Pub Date : 2021-09-01DOI: 10.1109/icdh52753.2021.00012
H. Jamil
Personal health libraries (PHL) are increasingly becoming the mainstay as a single point for patient centered health information management and services. However, the transition to a solely PHL based health information management (HIM) will, at the very least, take a very long time. It is more likely therefore to co-evolve with our current systems for HIMs. In this emerging scenario, the traditional obstacles of data integration among autonomous HIMs face novel challenges. Additionally, the goal to make PHLs responsive to open-ended and personalized health information needs adds unknown wrinkles to current challenges. In this paper, we propose a new architecture, and a knowledge-based information retrieval and processing model for PHLs. We show that by using a declarative data integration language, a knowledge representation scheme and knowledge graph induction technique from health information texts, we are able to respond to patient queries in unprecedented ways in the context of their PHLs.
{"title":"Architecture of an Intelligent Personal Health Library for Improved Health Outcomes","authors":"H. Jamil","doi":"10.1109/icdh52753.2021.00012","DOIUrl":"https://doi.org/10.1109/icdh52753.2021.00012","url":null,"abstract":"Personal health libraries (PHL) are increasingly becoming the mainstay as a single point for patient centered health information management and services. However, the transition to a solely PHL based health information management (HIM) will, at the very least, take a very long time. It is more likely therefore to co-evolve with our current systems for HIMs. In this emerging scenario, the traditional obstacles of data integration among autonomous HIMs face novel challenges. Additionally, the goal to make PHLs responsive to open-ended and personalized health information needs adds unknown wrinkles to current challenges. In this paper, we propose a new architecture, and a knowledge-based information retrieval and processing model for PHLs. We show that by using a declarative data integration language, a knowledge representation scheme and knowledge graph induction technique from health information texts, we are able to respond to patient queries in unprecedented ways in the context of their PHLs.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"167 1","pages":"1-10"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73809484","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 : 2021-09-01DOI: 10.1007/s40747-022-00775-w
V. Bellandi, P. Ceravolo, E. Damiani, S. Maghool, M. Cesari, Ioannis Basdekis, Eleftheria Iliadou, Mircea Mărzan
{"title":"Engineering Continuous Monitoring of Intrinsic Capacity for Elderly People","authors":"V. Bellandi, P. Ceravolo, E. Damiani, S. Maghool, M. Cesari, Ioannis Basdekis, Eleftheria Iliadou, Mircea Mărzan","doi":"10.1007/s40747-022-00775-w","DOIUrl":"https://doi.org/10.1007/s40747-022-00775-w","url":null,"abstract":"","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"36 1","pages":"166-171"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75166188","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 : 2021-09-01DOI: 10.1109/icdh52753.2021.00023
Shreesha Narasimha Murthy, E. Agu
The world has been severely affected by COVID-19, an infectious disease caused by the SARS-Cov-2 coronavirus. COVID-19 incubates in a patient for 7 days before symptoms manifest. The identification of the presence of COVID-19 is challenging as its symptoms are similar to influenza symptoms such as cough, cold, runny nose, and chills. COVID-19 affects human speech sub-systems involved in respiration, phonation, and articulation. We propose a deep anomaly detection framework for passive, speech-based detection of COVID-related anomalies in voice samples of COVID-19 affected individuals. The low percentage of positive cases and extreme imbalance in available COVID audio datasets present a challenge to machine learning classifiers but create an opportunity to utilize anomaly detection techniques. We investigate COVID detection from audio using various types of deep anomaly detectors and convolutional autoencoders. Contrastive loss methods are also explored to force our models to learn discrepancies between COVID and non-COVID cough data representations. In contrast with prior work that controlled data collection, our work focuses on crowdsourced datasets that are true representatives of the general population. In rigorous evaluation, the variational autoencoder with the elliptic envelope as the anomaly detector analyzing Mel Filterbanks audio representations performed best with an AUC of 0.65, outperforming the state of the art.
