Osamu Sugiyama - Academia.edu (original) (raw)

Papers by Osamu Sugiyama

PubMed, Aug 21, 2019

Redesigning Electronic Medical Record (EMR) systems is needed to improve their usefulness and usa... more Redesigning Electronic Medical Record (EMR) systems is needed to improve their usefulness and usability. For user-centered redesign, designers should consider which EMR features are the most important to the users. However, prioritizing the EMR features is complicated because: (i) EMR systems involve multiple users with different, and sometimes conflicting, priorities and (ii) targeting one feature will affect other features of the EMR system. In this work, we propose a method for prioritizing the features to target when redesigning an EMR system. The method takes into consideration the different priorities of the users and the relationships between the different features. We illustrate the method through a case study on redesigning EMR systems in Japanese antenatal care settings. Our results show the importance of considering the different types of EMR users and the relationships between different EMR features. Designers could use the proposed method as a decision-aid tool in EMR redesign projects.

European Journal for Biomedical Informatics, 2018

Electronic medical records (hereafter called "EMRs") are designed for the primary use of data, su... more Electronic medical records (hereafter called "EMRs") are designed for the primary use of data, such as the efficiency of medical work by order entry systems and medical accounting systems. Secondary data, such as medical research, decision support, and educational support for young doctors, are expected to accumulate via promoting the computerization of medical data. Especially, computer-based clinical decision support systems (hereafter called "CDSS") to support doctor's clinical decision making is an important example of secondary use of EMRs [1]. The CDSS is a system that provides better medical care to patients by reducing decision mistakes and sharing medical evidence when medical staff makes decisions, such as diagnosis and prescription [2].

JMIR serious games, Jul 22, 2019

Background: The novel genre of pervasive games, which aim to create more fun and engaging experie... more Background: The novel genre of pervasive games, which aim to create more fun and engaging experiences by promoting deeper immersion, could be a powerful strategy to stimulate physical activity among older adults. To use these games more effectively, it is necessary to understand how different design elements affect player behavior. Objective: The aim was to vary a specific design element of pervasive games for older adults, namely social interaction, to test the effect on levels of physical activity. Methods: Over 4 weeks, two variations of the same pervasive game were compared: social interaction for the test group and no social interaction for the control group. In both versions, players had to walk to physical locations and collect virtual cards, but the social interaction version allowed people to collaborate to obtain more cards. Weekly step counts were used to evaluate the effect on each group, and the number of places visited was used as an indicator of play activity. Results: A total of 32 participants were recruited (no social interaction=15, social interaction=17); 18 remained until the end of the study (no social interaction=7, social interaction=11). Step counts during the first week were used as the baseline (no social interaction: mean 17,099.4, SE 3906.5; social interaction: mean 17,981.9, SE 2171.1). For the following weeks, changes to individual baseline were as follows for no social interaction (absolute/proportional): 383.8 (SE 563.8)/1.1% (SE 4.3%), 435.9 (SE 574.5)/2.2% (SE 4.6%), and −106.1 (SE 979.9)/−2.6% (SE 8.1%) for weeks 2, 3, and 4, respectively. For social interaction they were 3841.9 (SE 1425.4)/21.7% (SE 5.1%), 2270.6 (SE 947.1)/16.5% (SE 4.4%), and 2443.4 (SE 982.6)/17.9% (SE 4.7%) for weeks 2, 3, and 4, respectively. Analysis of group effect was significant (absolute change: η 2 =.19, P=.01; proportional change: η 2 =.27, P=.009). Correlation between the proportional change and the play activity was significant (r=.34, 95% CI 0.08 to 0.56), whereas for absolute change it was not.

PLOS ONE, Jul 27, 2018

We developed a computer-aided diagnosis (CADx) method for classification between benign nodule, p... more We developed a computer-aided diagnosis (CADx) method for classification between benign nodule, primary lung cancer, and metastatic lung cancer and evaluated the following: (i) the usefulness of the deep convolutional neural network (DCNN) for CADx of the ternary classification, compared with a conventional method (hand-crafted imaging feature plus machine learning), (ii) the effectiveness of transfer learning, and (iii) the effect of image size as the DCNN input. Among 1240 patients of previously-built database, computed tomography images and clinical information of 1236 patients were included. For the conventional method, CADx was performed by using rotation-invariant uniform-pattern local binary pattern on three orthogonal planes with a support vector machine. For the DCNN method, CADx was evaluated using the VGG-16 convolutional neural network with and without transfer learning, and hyperparameter optimization of the DCNN method was performed by random search. The best averaged validation accuracies of CADx were 55.9%, 68.0%, and 62.4% for the conventional method, the DCNN method with transfer learning, and the DCNN method without transfer learning, respectively. For image size of 56, 112, and 224, the best averaged validation accuracy for the DCNN with transfer learning were 60.7%, 64.7%, and 68.0%, respectively. DCNN was better than the conventional method for CADx, and the accuracy of DCNN improved when using transfer learning. Also, we found that larger image sizes as inputs to DCNN improved the accuracy of lung nodule classification.

