. Firstly, according to the test data of different athletes, the OmPD model is used to establish the rider’s own power profile. Through force analysis, after discretization, the relationship between power and position is numerically simulated. In addition, the limitation of anaerobic work ability to decision-making is added. In order to facilitate the calculation, the two-dimensional situation is considered first, and then the three-dimensional situation of the turning is corrected separately. For different regions and courses, after determining the local environmental parameters according to the data, the optimization goal is to take the shortest time after spline interpolation. Anaerobic working capacity and maximum power are the constraints. The optimal numerical solution is carried out by using Method of Moving Asymptotes. The 2020 Olympic Games and the 2021 UCI World Championship are simulated, and the power-position curves are obtained.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128128518","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}
Raoul D'Alessio, Teresa Angela Trunfio, M. Leonetti, A. Laino, R. Deli, L. Galantucci
Introduction. The beauty of the face is a mystery well rooted in history. Several studies have followed over the years to search for objective characteristics that help define facial attractiveness. Photogrammetry is a valid tool, also used in clinical practice, to acquire multiple images of the face, which, through a special software, can be processed. Methods. In this study, starting from the 3D reconstruction of the faces of 7 women considered attractive at national level, i.e. finalists in the years 2019 and 2020 in the national beauty contest Miss Italia, and 7 women considered attractive by medical experts, 58 reference points were acquired and from these were obtained 10 linear measures and 5 angular measures. The U-Mann Whitney test was used to compare the two groups. Results. The data, analyzed with the use of statistical test, show that there are no significant differences between the measurements of the two samples. Conclusions. The study confirms the validity of the judgment provided by medical experts as well as the various selection steps made for the competition. Furthermore, it is possible to conclude that there are features that conventionally, for this nationality and in this era, are recognized as a standard of facial attractiveness.
{"title":"Facial attractiveness: are there features recognized as a standard?","authors":"Raoul D'Alessio, Teresa Angela Trunfio, M. Leonetti, A. Laino, R. Deli, L. Galantucci","doi":"10.1145/3545729.3545735","DOIUrl":"https://doi.org/10.1145/3545729.3545735","url":null,"abstract":"Introduction. The beauty of the face is a mystery well rooted in history. Several studies have followed over the years to search for objective characteristics that help define facial attractiveness. Photogrammetry is a valid tool, also used in clinical practice, to acquire multiple images of the face, which, through a special software, can be processed. Methods. In this study, starting from the 3D reconstruction of the faces of 7 women considered attractive at national level, i.e. finalists in the years 2019 and 2020 in the national beauty contest Miss Italia, and 7 women considered attractive by medical experts, 58 reference points were acquired and from these were obtained 10 linear measures and 5 angular measures. The U-Mann Whitney test was used to compare the two groups. Results. The data, analyzed with the use of statistical test, show that there are no significant differences between the measurements of the two samples. Conclusions. The study confirms the validity of the judgment provided by medical experts as well as the various selection steps made for the competition. Furthermore, it is possible to conclude that there are features that conventionally, for this nationality and in this era, are recognized as a standard of facial attractiveness.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"264 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133849931","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}
The ankle rehabilitation robot is essential equipment for patients with clubfoot and talipes valgus to make up for deficiencies of the manual rehabilitation training and reduce the workload of rehabilitation physicians. Based on the physiological model of the ankle joint and the requirement of rehabilitation in physical therapy, an ankle rehabilitation parallel mechanism has three limbs with a universal joint, rotation joint, and spherical joint configures (3-URS ARPM), which had 6-DOF was analyzed and developed. The inverse kinematics problem of 3-URS ARPM was then solved using GRG optimization methods combined with the Banana objective function. As a result, six control solutions of the inverse kinematics of 3-URS ARPM are obtained. Furthermore, the forward kinematics problem is also analyzed through optimization approaches suitable for motor position control. Finally, the kinematic control characteristic of joints variable for 3-URS ARPM is presented in detail, comparing its motion range to the ADAMS software. The numerical simulation results showed an excellent smooth trajectory tracking in real-time control, indicating that this mechanism for ankle rehabilitation with a simple structure has precise control characteristics with the accuracy achieved is . Hence, the developed 3-URS ARPM can be applied to ankle rehabilitation widely.