{"title":"Deep Learning Anomaly Detection methods to passively detect COVID-19 from Audio","authors":"Shreesha Narasimha Murthy, E. Agu","doi":"10.1109/icdh52753.2021.00023","DOIUrl":"https://doi.org/10.1109/icdh52753.2021.00023","url":null,"abstract":"The world has been severely affected by COVID-19, an infectious disease caused by the SARS-Cov-2 coronavirus. COVID-19 incubates in a patient for 7 days before symptoms manifest. The identification of the presence of COVID-19 is challenging as its symptoms are similar to influenza symptoms such as cough, cold, runny nose, and chills. COVID-19 affects human speech sub-systems involved in respiration, phonation, and articulation. We propose a deep anomaly detection framework for passive, speech-based detection of COVID-related anomalies in voice samples of COVID-19 affected individuals. The low percentage of positive cases and extreme imbalance in available COVID audio datasets present a challenge to machine learning classifiers but create an opportunity to utilize anomaly detection techniques. We investigate COVID detection from audio using various types of deep anomaly detectors and convolutional autoencoders. Contrastive loss methods are also explored to force our models to learn discrepancies between COVID and non-COVID cough data representations. In contrast with prior work that controlled data collection, our work focuses on crowdsourced datasets that are true representatives of the general population. In rigorous evaluation, the variational autoencoder with the elliptic envelope as the anomaly detector analyzing Mel Filterbanks audio representations performed best with an AUC of 0.65, outperforming the state of the art.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"152 1","pages":"114-121"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75840722","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 : 2021-09-01DOI: 10.1109/ICDH52753.2021.00052
L. Migiro, H. Shahriar, S. Sneha
The spread of COVD-19 has affected normal life like no other pandemic in the 21 st century. This has seen the evolution and adoption of digital contact tracing applications, majority of which rely on google and apple exposure notification and can easily be downloaded for use in any smartphone. It is imperative to protect personal health information transmitted in these apps. Developers have been criticized for slacking in protecting personal health information and on being non-compliant to HIPAA. Using MobSF, we interact with these apps to detect security vulnerabilities and demonstrate whether they are complying with their privacy policies. Our analysis showed that contact tracing applications have poor security features and not safe.
{"title":"Analyzing Security and Privacy Concerns of Contact Tracing Applications","authors":"L. Migiro, H. Shahriar, S. Sneha","doi":"10.1109/ICDH52753.2021.00052","DOIUrl":"https://doi.org/10.1109/ICDH52753.2021.00052","url":null,"abstract":"The spread of COVD-19 has affected normal life like no other pandemic in the 21 st century. This has seen the evolution and adoption of digital contact tracing applications, majority of which rely on google and apple exposure notification and can easily be downloaded for use in any smartphone. It is imperative to protect personal health information transmitted in these apps. Developers have been criticized for slacking in protecting personal health information and on being non-compliant to HIPAA. Using MobSF, we interact with these apps to detect security vulnerabilities and demonstrate whether they are complying with their privacy policies. Our analysis showed that contact tracing applications have poor security features and not safe.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"67 1","pages":"283-292"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77465582","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 : 2021-09-01DOI: 10.1109/icdh52753.2021.00017
Nur Imtiazul Haque, Alvi Ataur Khalil, M. Rahman, M. Amini, Sheikh Iqbal Ahamed
The modern smart digital healthcare system (SDHS) is leaning towards automation of patient disease monitoring and treatment with the advent of wireless body sensor networks (WBSN) and the internet of medical things (IoMT). However, the open communication network for sensitive medical data transfer is giving rise to vulnerabilities and security concerns. To prevent adversarial manipulation of sensor measurements, SDHS IoMT controllers leverage anomaly detection systems on top of the disease classification systems. Machine learning (ML) is one of the most effective techniques for providing experience-based automated decision-making models. These models generalize well to produce the expected output for the unseen inputs from the learned patterns. Therefore, ML-based models are currently being adopted to automate the anomaly detection and disease classification tasks of SDHS. In this work, we consider a SDHS that uses supervised ML models for patient status/disease classification and unsupervised ML models for anomaly detection. However, the performance of the ML models largely depends on hyper-parameter tuning. Finding the optimal hyper-parameter is a challenging task, and it becomes more difficult and time-consuming in high-dimensional feature space. In this work, we propose BIOCAD, a comprehensive bio-inspired optimization framework for SDHS data classification and anomaly detection. The framework leverages a novel fitness function for unsu-pervised anomaly detection ML models. We experiment with state-of-the-art datasets - the Pima Indians diabetes dataset, the Parkinson dataset, and the University of Queensland vital signs (UQVS) dataset for validating our proposed strategy.