Value in Health, Oct 1, 2018

Background: To parse free text medical notes into structured data such as disease names, drugs, p... more Background: To parse free text medical notes into structured data such as disease names, drugs, procedures, and other important medical information first, it is necessary to detect medical entities. It is important for an Electronic Medical Record (EMR) to have structured data with semantic interoperability to serve as a seamless communication platform whenever a patient migrates from one physician to another. However, in free text notes, medical entities are often expressed using informal abbreviations. An informal abbreviation is a non-standard or undetermined abbreviation, made in diverse writing styles, which may burden the semantic interoperability between EMR systems. Therefore, a detection of informal abbreviations is required to tackle this issue. Objectives: We attempt to achieve highly reliable detection of informal abbreviations made in diverse writing styles. Methods: In this study, we apply the Long Short-Term Memory (LSTM) model to detect informal abbreviations in free text medical notes. Additionally, we use sliding windows to tackle the limited data issue and sample generator for the imbalance class issue, while introducing additional pre-trained features (bag of words and word2vec vectors) to the model. Results: The LSTM model was able to detect informal abbreviations with precision of 93.6%, recall of 57.6%, and F1-score of 68.9%. Conclusion: Our method was able to recognize informal abbreviations using small data set with high precision. The detection can be used to recognize informal abbreviations in real-time while the physician is typing it and raise appropriate indicators for the informal abbreviation meaning confirmation, thus increase the semantic interoperability.

Walking 8,000 steps in a day is one of the important criteria to maintain our health. However, we... more Walking 8,000 steps in a day is one of the important criteria to maintain our health. However, we often miss a chance to walk due to the difficulty to keep our motivation toward our health in a daily life. We propose an algorithm to search an appropriate daily walking pattern from the user's past walking record. The searched walking patterns are used for making a health promotion agent recommend an effective timing to walk. With this recommendation, users will not miss the timing when they can walk. In this study, we focused on designing the algorithm for searching a daily walking pattern, which satisfied both conditions, achieving 8,000 steps a day and being similar to the current user walking record. In a pilot performance study, it was revealed that the proposed algorithm can narrow down the walking pattern candidates and properly search the walking pattern similar to the current user walking record.

Smart innovation, systems and technologies, May 21, 2017

During the Kumamoto earthquakes in Japan, disaster medical assistance teams (DMATs) were dispatch... more During the Kumamoto earthquakes in Japan, disaster medical assistance teams (DMATs) were dispatched for emergency support. Communication among DMAT members were primarily done via emails and phones, however, during this disaster, some teams also used LINE, a popular social networking service in Japan. Although this tool is simple to use, the teams had problems organizing various topics in a single chat room. In this paper, we propose an application that uses hashtags, which consists of two main units: (1) a bot that redirects messages to specific groups according to hashtags input by users; and (2) a system for logistic-support to manually apply hashtags to messages without tags, and to manually edit hashtags of already-tagged messages. User studies of two Kyoto University Hospital DMAT members were conducted, and through discussion, we found that the generality of the proposed application should be further considered for usage in other activities.

Smart innovation, systems and technologies, 2019

We present the design process and evaluation of a set of new social interaction mechanics in a mo... more We present the design process and evaluation of a set of new social interaction mechanics in a mobile, location-based game, developed to explore the effect of variations in design elements in elderly people’s levels of physical activity. The game had previously been evaluated in Kyoto, Japan, and new social interaction mechanics were proposed and evaluated in a different cultural context, Brazil, with a group of elderly volunteers in the Brazilian city of Pelotas. The cultural adaptation was made in a way to preserve the core design principles of the game and allow for evaluation of the new proposed social interactions, which were found to create a more enjoyable and engaging experience for the players.

