Pub Date : 2020-12-01DOI: 10.1109/CSCI51800.2020.00334
Mounika Vanamala, Jairen Gilmore, Xiaohong Yuan, K. Roy
To develop secure software, software developers need to know the potential threats to the software. Knowledge captured in the Common Attack Pattern Enumeration and Classification (CAPEC) database can help software developers to understand how attackers target application weaknesses. In this paper, we present a method of recommending CAPEC attack patterns based on software requirement specification (SRS) documents. The method uses topic modelling to extract topics from each attack pattern and to extract topics from the software system description, user classes, use cases, and function requirements within the SRS documents. Attack patterns are recommended by calculating the distance measure of each attack pattern topic distribution and each SRS topic distribution using cosine similarity. Attack patterns are then ranked from maximum to minimum. The top attack patterns are then recommended to the software developers as the most relevant to the software system under development.
{"title":"Recommending Attack Patterns for Software Requirements Document","authors":"Mounika Vanamala, Jairen Gilmore, Xiaohong Yuan, K. Roy","doi":"10.1109/CSCI51800.2020.00334","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00334","url":null,"abstract":"To develop secure software, software developers need to know the potential threats to the software. Knowledge captured in the Common Attack Pattern Enumeration and Classification (CAPEC) database can help software developers to understand how attackers target application weaknesses. In this paper, we present a method of recommending CAPEC attack patterns based on software requirement specification (SRS) documents. The method uses topic modelling to extract topics from each attack pattern and to extract topics from the software system description, user classes, use cases, and function requirements within the SRS documents. Attack patterns are recommended by calculating the distance measure of each attack pattern topic distribution and each SRS topic distribution using cosine similarity. Attack patterns are then ranked from maximum to minimum. The top attack patterns are then recommended to the software developers as the most relevant to the software system under development.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133890418","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00028
M. Blair, Davis Jeffords, Eric Lilling, S. Banik
Due to increasing number of attacks in the cyberspace that deals with different types of users, it is imperative that an automated responder service will be efficient to help the users detect and mitigate different types of attacks in their systems. In this research, we propose to replicate the framework of emergency responder service (911) of the physical space to the cyberspace. Towards this we propose a framework for Cyber-as-a-Service for end-users. In our proposed model, we have three entities: the Dispatch Center, the Guard, and the Client Software. These entities will communicate with each other to detect and extinguish any malicious activity on the host computer. The host machine will run a software that scans and detects any abnormal or malicious activity and communicates this activity to the Guard, which then replies with an executable resolution back to the host. Meanwhile, the Dispatch Center manages connections between hosts and Guards to ensure that hosts are connected to the optimal Guard. We propose algorithms that will place and distribute the Dispatch Center and the Guards. These algorithms allow for fair distribution of Guards, as well as balance the workload among the Guards. We propose the communication protocol that will take place between the Client software, Guards and the Dispatch Center. Our goal is to design the framework for Cyber-as-a-Service for everyday users in the cyberspace who do not have sufficient technical skills to manage tools to detect different attacks.