{"title":"BIOCAD: Bio-Inspired Optimization for Classification and Anomaly Detection in Digital Healthcare Systems","authors":"Nur Imtiazul Haque, Alvi Ataur Khalil, M. Rahman, M. Amini, Sheikh Iqbal Ahamed","doi":"10.1109/icdh52753.2021.00017","DOIUrl":"https://doi.org/10.1109/icdh52753.2021.00017","url":null,"abstract":"The modern smart digital healthcare system (SDHS) is leaning towards automation of patient disease monitoring and treatment with the advent of wireless body sensor networks (WBSN) and the internet of medical things (IoMT). However, the open communication network for sensitive medical data transfer is giving rise to vulnerabilities and security concerns. To prevent adversarial manipulation of sensor measurements, SDHS IoMT controllers leverage anomaly detection systems on top of the disease classification systems. Machine learning (ML) is one of the most effective techniques for providing experience-based automated decision-making models. These models generalize well to produce the expected output for the unseen inputs from the learned patterns. Therefore, ML-based models are currently being adopted to automate the anomaly detection and disease classification tasks of SDHS. In this work, we consider a SDHS that uses supervised ML models for patient status/disease classification and unsupervised ML models for anomaly detection. However, the performance of the ML models largely depends on hyper-parameter tuning. Finding the optimal hyper-parameter is a challenging task, and it becomes more difficult and time-consuming in high-dimensional feature space. In this work, we propose BIOCAD, a comprehensive bio-inspired optimization framework for SDHS data classification and anomaly detection. The framework leverages a novel fitness function for unsu-pervised anomaly detection ML models. We experiment with state-of-the-art datasets - the Pima Indians diabetes dataset, the Parkinson dataset, and the University of Queensland vital signs (UQVS) dataset for validating our proposed strategy.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"1 1","pages":"48-58"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76049055","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 : 2021-09-01DOI: 10.1109/icdh52753.2021.00020
Atifa Sarwar, E. Agu
COVID-19 has now infected over 165 million people and killed over 3.5 million people. While public health interventions have reduced its spread and vaccines are being deployed, passive detection methods are needed to detect infections and early track its resurgence. Wearables that are widely owned can gather various physiological and activity data, presenting an opportunity to detect COVID-19 unobtrusively. COVID-19 infection causes deviations in the vital physiological signs and activity patterns of infected users. However, similar deviations of these same variables can also be affected by non-COVID factors, confounding the signals. In this paper, we investigate the feasibility of predicting COVID-19 infection to detect abnormalities in heart rate, activity (steps), and sleep data available on low-end wearables by using machine learning. Prior work utilized data such as oxygen saturation that is only available on clinical-grade equipment or expensive wearables. We extracted 43 statistical features (standard deviation, mean, slope) and behavioral (min/max/avg length of sedentary and active bouts, sleep duration, no. of awake/asleep/restless samples) from wearable sensor data. We classified these features using machine learning classification and anomaly detection algorithms. Physical activity features were the most predictive (min length of the sedentary and active bout), yielding an AUC-ROC of 78% [specificity=74%, sensitivity=69%] when classified using Gradient Boosting Machines (GBMs). We also found that sleep irregularities had low discriminative performance. COVID-19 detection using inexpensive wearables can facilitate population-level interventions.