Companion of the 2018 ACM Conference on Computer Supported Cooperative Work and Social Computing

2019 International Conference on Computer, Control, Informatics and its Applications (IC3INA)

Clinical research has led to the development of new medications and medication strategy may chang... more Clinical research has led to the development of new medications and medication strategy may change to gain better treatment outcomes. With the increasing attention to the evidence based medical guideline, it is important for clinicians to study the medication strategy changes. This study is interested in evaluating long-term prescriptions of type 2 diabetes patients provided by Kyoto University Hospital following the release of a new medication in 2010. Frequent sequential pattern mining (FSPM) is a prominent tools for extracting frequent patterns. However, the number of the result set could be enormous and may inhibit clinicians in assessing the results. To help the clinicians, a medication progression graph (MPG) is constructed using adjacent (1-sequence) frequent patterns produced by the singleton pattern mining. Compared to conventional frequent patterns, singleton’s patterns features full itemsets and 1-step distance. Hence, the pattern represents the true medication transition event. Using the graph, a clinical physician was able to observe a significant increase of physician’s activities in changing the medication after the release (i.e., in MPG that start with Sulfonylurea after 2010, the number of edges is increase by 2.83 and the number of nodes is increase by 2.27 compared to before 2010). Our preliminary results show that the new visualization method enables a clinician to analyze a medication strategy changes after the new medication released.

Advanced Biomedical Engineering, 2022

Appropriate evaluation of the intraoperative state of a surgical team is essential for the improv... more Appropriate evaluation of the intraoperative state of a surgical team is essential for the improvement of teamwork and hence a safe surgical environment. Traditional methods to evaluate intraoperative team states such as interview and self-check questionnaire on each surgical team member often require human efforts, which are time-consuming and can be biased by individual recall. One effective solution is to analyze the surgical video and track the important team activities, such as whether the members are complying with the surgical procedure or are being distracted by unexpected events. However, due to the complexity of the situations in an operating room, identifying the team activities without any human effort remains challenging. In this work, we propose a novel approach that automatically recognizes and quanti es intraoperative activities from surgery videos. As a rst step, we focus on recognizing two activities that especially involve multiple individuals: (a) passing of clean-packaged surgery instruments which is a representative interaction between the surgical technologists such as the circulating nurse and scrub nurse, and (b) group attention that may be attracted by unexpected events. We record surgical videos as input, and apply pose estimation and particle lters to extract individual s face orientation, body orientation, and arm raise. These results coupled with individual IDs are then sent to an estimation model that provides the probability of each target activity. Simultaneously, a person model is generated and bound to each individual, which describes all the involved activities along the timeline. We tested our method using videos of simulated activities. The results showed that the system was able to recognize instrument passing and group attention with F1 = 0.95 and F1 = 0.66, respectively. We also implemented a system with an interface that automatically annotated intraoperative activities along the video timeline, and invited feedback from surgical technologists. The results suggest that the quanti ed and visualized activities can help improve understanding of the intraoperative state of the surgical team.

Advanced Biomedical Engineering, 2022

Designing a deep neural network model that integrates clinical images with other electronic medic... more Designing a deep neural network model that integrates clinical images with other electronic medical records entails various preprocessing operations. Preprocessing of clinical images often requires trimming of parts of the lesions shown in the images, whereas preprocessing of other electronic medical records requires vectorization of these records; for example, patient age is often converted into a categorical vector of 10-year intervals. Although these preprocessing operations are critical to the performance of the classi cation model, there is no guarantee that the preprocessing step chosen is appropriate for model training. The ability to integrate these preprocessing operations into a deep neural network model and to train the model, including the preprocessing operations, can help design a multi-modal medical classi cation model. This study proposes integration layers of preprocessing, both for clinical images and electronic medical records, in deep neural network models. Preprocessing of clinical images is realized by a vision transformer layer that selectively adopts the parts of the images requiring attention. The preprocessing of other medical electrical records is performed by adopting full-connection layers and normalizing these layers. These proposed preprocessing-integrated layers were veri ed using a posttreatment visual acuity prediction task in ophthalmology as a case study. This prediction task requires clinical images as well as patient pro le data corresponding to each patient s posttreatment logMAR visual acuity. The performance of a heuristically designed prediction model was compared with the performance of the prediction model that includes the proposed preprocessing integration layers. The mean square errors between predicted and correct results were 0.051 for the heuristic model and 0.054 for the proposed model. Experimental results showed that the proposed model utilizing preprocessing integration layers achieved nearly the same performance as the heuristically designed model.