{"title":"Design and Kinematics Analysis of 3-URS Ankle Rehabilitation Parallel Robot","authors":"ThanhTrung Trang, Yueming Hu, Thanh-Long Pham, Quoc-Khanh Duong","doi":"10.1145/3545729.3545764","DOIUrl":"https://doi.org/10.1145/3545729.3545764","url":null,"abstract":"The ankle rehabilitation robot is essential equipment for patients with clubfoot and talipes valgus to make up for deficiencies of the manual rehabilitation training and reduce the workload of rehabilitation physicians. Based on the physiological model of the ankle joint and the requirement of rehabilitation in physical therapy, an ankle rehabilitation parallel mechanism has three limbs with a universal joint, rotation joint, and spherical joint configures (3-URS ARPM), which had 6-DOF was analyzed and developed. The inverse kinematics problem of 3-URS ARPM was then solved using GRG optimization methods combined with the Banana objective function. As a result, six control solutions of the inverse kinematics of 3-URS ARPM are obtained. Furthermore, the forward kinematics problem is also analyzed through optimization approaches suitable for motor position control. Finally, the kinematic control characteristic of joints variable for 3-URS ARPM is presented in detail, comparing its motion range to the ADAMS software. The numerical simulation results showed an excellent smooth trajectory tracking in real-time control, indicating that this mechanism for ankle rehabilitation with a simple structure has precise control characteristics with the accuracy achieved is . Hence, the developed 3-URS ARPM can be applied to ankle rehabilitation widely.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127131892","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}
E. Montella, Marta Rosaria Marino, Massimo Majolo, E. Raiola, Giuseppe Russo, G. Longo, A. Lombardi, A. Borrelli, M. Triassi
In recent years, the use of caesarean sections (CS) has grown, leading more women, especially in developed countries, to choose it as a preferential modality, even without clear clinical needs. Although Caesarean sections are effective in reducing maternal and infant mortality, they can cause significant and sometimes permanent complications. The increase in the CS rate, increases hospital stay and therefore hospital costs. Being able to analyze and even predict the length of stay (LOS) for a rapidly growing procedure becomes a valuable information resource for healthcare managers. The purpose of this study is to study LOS for all patients undergoing CS both in the "San Giovanni di Dio e Ruggi d'Aragona" University Hospital of Salerno and in the A.O.R.N. "Antonio Cardarelli" of Naples. With multiple linear regression analysis and machine learning algorithms we can create a model for LOS prediction.
近年来，剖腹产的使用有所增加，导致更多的妇女，特别是在发达国家，即使没有明确的临床需要，也将其作为一种优先方式选择。虽然剖腹产在降低孕产妇和婴儿死亡率方面是有效的，但它们可能导致严重的，有时是永久性的并发症。CS比率的增加增加了住院时间，从而增加了医院费用。能够分析甚至预测快速增长的手术的住院时间(LOS)成为医疗保健管理人员的宝贵信息资源。本研究的目的是研究萨勒诺“San Giovanni di Dio e Ruggi d'Aragona”大学医院和a.o.r.n接受CS的所有患者的LOS那不勒斯的安东尼奥·卡达雷利。通过多元线性回归分析和机器学习算法，我们可以创建一个LOS预测模型。
{"title":"Regression and classification methods for predicting the length of hospital stay after cesarean section: a bicentric study","authors":"E. Montella, Marta Rosaria Marino, Massimo Majolo, E. Raiola, Giuseppe Russo, G. Longo, A. Lombardi, A. Borrelli, M. Triassi","doi":"10.1145/3545729.3545757","DOIUrl":"https://doi.org/10.1145/3545729.3545757","url":null,"abstract":"In recent years, the use of caesarean sections (CS) has grown, leading more women, especially in developed countries, to choose it as a preferential modality, even without clear clinical needs. Although Caesarean sections are effective in reducing maternal and infant mortality, they can cause significant and sometimes permanent complications. The increase in the CS rate, increases hospital stay and therefore hospital costs. Being able to analyze and even predict the length of stay (LOS) for a rapidly growing procedure becomes a valuable information resource for healthcare managers. The purpose of this study is to study LOS for all patients undergoing CS both in the \"San Giovanni di Dio e Ruggi d'Aragona\" University Hospital of Salerno and in the A.O.R.N. \"Antonio Cardarelli\" of Naples. With multiple linear regression analysis and machine learning algorithms we can create a model for LOS prediction.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129988701","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}
Noratikah Nordin, Z. Zainol, M. H. M. Noor, Chan Lai Fong
In the healthcare setting, suicidal behavior prediction plays an important role in clinical decision making due to the suicide rate increasing day by day contributes to a decrease in productivity and increase in national expenditure. Several machine learning models are being developed to generate accurate predictions in a suicide attempt. However, there is a lack of interpretability, explainability and transparency with these predictive models. Therefore, the aim of this study is to improve explanations of machine learning models for predicting suicidal behavior based on clinical data using the Shapley Additive exPlanations (SHAP) approach. The experiment shows that machine learning models with SHAP are able to interpret and understand the nature of an individual's predictions of suicidal behavior.