{"title":"Passive COVID-19 Assessment using Machine Learning on Physiological and Activity Data from Low End Wearables","authors":"Atifa Sarwar, E. Agu","doi":"10.1109/icdh52753.2021.00020","DOIUrl":"https://doi.org/10.1109/icdh52753.2021.00020","url":null,"abstract":"COVID-19 has now infected over 165 million people and killed over 3.5 million people. While public health interventions have reduced its spread and vaccines are being deployed, passive detection methods are needed to detect infections and early track its resurgence. Wearables that are widely owned can gather various physiological and activity data, presenting an opportunity to detect COVID-19 unobtrusively. COVID-19 infection causes deviations in the vital physiological signs and activity patterns of infected users. However, similar deviations of these same variables can also be affected by non-COVID factors, confounding the signals. In this paper, we investigate the feasibility of predicting COVID-19 infection to detect abnormalities in heart rate, activity (steps), and sleep data available on low-end wearables by using machine learning. Prior work utilized data such as oxygen saturation that is only available on clinical-grade equipment or expensive wearables. We extracted 43 statistical features (standard deviation, mean, slope) and behavioral (min/max/avg length of sedentary and active bouts, sleep duration, no. of awake/asleep/restless samples) from wearable sensor data. We classified these features using machine learning classification and anomaly detection algorithms. Physical activity features were the most predictive (min length of the sedentary and active bout), yielding an AUC-ROC of 78% [specificity=74%, sensitivity=69%] when classified using Gradient Boosting Machines (GBMs). We also found that sleep irregularities had low discriminative performance. COVID-19 detection using inexpensive wearables can facilitate population-level interventions.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"26 1","pages":"80-90"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82871389","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 : 2021-09-01DOI: 10.1109/ICDH52753.2021.00047
Zhanqiang Cao, Lin Liu, Jianmin Wang
For the continuous improvement of healthcare quality and efficiency, much research efforts have been spent on adding intelligent modules in hospital information systems to ensure timely access to health data, monitor the quality of health care services, continuously improve service outcomes. This paper reports our experience in analyzing and implementing intelligent health data services based on actor model. The key lessons are 1) To adapt to the continuously changing health information needs, a system instrumented with actor-based conceptual model is a natural fit; 2) A dedicated domain actor-based data model provides the foundation for streamlining the stakeholders' needs and effective management of information systems assets of a hospital, and the key is bridging the gaps between the different levels of abstraction and the multi-perspectives of actors; 3) actors with learning ability can help the continuous observation and optimization of system-level qualities. Regardless of the discrepancies in medical processes and IT maturity, the evolution of hospital information systems to intelligent and learning organizations based on actor model can help provide clinical data collection services on demand, account for different types of medical events, and build a closed-loop management process.
{"title":"Intelligent Health Information Services Requirements Revisited from an Actor's Perspective","authors":"Zhanqiang Cao, Lin Liu, Jianmin Wang","doi":"10.1109/ICDH52753.2021.00047","DOIUrl":"https://doi.org/10.1109/ICDH52753.2021.00047","url":null,"abstract":"For the continuous improvement of healthcare quality and efficiency, much research efforts have been spent on adding intelligent modules in hospital information systems to ensure timely access to health data, monitor the quality of health care services, continuously improve service outcomes. This paper reports our experience in analyzing and implementing intelligent health data services based on actor model. The key lessons are 1) To adapt to the continuously changing health information needs, a system instrumented with actor-based conceptual model is a natural fit; 2) A dedicated domain actor-based data model provides the foundation for streamlining the stakeholders' needs and effective management of information systems assets of a hospital, and the key is bridging the gaps between the different levels of abstraction and the multi-perspectives of actors; 3) actors with learning ability can help the continuous observation and optimization of system-level qualities. Regardless of the discrepancies in medical processes and IT maturity, the evolution of hospital information systems to intelligent and learning organizations based on actor model can help provide clinical data collection services on demand, account for different types of medical events, and build a closed-loop management process.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"2 1","pages":"244-253"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89553076","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 : 2021-09-01DOI: 10.1109/icdh52753.2021.00045
Md. Saiful Islam, Shahriar Sobhan, Maria Valero, H. Shahriar, Liang Zhao, S. Ahamed
The Internet of Things (IoT) is the most significant and blooming technology in the 21st century while rapidly developed by covering hundreds of applications in the civil, health, military, and agriculture areas. IoT is based on the collection of sensor data through an embedded system, and this embedded system uploads the data on the internet. Devices and sensor technologies connected over a network can monitor and measure data in real-time. The main challenge is to collect data from IoT devices, transmit them to store in the Cloud, and later retrieve them at any time for visualization and data analysis. All these phases need to be secure by following security protocol to ensure data integrity. In this paper, we present the design of a lightweight and easy-to-use data collection framework for IoT devices, that can potentially be applied to sensors that monitor healthcare. This framework consists of collecting data from sensors and sending them to Cloud storage securely and in realtime for further processing and visualization. Our main objective is to make a data-collecting platform that will be plug-and-play and secure so that any healthcare organization or research team can use it to collect data from any IoT device for further data analysis.