Lecture Notes in Computer Science, 2019

We present the design process and evaluation of a pervasive, location-based mobile game created t... more We present the design process and evaluation of a pervasive, location-based mobile game created to act as an experiment system and allow evaluation of how different design elements can influence player behaviour, using social interaction as a study case. A feasibility study with a group of community dwelling elderly volunteers from the city of Kyoto, Japan, was performed to evaluate the system. Results showed that the choice of theme and overall design of game was adequate, and that elderly people could understand the game rules and their goals while playing. Points of improvement included reducing the complexity of game controls and changing social interaction mechanics to account for situations when there are only a few players active or players are too far apart.

BACKGROUND Pervasive games aim to create more fun and engaging experiences by mixing elements fro... more BACKGROUND Pervasive games aim to create more fun and engaging experiences by mixing elements from the real world into the game world. Because they intermingle with players’ lives and naturally promote more casual gameplay, they could be a powerful strategy to stimulate physical activity among older adults. However, to use these games more effectively, it is necessary to understand how design elements of the game affect player behavior. OBJECTIVE The aim of this study was to evaluate how the presence of a specific design element, namely social interaction, would affect levels of physical activity. METHODS Participants were recruited offline and randomly assigned to control and intervention groups in a single-blind design. Over 4 weeks, two variations of the same pervasive game were compared: with social interaction (intervention group) and with no social interaction (control group). In both versions, players had to walk to physical locations and collect virtual cards, but the social...

Studies in health technology and informatics, 2020

Effective bed management is important for hospital management. Until now, bed allocation process ... more Effective bed management is important for hospital management. Until now, bed allocation process is generally controlled by administrative staffs in centralized manner but it is not always effective. In the present study, we proposed and evaluated new method for bed allocation applying market mechanism via token. Evaluation was performed with newly-developed game-type simulation. Nurse managers as research participants played it and answered for survey. The result showed that the proposed method can be useful with appropriate operational design.

Studies in health technology and informatics, 2020

Electronic Medical Record (EMR) systems are complex systems with interdependent features. Redesig... more Electronic Medical Record (EMR) systems are complex systems with interdependent features. Redesigning one feature of the system can create a cascade effect affecting the other features. By calculating the cascade effect, the designers can understand how each individual feature could be affected. This understanding allows them to maximize the positive effects and avoid negative consequences of their redesign activities. To understand the cascade effect, the designers can look at their computations' results; a task that becomes more difficult when the number of features grows. To reduce their task load, we propose a tool for visualizing the cascade effect of redesigning features in an EMR system. Our preliminary evaluation with six graduate students shows that visualizing the cascade effect reduces the task load and slightly improves their performance when analyzing the cascade effect. Ways for improving the tool include (i) showing the computation results within the visualization...

Studies in health technology and informatics, 2020

The goal of this research was to design a solution to detect non-reported incidents, especially s... more The goal of this research was to design a solution to detect non-reported incidents, especially severe incidents. To achieve this goal, we proposed a method to process electronic medical records and automatically extract clinical notes describing severe incidents. To evaluate the proposed method, we implemented a system and used the system. The system successfully detected a non-reported incident to the safety management department.

PubMed, Aug 21, 2019

Redesigning Electronic Medical Record (EMR) systems is needed to improve their usefulness and usa... more Redesigning Electronic Medical Record (EMR) systems is needed to improve their usefulness and usability. For user-centered redesign, designers should consider which EMR features are the most important to the users. However, prioritizing the EMR features is complicated because: (i) EMR systems involve multiple users with different, and sometimes conflicting, priorities and (ii) targeting one feature will affect other features of the EMR system. In this work, we propose a method for prioritizing the features to target when redesigning an EMR system. The method takes into consideration the different priorities of the users and the relationships between the different features. We illustrate the method through a case study on redesigning EMR systems in Japanese antenatal care settings. Our results show the importance of considering the different types of EMR users and the relationships between different EMR features. Designers could use the proposed method as a decision-aid tool in EMR redesign projects.