{"title":"Cyber as a Service: Automating First Responders’ Service in the Cyberspace","authors":"M. Blair, Davis Jeffords, Eric Lilling, S. Banik","doi":"10.1109/CSCI51800.2020.00028","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00028","url":null,"abstract":"Due to increasing number of attacks in the cyberspace that deals with different types of users, it is imperative that an automated responder service will be efficient to help the users detect and mitigate different types of attacks in their systems. In this research, we propose to replicate the framework of emergency responder service (911) of the physical space to the cyberspace. Towards this we propose a framework for Cyber-as-a-Service for end-users. In our proposed model, we have three entities: the Dispatch Center, the Guard, and the Client Software. These entities will communicate with each other to detect and extinguish any malicious activity on the host computer. The host machine will run a software that scans and detects any abnormal or malicious activity and communicates this activity to the Guard, which then replies with an executable resolution back to the host. Meanwhile, the Dispatch Center manages connections between hosts and Guards to ensure that hosts are connected to the optimal Guard. We propose algorithms that will place and distribute the Dispatch Center and the Guards. These algorithms allow for fair distribution of Guards, as well as balance the workload among the Guards. We propose the communication protocol that will take place between the Client software, Guards and the Dispatch Center. Our goal is to design the framework for Cyber-as-a-Service for everyday users in the cyberspace who do not have sufficient technical skills to manage tools to detect different attacks.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"146 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133965682","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 : 2020-12-01DOI: 10.1109/csci51800.2020.00022
Guillermo A. Francia, E. El-Sheikh, H. Chi
The rapid pace with which connected and autonomous vehicles is evolving presents security challenges that are prevalent on communication technologies. Although it is universally accepted that tremendous benefits can be derived from this emerging technology, we need to make sure that this critical infrastructure is secured and protected. Recent attacks on vehicle networks have validated the urgent need for a robust and sustained effort to stem the tide of these debilitating incursions. Our ever-increasing dependence on this type of transport system brings us to new crossroads and challenges that are confronting our economic security, privacy protection, and well-being. One major challenge is the education and training of the current and future workforce in this emerging technology. This paper explores key curriculum issues in securing modem automobiles, including the essential tools necessary to implement meaningful hands-on laboratory experiments and learning scenarios.
{"title":"Vehicle Security Learning Tools and Scenarios","authors":"Guillermo A. Francia, E. El-Sheikh, H. Chi","doi":"10.1109/csci51800.2020.00022","DOIUrl":"https://doi.org/10.1109/csci51800.2020.00022","url":null,"abstract":"The rapid pace with which connected and autonomous vehicles is evolving presents security challenges that are prevalent on communication technologies. Although it is universally accepted that tremendous benefits can be derived from this emerging technology, we need to make sure that this critical infrastructure is secured and protected. Recent attacks on vehicle networks have validated the urgent need for a robust and sustained effort to stem the tide of these debilitating incursions. Our ever-increasing dependence on this type of transport system brings us to new crossroads and challenges that are confronting our economic security, privacy protection, and well-being. One major challenge is the education and training of the current and future workforce in this emerging technology. This paper explores key curriculum issues in securing modem automobiles, including the essential tools necessary to implement meaningful hands-on laboratory experiments and learning scenarios.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128925915","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00087
Zhantong Liang, A. Youssef
It takes more than correct grammar to speak good English. In this paper, we describe the sentence denoising task that reduces the vagueness, redundancy, and irrationality of a grammatical sentence. We define a rich, linguistics-inspired noise taxonomy and establish the formal definition of the problem. A unified end-to-end model based on Transformer is proposed and an efficient algorithm for constructing the training data is given, together with a separate fine-tuning step to get the ideal model. Our method outperforms previous results and keeps good accuracy as the noise composition gets more complicated.
{"title":"Unified End-to-End Sentence Denoising","authors":"Zhantong Liang, A. Youssef","doi":"10.1109/CSCI51800.2020.00087","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00087","url":null,"abstract":"It takes more than correct grammar to speak good English. In this paper, we describe the sentence denoising task that reduces the vagueness, redundancy, and irrationality of a grammatical sentence. We define a rich, linguistics-inspired noise taxonomy and establish the formal definition of the problem. A unified end-to-end model based on Transformer is proposed and an efficient algorithm for constructing the training data is given, together with a separate fine-tuning step to get the ideal model. Our method outperforms previous results and keeps good accuracy as the noise composition gets more complicated.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130929573","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00315
Jinhyung Park, Hyun-su Lim, Jun-Yeong Bok
Korea Aerospace Research Institute launched GEO-KOMPSAT-2B, the second satellite of its Geostationary Observation Satellite series, in February 2020. The GOCI-II is a payload embedded on GEO-KOMPSAT-2B continuing GOCI’s geostationary ocean monitoring for Korean Peninsular as well as Earth fulldisk. It is expected that more various marine information production with its 4-time improved performance in spatial resolution. In this paper, we introduce development of DPS to pre-process image data of GOCI-II. The GOCI-II DPS is designed to process in real-time and reliable without operator’s interference. Completing in-orbit tests, the GOCI-II DPS currently starts its nominal operations during the next 10 years of GEO-KOMPSAT-2B lifetime to provide useful ocean image data.