物联网(Internet of Things, IoT)是21世纪最重要、最蓬勃发展的技术，在民用、卫生、军事和农业等领域得到了广泛的应用。物联网是基于通过嵌入式系统收集传感器数据，该嵌入式系统将数据上传到互联网上。通过网络连接的设备和传感器技术可以实时监控和测量数据。主要的挑战是从物联网设备收集数据，将其传输到云中存储，然后随时检索它们以进行可视化和数据分析。所有这些阶段都需要遵循安全协议来确保数据的完整性。在本文中，我们为物联网设备设计了一个轻量级且易于使用的数据收集框架，该框架可以潜在地应用于监控医疗保健的传感器。该框架包括从传感器收集数据并将其安全实时地发送到云存储，以便进一步处理和可视化。我们的主要目标是制作一个即插即用且安全的数据收集平台，以便任何医疗机构或研究团队都可以使用它从任何物联网设备收集数据以进行进一步的数据分析。
{"title":"Framework for Collecting Data from specialized IoT devices - An application to enhance Healthcare Systems","authors":"Md. Saiful Islam, Shahriar Sobhan, Maria Valero, H. Shahriar, Liang Zhao, S. Ahamed","doi":"10.1109/icdh52753.2021.00045","DOIUrl":"https://doi.org/10.1109/icdh52753.2021.00045","url":null,"abstract":"The Internet of Things (IoT) is the most significant and blooming technology in the 21st century while rapidly developed by covering hundreds of applications in the civil, health, military, and agriculture areas. IoT is based on the collection of sensor data through an embedded system, and this embedded system uploads the data on the internet. Devices and sensor technologies connected over a network can monitor and measure data in real-time. The main challenge is to collect data from IoT devices, transmit them to store in the Cloud, and later retrieve them at any time for visualization and data analysis. All these phases need to be secure by following security protocol to ensure data integrity. In this paper, we present the design of a lightweight and easy-to-use data collection framework for IoT devices, that can potentially be applied to sensors that monitor healthcare. This framework consists of collecting data from sensors and sending them to Cloud storage securely and in realtime for further processing and visualization. Our main objective is to make a data-collecting platform that will be plug-and-play and secure so that any healthcare organization or research team can use it to collect data from any IoT device for further data analysis.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"31 1","pages":"231-233"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76827991","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 : 2021-09-01DOI: 10.1109/ICDH52753.2021.00036
M. M. Baldi, Petar Aleksandrov Mavrodiev, B. Galuzzi, F. Mantovani, O. Realdon, E. Messina
Telerehabilitation is a growing area of research and clinical practice which attempts to mitigate some of the major problems in chronic disease rehabilitation programs: short-staffed clinical care teams, great demand for complex face-to-face treatments, and lack of tools for reaching, monitoring and aiding target clinical populations. Telerehabilitation attempts to solve this through the use of easily accessible digital tools such as mobile and web-based applications which often rely on some form of data collection and analysis. Empirically tested perspectives on the integration of those data-driven tools in the real-time decision-making process of clinical care practitioners are still lacking. In this paper, we present a Decision Support System prototype, designed in the context of an applied game as a part of a comprehensive telerehabilitation software system, with the purpose of supporting real-time dynamic data visualization, understanding of patient gameplay and care routine patterns and, ultimately, enhancing the clinical care and design teams' decision- making processes.