European Journal for Biomedical Informatics, 2018

Electronic medical records (hereafter called "EMRs") are designed for the primary use of data, su... more Electronic medical records (hereafter called "EMRs") are designed for the primary use of data, such as the efficiency of medical work by order entry systems and medical accounting systems. Secondary data, such as medical research, decision support, and educational support for young doctors, are expected to accumulate via promoting the computerization of medical data. Especially, computer-based clinical decision support systems (hereafter called "CDSS") to support doctor's clinical decision making is an important example of secondary use of EMRs [1]. The CDSS is a system that provides better medical care to patients by reducing decision mistakes and sharing medical evidence when medical staff makes decisions, such as diagnosis and prescription [2].

JMIR serious games, Jul 22, 2019

Background: The novel genre of pervasive games, which aim to create more fun and engaging experie... more Background: The novel genre of pervasive games, which aim to create more fun and engaging experiences by promoting deeper immersion, could be a powerful strategy to stimulate physical activity among older adults. To use these games more effectively, it is necessary to understand how different design elements affect player behavior. Objective: The aim was to vary a specific design element of pervasive games for older adults, namely social interaction, to test the effect on levels of physical activity. Methods: Over 4 weeks, two variations of the same pervasive game were compared: social interaction for the test group and no social interaction for the control group. In both versions, players had to walk to physical locations and collect virtual cards, but the social interaction version allowed people to collaborate to obtain more cards. Weekly step counts were used to evaluate the effect on each group, and the number of places visited was used as an indicator of play activity. Results: A total of 32 participants were recruited (no social interaction=15, social interaction=17); 18 remained until the end of the study (no social interaction=7, social interaction=11). Step counts during the first week were used as the baseline (no social interaction: mean 17,099.4, SE 3906.5; social interaction: mean 17,981.9, SE 2171.1). For the following weeks, changes to individual baseline were as follows for no social interaction (absolute/proportional): 383.8 (SE 563.8)/1.1% (SE 4.3%), 435.9 (SE 574.5)/2.2% (SE 4.6%), and −106.1 (SE 979.9)/−2.6% (SE 8.1%) for weeks 2, 3, and 4, respectively. For social interaction they were 3841.9 (SE 1425.4)/21.7% (SE 5.1%), 2270.6 (SE 947.1)/16.5% (SE 4.4%), and 2443.4 (SE 982.6)/17.9% (SE 4.7%) for weeks 2, 3, and 4, respectively. Analysis of group effect was significant (absolute change: η 2 =.19, P=.01; proportional change: η 2 =.27, P=.009). Correlation between the proportional change and the play activity was significant (r=.34, 95% CI 0.08 to 0.56), whereas for absolute change it was not.

PLOS ONE, Jul 27, 2018

We developed a computer-aided diagnosis (CADx) method for classification between benign nodule, p... more We developed a computer-aided diagnosis (CADx) method for classification between benign nodule, primary lung cancer, and metastatic lung cancer and evaluated the following: (i) the usefulness of the deep convolutional neural network (DCNN) for CADx of the ternary classification, compared with a conventional method (hand-crafted imaging feature plus machine learning), (ii) the effectiveness of transfer learning, and (iii) the effect of image size as the DCNN input. Among 1240 patients of previously-built database, computed tomography images and clinical information of 1236 patients were included. For the conventional method, CADx was performed by using rotation-invariant uniform-pattern local binary pattern on three orthogonal planes with a support vector machine. For the DCNN method, CADx was evaluated using the VGG-16 convolutional neural network with and without transfer learning, and hyperparameter optimization of the DCNN method was performed by random search. The best averaged validation accuracies of CADx were 55.9%, 68.0%, and 62.4% for the conventional method, the DCNN method with transfer learning, and the DCNN method without transfer learning, respectively. For image size of 56, 112, and 224, the best averaged validation accuracy for the DCNN with transfer learning were 60.7%, 64.7%, and 68.0%, respectively. DCNN was better than the conventional method for CADx, and the accuracy of DCNN improved when using transfer learning. Also, we found that larger image sizes as inputs to DCNN improved the accuracy of lung nodule classification.