{"title":"Explainable Machine Learning Models for Suicidal Behavior Prediction","authors":"Noratikah Nordin, Z. Zainol, M. H. M. Noor, Chan Lai Fong","doi":"10.1145/3545729.3545754","DOIUrl":"https://doi.org/10.1145/3545729.3545754","url":null,"abstract":"In the healthcare setting, suicidal behavior prediction plays an important role in clinical decision making due to the suicide rate increasing day by day contributes to a decrease in productivity and increase in national expenditure. Several machine learning models are being developed to generate accurate predictions in a suicide attempt. However, there is a lack of interpretability, explainability and transparency with these predictive models. Therefore, the aim of this study is to improve explanations of machine learning models for predicting suicidal behavior based on clinical data using the Shapley Additive exPlanations (SHAP) approach. The experiment shows that machine learning models with SHAP are able to interpret and understand the nature of an individual's predictions of suicidal behavior.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129629271","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}
Diana Guadalupe González-Rodríguez, D. Martinez-Peon, Xochitl Angélica Ortiz Jiménez, J. F. Góngora-Rivera, F. Benavides-Bravo, Brayan Soria Rodríguez, Mariana Ruíz Velázquez
Numerous studies are tackling mental fatigue due to the high number of accidents caused by mental fatigue. The requirement of real-time detection allows that EEG measurement is one of the most feasible options because it is non-invasive and low cost, against other techniques. In this paper, it is proposed a drivers’ mental fatigue detection through EEG signals using the Hurst exponent, because of the simple calculations to obtain it and the fractal nature of EEG signals, and it was compared with the event-related desynchronization/ synchronization (ERD/ERS). Frontocentral (FC3) right and left parietal (P3 and P4) regions in the alpha-band, which are regions where mental fatigue is detected, were analyzed. The task was divided into 3 blocks at 2, 25, and 40 minutes. The results for ERD/ERS in block 2 (25 minutes) showed a desynchronization in electrode FC3 and synchronization in electrodes P3 and P4, these changes indicate the subjects presented mental fatigue in that block. The results using Hurst’s exponent showed for block 2 that persistence decays at electrode FC3, while for electrodes P3 and P4 persistence increases. The results found for ERD/ERS and the Hurst exponent showed a positive correlation in block 2, which is where the first symptoms of mental fatigue appear. It is concluded that the Hurst exponent can be a potential tool that can be used as an indicator to detect mental fatigue.
{"title":"EEG-Based Drivers Mental Fatigue Detection Using ERD/ERS and Hurst Exponent","authors":"Diana Guadalupe González-Rodríguez, D. Martinez-Peon, Xochitl Angélica Ortiz Jiménez, J. F. Góngora-Rivera, F. Benavides-Bravo, Brayan Soria Rodríguez, Mariana Ruíz Velázquez","doi":"10.1145/3545729.3545763","DOIUrl":"https://doi.org/10.1145/3545729.3545763","url":null,"abstract":"Numerous studies are tackling mental fatigue due to the high number of accidents caused by mental fatigue. The requirement of real-time detection allows that EEG measurement is one of the most feasible options because it is non-invasive and low cost, against other techniques. In this paper, it is proposed a drivers’ mental fatigue detection through EEG signals using the Hurst exponent, because of the simple calculations to obtain it and the fractal nature of EEG signals, and it was compared with the event-related desynchronization/ synchronization (ERD/ERS). Frontocentral (FC3) right and left parietal (P3 and P4) regions in the alpha-band, which are regions where mental fatigue is detected, were analyzed. The task was divided into 3 blocks at 2, 25, and 40 minutes. The results for ERD/ERS in block 2 (25 minutes) showed a desynchronization in electrode FC3 and synchronization in electrodes P3 and P4, these changes indicate the subjects presented mental fatigue in that block. The results using Hurst’s exponent showed for block 2 that persistence decays at electrode FC3, while for electrodes P3 and P4 persistence increases. The results found for ERD/ERS and the Hurst exponent showed a positive correlation in block 2, which is where the first symptoms of mental fatigue appear. It is concluded that the Hurst exponent can be a potential tool that can be used as an indicator to detect mental fatigue.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131952940","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}
This paper aims to study the care burden and influencing factors of the family caregivers who work for disabled elderly. From January 2021 to April 2021, 266 caregivers of disabled elderly from four communities in Dalian (Both deputy province and Independent Plan city in P.R.China) were enrolled as respondents in this study. All the respondents were evaluated by general data questionnaire and improved Zarit Caregiver Burden Interview(ZBI), then analyzed and interpreted by the main influencing factors of care burden of the disabled-elders’ family caregivers. First, it is shown that the care burden of disabled-elders’ family caregivers are generally at a moderate level, and they often work with heaviness. Employing the multivariate linear analysis in statistics, the second founding appears that age, education level, average daily care time were positively correlated with the care burden of disabled-elders’ family caregivers, while there is a negative correlation between the number of children, the self-care ability of the elders themselves and the burden of their family caregivers. Finally, the family care of disabled-elders is mainly provided by their family members, whose order of main family caregivers are daughters, spouses, daughter-in-law and sons in turn. The care burden of disabled-elders’ caregivers is far more heavier, but current pension policy has not made corresponding arrangements for the special care of disabled elderly. The government ought to pay more attention to the situation of family caregivers of disabled-elders and increase the support by establishing a long-term family care system for disabled elderly.