{"title":"A Decision Support System in the Context of an Applied Game for Telerehabilitation","authors":"M. M. Baldi, Petar Aleksandrov Mavrodiev, B. Galuzzi, F. Mantovani, O. Realdon, E. Messina","doi":"10.1109/ICDH52753.2021.00036","DOIUrl":"https://doi.org/10.1109/ICDH52753.2021.00036","url":null,"abstract":"Telerehabilitation is a growing area of research and clinical practice which attempts to mitigate some of the major problems in chronic disease rehabilitation programs: short-staffed clinical care teams, great demand for complex face-to-face treatments, and lack of tools for reaching, monitoring and aiding target clinical populations. Telerehabilitation attempts to solve this through the use of easily accessible digital tools such as mobile and web-based applications which often rely on some form of data collection and analysis. Empirically tested perspectives on the integration of those data-driven tools in the real-time decision-making process of clinical care practitioners are still lacking. In this paper, we present a Decision Support System prototype, designed in the context of an applied game as a part of a comprehensive telerehabilitation software system, with the purpose of supporting real-time dynamic data visualization, understanding of patient gameplay and care routine patterns and, ultimately, enhancing the clinical care and design teams' decision- making processes.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"103 1","pages":"203-208"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81311698","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 : 2021-09-01DOI: 10.1109/icdh52753.2021.00032
Haochen Jiang, Ziqi Wei, Jun Chen
In the industry of medical intelligence, classification is one of the most common tasks. It appears in various medical jobs, such as triage, diagnosis, and pathologic analysis. Many classification algorithms studied in machine learning can be chosen to help solve these tasks. However, due to the special nature of the medical industry, its data sets show a character of imbalance. Namely, the data are skewed distributed in different classes. Unfortunately, the classification problem of imbalanced data has a reputation of classic and hard-to-solve in data mining and artificial intelligence research community. What's worse, most proposed classification methods are designed to deal with binary classification case, while the common scenario in medical intelligence applications is multi-classification. To deal with this, a pre-processing structure called Cost-Sensitive Variable Neighbour Search (CSVNS) is proposed in this paper. It combines the ideas of sampling and cost-sensitive, which are two most commonly used strategies for multi-class imbalanced data classification tasks. As for the sampling process, a double-stack Variable Neighbour Search (VNS) structure is introduced and 15 different neighborhood structures are designed to help optimizing the process. Also, the classes are allocated different weights to improve the classifier's classification capacity. In the experiment part, the proposed method is evaluated on 4 medical data sets. $G$ - mean and mAUC are selected to represent the method's performance in medical classification tasks. Experimental results show the proposed method outperforms the classic methods in most situations. In the end, 3 extra data sets are tested to demonstrate the algorithms' scalability.
{"title":"A Novel Pre-processing Method for Classification Problems in Medical Intelligent Tasks","authors":"Haochen Jiang, Ziqi Wei, Jun Chen","doi":"10.1109/icdh52753.2021.00032","DOIUrl":"https://doi.org/10.1109/icdh52753.2021.00032","url":null,"abstract":"In the industry of medical intelligence, classification is one of the most common tasks. It appears in various medical jobs, such as triage, diagnosis, and pathologic analysis. Many classification algorithms studied in machine learning can be chosen to help solve these tasks. However, due to the special nature of the medical industry, its data sets show a character of imbalance. Namely, the data are skewed distributed in different classes. Unfortunately, the classification problem of imbalanced data has a reputation of classic and hard-to-solve in data mining and artificial intelligence research community. What's worse, most proposed classification methods are designed to deal with binary classification case, while the common scenario in medical intelligence applications is multi-classification. To deal with this, a pre-processing structure called Cost-Sensitive Variable Neighbour Search (CSVNS) is proposed in this paper. It combines the ideas of sampling and cost-sensitive, which are two most commonly used strategies for multi-class imbalanced data classification tasks. As for the sampling process, a double-stack Variable Neighbour Search (VNS) structure is introduced and 15 different neighborhood structures are designed to help optimizing the process. Also, the classes are allocated different weights to improve the classifier's classification capacity. In the experiment part, the proposed method is evaluated on 4 medical data sets. $G$ - mean and mAUC are selected to represent the method's performance in medical classification tasks. Experimental results show the proposed method outperforms the classic methods in most situations. In the end, 3 extra data sets are tested to demonstrate the algorithms' scalability.","PeriodicalId":93401,"journal":{"name":"2021 IEEE International Conference on Digital Health (ICDH)","volume":"1 1","pages":"178-183"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90202190","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：