Pub Date : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668397
M. Siket, G. Eigner, L. Kovács, Imre J. Rudas
In the artificial pancreas (AP) concept physiological models prove to play an important role. The high complexity of the physiological processes, intrapatient variability, data scarcity and other factors (physical activity, stress, quality of meal) pose difficulties to achieve an accurate, and virtual representation of the patient. In this pilot study, experiment-and patient-related considerations were taken into account during the identification to restrict the parameter space in a realistic way. Effects of the variabilites have been investigated on the prediction accuracy by using sensitivity analysis.
{"title":"Analysis of the Effect of Variability on the Blood Glucose Prediction Accuracy*","authors":"M. Siket, G. Eigner, L. Kovács, Imre J. Rudas","doi":"10.1109/CINTI53070.2021.9668397","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668397","url":null,"abstract":"In the artificial pancreas (AP) concept physiological models prove to play an important role. The high complexity of the physiological processes, intrapatient variability, data scarcity and other factors (physical activity, stress, quality of meal) pose difficulties to achieve an accurate, and virtual representation of the patient. In this pilot study, experiment-and patient-related considerations were taken into account during the identification to restrict the parameter space in a realistic way. Effects of the variabilites have been investigated on the prediction accuracy by using sensitivity analysis.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124509684","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 : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668496
Ákos Holló-Szabó, I. Albert, J. Botzheim
Genetic algorithms are modular metaheuristics simulating the evolutionary process over a solution set. The optimization is very adaptive but slow, making statistical research difficult. In this paper an algorithm is proposed where different variants are racing against each other while statistics are gathered. Our results show that this algorithm is an efficient, standalone, and even more adaptive solution. Those variants that result in faster convergence lead the race, but get stuck in local minima. In these cases, the more agile combinations with slower convergence gain higher probability and find better solutions farther from the local minimum. The hybrid is capable of faster convergence with minimal additional runtime. We also provide complexity estimations for resource requirements.
{"title":"Statistical Racing Crossover Based Genetic Algorithm for Vehicle Routing Problem","authors":"Ákos Holló-Szabó, I. Albert, J. Botzheim","doi":"10.1109/CINTI53070.2021.9668496","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668496","url":null,"abstract":"Genetic algorithms are modular metaheuristics simulating the evolutionary process over a solution set. The optimization is very adaptive but slow, making statistical research difficult. In this paper an algorithm is proposed where different variants are racing against each other while statistics are gathered. Our results show that this algorithm is an efficient, standalone, and even more adaptive solution. Those variants that result in faster convergence lead the race, but get stuck in local minima. In these cases, the more agile combinations with slower convergence gain higher probability and find better solutions farther from the local minimum. The hybrid is capable of faster convergence with minimal additional runtime. We also provide complexity estimations for resource requirements.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128922502","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 : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668436
Ilya Makarov, Ivan Guschenko-Cheverda
Accurate depth estimation from images is a fundamental task in deep learning. It has many applications including scene understanding and reconstruction. Datasets for supervised depth estimation are hard to obtain and usually do not contain a sufficient number of images or a sufficient variety of scenes. Since inputs for depth estimation are simple RGB images, it is easy to obtain a large number of various unlabeled images. We consider that depth masks can be labeled by using manual marking. Thus, we researched the possibility of performing an active learning approach for selecting unlabeled samples to be labeled. In this work, we concentrated on using the learning loss method to perform active learning train selection. We performed multiple experiments with the learning loss algorithm and evaluated the resulting model.
{"title":"Learning Loss for Active Learning in Depth Reconstruction Problem","authors":"Ilya Makarov, Ivan Guschenko-Cheverda","doi":"10.1109/CINTI53070.2021.9668436","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668436","url":null,"abstract":"Accurate depth estimation from images is a fundamental task in deep learning. It has many applications including scene understanding and reconstruction. Datasets for supervised depth estimation are hard to obtain and usually do not contain a sufficient number of images or a sufficient variety of scenes. Since inputs for depth estimation are simple RGB images, it is easy to obtain a large number of various unlabeled images. We consider that depth masks can be labeled by using manual marking. Thus, we researched the possibility of performing an active learning approach for selecting unlabeled samples to be labeled. In this work, we concentrated on using the learning loss method to perform active learning train selection. We performed multiple experiments with the learning loss algorithm and evaluated the resulting model.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"208 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128563770","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 : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668479
Anton Zakharenkov, Ilya Makarov
VizDoom is a flexible and easy-to-use 3D reinforcement learning research platform based on the well-known Doom first-person shooter. The challenge is to create bots that compete in the DeathMatch track, making decisions based solely on visual in-formation from the screen. The paper offers a com-parison of different approaches with reinforcement learning: Q-learning and policy-gradient algorithms. We explore the distributed learning paradigm in re-inforcement learning, and also discuss the differences in speed and quality of convergence when adding an object detection module.