{"title":"Development of Image Pre-processing System for GEO-KOMPSAT-2 GOCI-II","authors":"Jinhyung Park, Hyun-su Lim, Jun-Yeong Bok","doi":"10.1109/CSCI51800.2020.00315","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00315","url":null,"abstract":"Korea Aerospace Research Institute launched GEO-KOMPSAT-2B, the second satellite of its Geostationary Observation Satellite series, in February 2020. The GOCI-II is a payload embedded on GEO-KOMPSAT-2B continuing GOCI’s geostationary ocean monitoring for Korean Peninsular as well as Earth fulldisk. It is expected that more various marine information production with its 4-time improved performance in spatial resolution. In this paper, we introduce development of DPS to pre-process image data of GOCI-II. The GOCI-II DPS is designed to process in real-time and reliable without operator’s interference. Completing in-orbit tests, the GOCI-II DPS currently starts its nominal operations during the next 10 years of GEO-KOMPSAT-2B lifetime to provide useful ocean image data.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133858495","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00260
Segundo Moisés Toapanta Toapanta, Stefania Nefali Guaranda Lara, Joseph Alexander Guamán Seis, Luis Enrique Mafla Gallegos, Jose Antonio Orizaga Trejo, Ma. Roció Maciel Arellano
Many companies have opted for the implementation of information and communication technologies (ICT) in their business processes, in order to achieve their corporate goals and obtain excellent results, but this has not always been achieved. The objective of this study was to present the prototype of a model that can align corporate strategies with ICT, in an Ecuadorian company. The quantitative approach was applied and the deductive method and exploration were used to analyze the information of articles and scientific. The result was a model that aligns ICT with corporate strategies in which a table was included with indicators of alignment between ICT and business strategy classified according to the strategic leadership style of the company for decision making, it was applied under the framework of Miles and Snow, in addition to the evaluation of the strategies, was effective using the Saaty scale and the Analytical Hierarchy Process (AHP) method applied for making objective decisions. It was concluded that the strategic alignment was achieved when the Scope of the Planning with the ICT Scope, under the Miles and Snow framework, the Analytical Hierarchy Process method was also applied, where a value of 0.08 for CR was obtained, which indicates that it is within the acceptable and consistent, transmission of cost reduction, standardization of processes, improvement in work flow and communications.
{"title":"Prototype of a Model for the Alignment of Corporate Strategies and Information and Communication Technologies","authors":"Segundo Moisés Toapanta Toapanta, Stefania Nefali Guaranda Lara, Joseph Alexander Guamán Seis, Luis Enrique Mafla Gallegos, Jose Antonio Orizaga Trejo, Ma. Roció Maciel Arellano","doi":"10.1109/CSCI51800.2020.00260","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00260","url":null,"abstract":"Many companies have opted for the implementation of information and communication technologies (ICT) in their business processes, in order to achieve their corporate goals and obtain excellent results, but this has not always been achieved. The objective of this study was to present the prototype of a model that can align corporate strategies with ICT, in an Ecuadorian company. The quantitative approach was applied and the deductive method and exploration were used to analyze the information of articles and scientific. The result was a model that aligns ICT with corporate strategies in which a table was included with indicators of alignment between ICT and business strategy classified according to the strategic leadership style of the company for decision making, it was applied under the framework of Miles and Snow, in addition to the evaluation of the strategies, was effective using the Saaty scale and the Analytical Hierarchy Process (AHP) method applied for making objective decisions. It was concluded that the strategic alignment was achieved when the Scope of the Planning with the ICT Scope, under the Miles and Snow framework, the Analytical Hierarchy Process method was also applied, where a value of 0.08 for CR was obtained, which indicates that it is within the acceptable and consistent, transmission of cost reduction, standardization of processes, improvement in work flow and communications.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114887467","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00293
Q. A. Al-Haija, M. Smadi, S. Zein-Sabatto
Severe circumstances of outdoor weather might have a significant influence on the road traffic. However, the early weather condition warning and detection can provide a significant chance for correct control and survival. Therefore, the auto-recognition models of weather situations with high level of confidence are essentially needed for several autonomous IoT systems, self-driving vehicles and transport control systems. In this work, we propose an accurate and precise self-reliant framework for weather recognition using ResNet-18 convolutional neural network to provide multi-class weather classification. The proposed model employs transfer learning technique of the powerful ResNet-18 CNN pretrained on ImageNet to train and classify weather recognition images dataset into four classes including: sunrise, shine, rain, and cloudy. The simulation results showed that our proposed model achieves remarkable classification accuracy of 98.22% outperforming other compared models trained on the same dataset.