{"title":"Study on the Care Burden and Influencing Factors of the Family Caregivers of Disabled Elders in Dalian","authors":"Xu Chen","doi":"10.1145/3545729.3545785","DOIUrl":"https://doi.org/10.1145/3545729.3545785","url":null,"abstract":"This paper aims to study the care burden and influencing factors of the family caregivers who work for disabled elderly. From January 2021 to April 2021, 266 caregivers of disabled elderly from four communities in Dalian (Both deputy province and Independent Plan city in P.R.China) were enrolled as respondents in this study. All the respondents were evaluated by general data questionnaire and improved Zarit Caregiver Burden Interview(ZBI), then analyzed and interpreted by the main influencing factors of care burden of the disabled-elders’ family caregivers. First, it is shown that the care burden of disabled-elders’ family caregivers are generally at a moderate level, and they often work with heaviness. Employing the multivariate linear analysis in statistics, the second founding appears that age, education level, average daily care time were positively correlated with the care burden of disabled-elders’ family caregivers, while there is a negative correlation between the number of children, the self-care ability of the elders themselves and the burden of their family caregivers. Finally, the family care of disabled-elders is mainly provided by their family members, whose order of main family caregivers are daughters, spouses, daughter-in-law and sons in turn. The care burden of disabled-elders’ caregivers is far more heavier, but current pension policy has not made corresponding arrangements for the special care of disabled elderly. The government ought to pay more attention to the situation of family caregivers of disabled-elders and increase the support by establishing a long-term family care system for disabled elderly.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"148 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131743880","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}
Electronic health records, as a repository, of patient information, is nowadays the most commonly used technology to record , store and review patient clinical records and perform other clinical tasks. However, the accurate identification and retrieval of relevant information from clinical records is a difficult task due to the unstructured nature of clinical documents, characterized in particular by lack of clear structure. Therefore, medical practice is facing a challenge thanks to rapid growth of health information in electronic health records (EHRs), mostly in narrative text form. As a result, it's becoming important to effectively manage the growing amount of data for a single patient and there is currently a requirement to visualize electronic health records (EHRs) in a way that aids physicians in clinical tasks and medical decision-making. Leveraging text visualization techniques to unstructured clinical narrative texts is a new area of research that aims to provide better information extraction and retrieval to support clinical decision support in scenarios where data generated continues to grow. Clinical datasets in electronic health records (EHR) offer a lot of potential for training accurate statistical models to classify facets of information which can then be used to improve patient care and outcomes. However, in many clinical note datasets, unstructured nature of clinical texts is a common problem. This paper examines to the very issue of getting raw clinical texts and mapping them into meaningful structures that can support healthcare professional utilizes narrative texts. Our work is the result of a collaborative design process that was aided by empirical data collected through formal usability testing.