{"title":"Deep Reinforcement Learning with DQN vs. PPO in VizDoom","authors":"Anton Zakharenkov, Ilya Makarov","doi":"10.1109/CINTI53070.2021.9668479","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668479","url":null,"abstract":"VizDoom is a flexible and easy-to-use 3D reinforcement learning research platform based on the well-known Doom first-person shooter. The challenge is to create bots that compete in the DeathMatch track, making decisions based solely on visual in-formation from the screen. The paper offers a com-parison of different approaches with reinforcement learning: Q-learning and policy-gradient algorithms. We explore the distributed learning paradigm in re-inforcement learning, and also discuss the differences in speed and quality of convergence when adding an object detection module.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129055973","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 : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668468
Ilya Makarov, Artem Oborevich
Graph visualization is an effective and efficient way to discover complex inter-connections between elements within the nested structure of data. To accomplish this type of representation machine learning algorithms use a technique called graph embedding and node embedding in particular. However, in this paper, we will compare well-known techniques to yet largely under-explored setting of graph embedding named community embedding: embedding individual communities instead of individual nodes. This type of embedding can be especially useful in graph visualization and community detection tasks. Despite the fact that graph embedding and clustering tasks are separate, a good solution to the first one tends to have a correlation with the solution of the second problem and may have a positive impact if knowledge is transferred.
{"title":"Network Embedding for Cluster Analysis","authors":"Ilya Makarov, Artem Oborevich","doi":"10.1109/CINTI53070.2021.9668468","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668468","url":null,"abstract":"Graph visualization is an effective and efficient way to discover complex inter-connections between elements within the nested structure of data. To accomplish this type of representation machine learning algorithms use a technique called graph embedding and node embedding in particular. However, in this paper, we will compare well-known techniques to yet largely under-explored setting of graph embedding named community embedding: embedding individual communities instead of individual nodes. This type of embedding can be especially useful in graph visualization and community detection tasks. Despite the fact that graph embedding and clustering tasks are separate, a good solution to the first one tends to have a correlation with the solution of the second problem and may have a positive impact if knowledge is transferred.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127717253","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 : 2021-11-18DOI: 10.1109/cinti53070.2021.9668472
{"title":"Copyriht Page","authors":"","doi":"10.1109/cinti53070.2021.9668472","DOIUrl":"https://doi.org/10.1109/cinti53070.2021.9668472","url":null,"abstract":"","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124126538","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 : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668590
Bence Takács, T. Haidegger
Medical robotics has become a major, rapidly expanding sector within medical devices. The development of medical/surgical robot systems is a diversified field, emerging at the cross-section of the clinical development and the machinery domain. Consequently, the core components of surgical robots can be clustered into two categories, custom-developed devices and commercially available components. Since the certification and clearance process of a medical technology is overwhelmingly complicated, there is a widespread trend to rely more on off-the-self parts, even for the main element of a system, the robot manipulator itself. Research and development can significantly speed up by integrating a robot that has the necessary certifications. Nevertheless, together with the additional components, the system shall still be certified as a new complete setup. Previously, it was not possible to obtain a robot manipulator certified for the surgical environment as a component. Companies that wanted to bring forward robot-assisted surgery spent millions of dollars just developing a new robot arm. As a result, many promising schemes did not come to market or at such high prices that they were not able to reach a wide penetration. This article introduces the state-of-the-art in component-based medical robot development, focusing on the only commercially available, certified, versatile collaborative robotic arm, the KUKA LBR med.
{"title":"Fasttracking Technology Transfer in Medical Robotics","authors":"Bence Takács, T. Haidegger","doi":"10.1109/CINTI53070.2021.9668590","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668590","url":null,"abstract":"Medical robotics has become a major, rapidly expanding sector within medical devices. The development of medical/surgical robot systems is a diversified field, emerging at the cross-section of the clinical development and the machinery domain. Consequently, the core components of surgical robots can be clustered into two categories, custom-developed devices and commercially available components. Since the certification and clearance process of a medical technology is overwhelmingly complicated, there is a widespread trend to rely more on off-the-self parts, even for the main element of a system, the robot manipulator itself. Research and development can significantly speed up by integrating a robot that has the necessary certifications. Nevertheless, together with the additional components, the system shall still be certified as a new complete setup. Previously, it was not possible to obtain a robot manipulator certified for the surgical environment as a component. Companies that wanted to bring forward robot-assisted surgery spent millions of dollars just developing a new robot arm. As a result, many promising schemes did not come to market or at such high prices that they were not able to reach a wide penetration. This article introduces the state-of-the-art in component-based medical robot development, focusing on the only commercially available, certified, versatile collaborative robotic arm, the KUKA LBR med.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131050667","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 : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668519
Bence Takács, Kristóf Takács, Tivadar Garamvölgyi, T. Haidegger
Soft tissue interaction and grasping is a widely researched field, nevertheless, autonomous robotics is a relatively new domain in delicate meat processing. The inner organs of animals are complex soft tissues with fuzzy boundaries and slippery surfaces, yet their precise manipulation might still be required for certain robotic processes. This paper presents a gripper development for pig inner organ gripping and manipulation. The customized mechanical design and the force measuring feature allow safe grasping, holding, stretching and moving of the slippery and easily torn tissues. The paper describes how a sensor-enabled, smart version of the gripper was engineered. The capabilities of the tool were primarily tested through laboratory dry tests and on pig-carcasses in a local slaughterhouse. Advanced features for in-device force, position and slip sensing are being developed for future use.