{"title":"Multi-Class Weather Classification Using ResNet-18 CNN for Autonomous IoT and CPS Applications","authors":"Q. A. Al-Haija, M. Smadi, S. Zein-Sabatto","doi":"10.1109/CSCI51800.2020.00293","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00293","url":null,"abstract":"Severe circumstances of outdoor weather might have a significant influence on the road traffic. However, the early weather condition warning and detection can provide a significant chance for correct control and survival. Therefore, the auto-recognition models of weather situations with high level of confidence are essentially needed for several autonomous IoT systems, self-driving vehicles and transport control systems. In this work, we propose an accurate and precise self-reliant framework for weather recognition using ResNet-18 convolutional neural network to provide multi-class weather classification. The proposed model employs transfer learning technique of the powerful ResNet-18 CNN pretrained on ImageNet to train and classify weather recognition images dataset into four classes including: sunrise, shine, rain, and cloudy. The simulation results showed that our proposed model achieves remarkable classification accuracy of 98.22% outperforming other compared models trained on the same dataset.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115041469","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00156
V. Veeraiah, G. Ravikumar
An integrated healthcare system, making use of Internet of things has immense benefits. A crucial factor for an efficient delivery of the health care system is the medical database of the patient on a real time basis. The mortality rate has deep rooted constrains in medical errors due to lack of medical information of the patient and unavailability of real time data. IoT enabled technology can bring about a drastic change in reducing the mortality rate. The value added is much beyond the money it can save people’s life, with the outbreak of Covid-19 pandemic the benefits derived from IoT enabled technology has been remarkable. A striking feature of IoT enabled tools in the field of health care sector is thait can cater to a large population simultaneously and remotely. If IoT platform is implemented in healthcare services it can reduce the expenditure of health care services to the national GDP and can reduce the mortality rate. With digital transformation E-health apps are cost effective and less time consuming for individuals to monitor vitals at will. An emerging economy like India can be benefitted immensely from using IoT enabled tools in the field of health care services.
{"title":"Integrated Health Care Delivery system with IoT Enabling Technology","authors":"V. Veeraiah, G. Ravikumar","doi":"10.1109/CSCI51800.2020.00156","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00156","url":null,"abstract":"An integrated healthcare system, making use of Internet of things has immense benefits. A crucial factor for an efficient delivery of the health care system is the medical database of the patient on a real time basis. The mortality rate has deep rooted constrains in medical errors due to lack of medical information of the patient and unavailability of real time data. IoT enabled technology can bring about a drastic change in reducing the mortality rate. The value added is much beyond the money it can save people’s life, with the outbreak of Covid-19 pandemic the benefits derived from IoT enabled technology has been remarkable. A striking feature of IoT enabled tools in the field of health care sector is thait can cater to a large population simultaneously and remotely. If IoT platform is implemented in healthcare services it can reduce the expenditure of health care services to the national GDP and can reduce the mortality rate. With digital transformation E-health apps are cost effective and less time consuming for individuals to monitor vitals at will. An emerging economy like India can be benefitted immensely from using IoT enabled tools in the field of health care services.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124824217","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00259
S. Demigha
Big Data refers to data that can’t be processed with traditional applications due the challenge of capturing, storing, transferring, querying, fast processing and updating data in such large amounts. The Big Data concept often uses analytics involving Artificial Intelligence (AI), Machine Learning and Deep Learning. The paper investigates the impact of Big Data in the use of AI methods and techniques.