{"title":"Using classification and visualization to support clinical texts review in electronic clinical documentation","authors":"Jonah Kenei, E. Opiyo","doi":"10.1145/3545729.3545746","DOIUrl":"https://doi.org/10.1145/3545729.3545746","url":null,"abstract":"Electronic health records, as a repository, of patient information, is nowadays the most commonly used technology to record , store and review patient clinical records and perform other clinical tasks. However, the accurate identification and retrieval of relevant information from clinical records is a difficult task due to the unstructured nature of clinical documents, characterized in particular by lack of clear structure. Therefore, medical practice is facing a challenge thanks to rapid growth of health information in electronic health records (EHRs), mostly in narrative text form. As a result, it's becoming important to effectively manage the growing amount of data for a single patient and there is currently a requirement to visualize electronic health records (EHRs) in a way that aids physicians in clinical tasks and medical decision-making. Leveraging text visualization techniques to unstructured clinical narrative texts is a new area of research that aims to provide better information extraction and retrieval to support clinical decision support in scenarios where data generated continues to grow. Clinical datasets in electronic health records (EHR) offer a lot of potential for training accurate statistical models to classify facets of information which can then be used to improve patient care and outcomes. However, in many clinical note datasets, unstructured nature of clinical texts is a common problem. This paper examines to the very issue of getting raw clinical texts and mapping them into meaningful structures that can support healthcare professional utilizes narrative texts. Our work is the result of a collaborative design process that was aided by empirical data collected through formal usability testing.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115871334","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}
William Yu Chung Wang, Philip Hong Wei Jiang, T. Goh, Chih-Chia Hsieh
It has been trendy to embedded machine learning techniques in enhancing decision supports in the organisations. The essential expectation for getting accurate prediction and estimation via such tool is the integrated systems and data quality – accuracy, completeness, consistency, timeliness, validity, and uniqueness. As suggested by the literature, however, hospital information systems are fragmented, and various departments implement various expert systems from different vendors due to the nature of medical complexity. Therefore, this paper proposes a conceptual framework that explains how data could be integrated from the separated systems for clinical decision support with a context of emergency department and how machine learning systems can be placed in the architecture of the completed hospital information systems.
{"title":"Gauging the Gaps for Decision Support - Data integration in the Hospital Information Systems with Machine Learning","authors":"William Yu Chung Wang, Philip Hong Wei Jiang, T. Goh, Chih-Chia Hsieh","doi":"10.1145/3545729.3545742","DOIUrl":"https://doi.org/10.1145/3545729.3545742","url":null,"abstract":"It has been trendy to embedded machine learning techniques in enhancing decision supports in the organisations. The essential expectation for getting accurate prediction and estimation via such tool is the integrated systems and data quality – accuracy, completeness, consistency, timeliness, validity, and uniqueness. As suggested by the literature, however, hospital information systems are fragmented, and various departments implement various expert systems from different vendors due to the nature of medical complexity. Therefore, this paper proposes a conceptual framework that explains how data could be integrated from the separated systems for clinical decision support with a context of emergency department and how machine learning systems can be placed in the architecture of the completed hospital information systems.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"30 7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116716462","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}
In recent years, with the application of deep learning technology in the field of medical image analysis, computer-aided medical image classification can help doctors diagnose and treat patients better. However, due to the particularity of medical images, the performance of traditional image processing is not satisfactory to all medical images. Self-supervised pretraining followed by supervised finetuning has seen success in image recognition, but has received limited attention in medical image classification. In this paper, we propose a method based on self-supervised pretraining and supervised finetuning. In the pretraining step, we train our backbone on unlabeled ImageNet and MedMNIST to learn different types of image features. In the finetuning step, we carefully compare our training method in two modalities with several mainstream methods. Our pretraining method outperforms supervised baselines pretrained on ImageNet. In addition, we show that with suitable pretraining method adopted, our proposed method could be reused on several similar tasks with little modification.
{"title":"Exploring a Universal Training Method for Medical Image Classification","authors":"Han Ding, Kun Yan, Zheyan Tu, Ping Wang","doi":"10.1145/3545729.3545731","DOIUrl":"https://doi.org/10.1145/3545729.3545731","url":null,"abstract":"In recent years, with the application of deep learning technology in the field of medical image analysis, computer-aided medical image classification can help doctors diagnose and treat patients better. However, due to the particularity of medical images, the performance of traditional image processing is not satisfactory to all medical images. Self-supervised pretraining followed by supervised finetuning has seen success in image recognition, but has received limited attention in medical image classification. In this paper, we propose a method based on self-supervised pretraining and supervised finetuning. In the pretraining step, we train our backbone on unlabeled ImageNet and MedMNIST to learn different types of image features. In the finetuning step, we carefully compare our training method in two modalities with several mainstream methods. Our pretraining method outperforms supervised baselines pretrained on ImageNet. In addition, we show that with suitable pretraining method adopted, our proposed method could be reused on several similar tasks with little modification.","PeriodicalId":432782,"journal":{"name":"Proceedings of the 6th International Conference on Medical and Health Informatics","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131719398","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：