{"title":"Inner Organ Manipulation During Automated Pig Slaughtering—Smart Gripping Approaches","authors":"Bence Takács, Kristóf Takács, Tivadar Garamvölgyi, T. Haidegger","doi":"10.1109/CINTI53070.2021.9668519","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668519","url":null,"abstract":"Soft tissue interaction and grasping is a widely researched field, nevertheless, autonomous robotics is a relatively new domain in delicate meat processing. The inner organs of animals are complex soft tissues with fuzzy boundaries and slippery surfaces, yet their precise manipulation might still be required for certain robotic processes. This paper presents a gripper development for pig inner organ gripping and manipulation. The customized mechanical design and the force measuring feature allow safe grasping, holding, stretching and moving of the slippery and easily torn tissues. The paper describes how a sensor-enabled, smart version of the gripper was engineered. The capabilities of the tool were primarily tested through laboratory dry tests and on pig-carcasses in a local slaughterhouse. Advanced features for in-device force, position and slip sensing are being developed for future use.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128079960","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 : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668569
Kristóf Druzsin, Rita Fleiner, A. Rusznák, P. Biró
Kidney exchange programmes (KEPs) have been organized for patients to exchange their willing, but incompatible donors among each other in a framework controlled by experts. Thousands of patients have already been matched and received kidneys from compatible donors in national and international KEPs in Europe, and elsewhere. To evaluate the performance of KEPs various simulator tools have been developed and tested on real historical and generated data for difference settings and optimization polices. In this paper, we propose a database model that can be used in such KEP simulator tools, and can serve as an example for databases in information systems of real applications.
{"title":"Database model for kidney exchange programmes simulation tool","authors":"Kristóf Druzsin, Rita Fleiner, A. Rusznák, P. Biró","doi":"10.1109/CINTI53070.2021.9668569","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668569","url":null,"abstract":"Kidney exchange programmes (KEPs) have been organized for patients to exchange their willing, but incompatible donors among each other in a framework controlled by experts. Thousands of patients have already been matched and received kidneys from compatible donors in national and international KEPs in Europe, and elsewhere. To evaluate the performance of KEPs various simulator tools have been developed and tested on real historical and generated data for difference settings and optimization polices. In this paper, we propose a database model that can be used in such KEP simulator tools, and can serve as an example for databases in information systems of real applications.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133553200","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 : 2021-11-18DOI: 10.1109/CINTI53070.2021.9668392
A. Mason, O. Korostynska, L. E. Cordova-Lopez, I. Esper, D. Romanov, S. Ross, K. Takacs, T. Haidegger
This paper provides a brief overview of the novel Meat Factory Cell and discusses its concept in the context of increasing sustainability in the meat sector. Job quality, environment, health risks, industrial development and education are discussed as sustainability goals that can be mapped against some of the United Nations Sustainable Development Goals (SDG). Technology can arguably help to improve related processes on a societal level, and to achieve the SDGs.
{"title":"Meat Factory Cell: Assisting meat processors address sustainability in meat production","authors":"A. Mason, O. Korostynska, L. E. Cordova-Lopez, I. Esper, D. Romanov, S. Ross, K. Takacs, T. Haidegger","doi":"10.1109/CINTI53070.2021.9668392","DOIUrl":"https://doi.org/10.1109/CINTI53070.2021.9668392","url":null,"abstract":"This paper provides a brief overview of the novel Meat Factory Cell and discusses its concept in the context of increasing sustainability in the meat sector. Job quality, environment, health risks, industrial development and education are discussed as sustainability goals that can be mapped against some of the United Nations Sustainable Development Goals (SDG). Technology can arguably help to improve related processes on a societal level, and to achieve the SDGs.","PeriodicalId":340545,"journal":{"name":"2021 IEEE 21st International Symposium on Computational Intelligence and Informatics (CINTI)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115619370","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：