{"title":"The impact of Big Data on AI","authors":"S. Demigha","doi":"10.1109/CSCI51800.2020.00259","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00259","url":null,"abstract":"Big Data refers to data that can’t be processed with traditional applications due the challenge of capturing, storing, transferring, querying, fast processing and updating data in such large amounts. The Big Data concept often uses analytics involving Artificial Intelligence (AI), Machine Learning and Deep Learning. The paper investigates the impact of Big Data in the use of AI methods and techniques.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129745299","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 : 2020-12-01DOI: 10.1109/CSCI51800.2020.00144
Ulyana Morar, Harold Martin, Walter Izquierdo, Parisa Forouzannezhad, Elaheh Zarafshan, R. Curiel, M. Roselli, D. Loewenstein, R. Duara, Elona Unger, M. Adjouadi
This study introduces a new multimodal deep regression method to predict cognitive test score in a 5-year longitudinal study on Alzheimer’s disease (AD). The proposed model takes advantage of multimodal data that includes cerebrospinal fluid (CSF) levels of tau and beta-amyloid, structural measures from magnetic resonance imaging (MRI), functional and metabolic measures from positron emission tomography (PET), and cognitive scores from neuropsychological tests (Cog), all with the aim of achieving highly accurate predictions of future Mini-Mental State Examination (MMSE) test scores up to five years after baseline biomarker collection. A novel data augmentation technique is leveraged to increase the numbers of training samples without relying on synthetic data. With the proposed method, the best and most encompassing regressor is shown to achieve better than state-of-the-art correlations of 85.07%(SD=1.59) for 6 months in the future, 87.39% (SD =1.48) for 12 months, 84.78% (SD=2.66) for 18 months, 85.13% (SD=2.19) for 24 months, 81.15% (SD=5.48) for 30 months, 81.17% (SD=4.44) for 36 months, 79.25% (SD=5.85) for 42 months, 78.98% (SD=5.79) for 48 months, 78.93%(SD=5.76) for 54 months, and 74.96% (SD=7.54) for 60 months.
{"title":"A Deep-Learning Approach for the Prediction of Mini-Mental State Examination Scores in a Multimodal Longitudinal Study","authors":"Ulyana Morar, Harold Martin, Walter Izquierdo, Parisa Forouzannezhad, Elaheh Zarafshan, R. Curiel, M. Roselli, D. Loewenstein, R. Duara, Elona Unger, M. Adjouadi","doi":"10.1109/CSCI51800.2020.00144","DOIUrl":"https://doi.org/10.1109/CSCI51800.2020.00144","url":null,"abstract":"This study introduces a new multimodal deep regression method to predict cognitive test score in a 5-year longitudinal study on Alzheimer’s disease (AD). The proposed model takes advantage of multimodal data that includes cerebrospinal fluid (CSF) levels of tau and beta-amyloid, structural measures from magnetic resonance imaging (MRI), functional and metabolic measures from positron emission tomography (PET), and cognitive scores from neuropsychological tests (Cog), all with the aim of achieving highly accurate predictions of future Mini-Mental State Examination (MMSE) test scores up to five years after baseline biomarker collection. A novel data augmentation technique is leveraged to increase the numbers of training samples without relying on synthetic data. With the proposed method, the best and most encompassing regressor is shown to achieve better than state-of-the-art correlations of 85.07%(SD=1.59) for 6 months in the future, 87.39% (SD =1.48) for 12 months, 84.78% (SD=2.66) for 18 months, 85.13% (SD=2.19) for 24 months, 81.15% (SD=5.48) for 30 months, 81.17% (SD=4.44) for 36 months, 79.25% (SD=5.85) for 42 months, 78.98% (SD=5.79) for 48 months, 78.93%(SD=5.76) for 54 months, and 74.96% (SD=7.54) for 60 months.","PeriodicalId":336929,"journal":{"name":"2020 International Conference on Computational Science and Computational Intelligence (CSCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128735282","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：