Value in Health, Oct 1, 2018

Background: To parse free text medical notes into structured data such as disease names, drugs, p... more Background: To parse free text medical notes into structured data such as disease names, drugs, procedures, and other important medical information first, it is necessary to detect medical entities. It is important for an Electronic Medical Record (EMR) to have structured data with semantic interoperability to serve as a seamless communication platform whenever a patient migrates from one physician to another. However, in free text notes, medical entities are often expressed using informal abbreviations. An informal abbreviation is a non-standard or undetermined abbreviation, made in diverse writing styles, which may burden the semantic interoperability between EMR systems. Therefore, a detection of informal abbreviations is required to tackle this issue. Objectives: We attempt to achieve highly reliable detection of informal abbreviations made in diverse writing styles. Methods: In this study, we apply the Long Short-Term Memory (LSTM) model to detect informal abbreviations in free text medical notes. Additionally, we use sliding windows to tackle the limited data issue and sample generator for the imbalance class issue, while introducing additional pre-trained features (bag of words and word2vec vectors) to the model. Results: The LSTM model was able to detect informal abbreviations with precision of 93.6%, recall of 57.6%, and F1-score of 68.9%. Conclusion: Our method was able to recognize informal abbreviations using small data set with high precision. The detection can be used to recognize informal abbreviations in real-time while the physician is typing it and raise appropriate indicators for the informal abbreviation meaning confirmation, thus increase the semantic interoperability.

Walking 8,000 steps in a day is one of the important criteria to maintain our health. However, we... more Walking 8,000 steps in a day is one of the important criteria to maintain our health. However, we often miss a chance to walk due to the difficulty to keep our motivation toward our health in a daily life. We propose an algorithm to search an appropriate daily walking pattern from the user's past walking record. The searched walking patterns are used for making a health promotion agent recommend an effective timing to walk. With this recommendation, users will not miss the timing when they can walk. In this study, we focused on designing the algorithm for searching a daily walking pattern, which satisfied both conditions, achieving 8,000 steps a day and being similar to the current user walking record. In a pilot performance study, it was revealed that the proposed algorithm can narrow down the walking pattern candidates and properly search the walking pattern similar to the current user walking record.

Smart innovation, systems and technologies, May 21, 2017

During the Kumamoto earthquakes in Japan, disaster medical assistance teams (DMATs) were dispatch... more During the Kumamoto earthquakes in Japan, disaster medical assistance teams (DMATs) were dispatched for emergency support. Communication among DMAT members were primarily done via emails and phones, however, during this disaster, some teams also used LINE, a popular social networking service in Japan. Although this tool is simple to use, the teams had problems organizing various topics in a single chat room. In this paper, we propose an application that uses hashtags, which consists of two main units: (1) a bot that redirects messages to specific groups according to hashtags input by users; and (2) a system for logistic-support to manually apply hashtags to messages without tags, and to manually edit hashtags of already-tagged messages. User studies of two Kyoto University Hospital DMAT members were conducted, and through discussion, we found that the generality of the proposed application should be further considered for usage in other activities.

Smart innovation, systems and technologies, 2019

We present the design process and evaluation of a set of new social interaction mechanics in a mo... more We present the design process and evaluation of a set of new social interaction mechanics in a mobile, location-based game, developed to explore the effect of variations in design elements in elderly people’s levels of physical activity. The game had previously been evaluated in Kyoto, Japan, and new social interaction mechanics were proposed and evaluated in a different cultural context, Brazil, with a group of elderly volunteers in the Brazilian city of Pelotas. The cultural adaptation was made in a way to preserve the core design principles of the game and allow for evaluation of the new proposed social interactions, which were found to create a more enjoyable and engaging experience for the players.

Companion of the 2018 ACM Conference on Computer Supported Cooperative Work and Social Computing

2019 International Conference on Computer, Control, Informatics and its Applications (IC3INA)

Clinical research has led to the development of new medications and medication strategy may chang... more Clinical research has led to the development of new medications and medication strategy may change to gain better treatment outcomes. With the increasing attention to the evidence based medical guideline, it is important for clinicians to study the medication strategy changes. This study is interested in evaluating long-term prescriptions of type 2 diabetes patients provided by Kyoto University Hospital following the release of a new medication in 2010. Frequent sequential pattern mining (FSPM) is a prominent tools for extracting frequent patterns. However, the number of the result set could be enormous and may inhibit clinicians in assessing the results. To help the clinicians, a medication progression graph (MPG) is constructed using adjacent (1-sequence) frequent patterns produced by the singleton pattern mining. Compared to conventional frequent patterns, singleton’s patterns features full itemsets and 1-step distance. Hence, the pattern represents the true medication transition event. Using the graph, a clinical physician was able to observe a significant increase of physician’s activities in changing the medication after the release (i.e., in MPG that start with Sulfonylurea after 2010, the number of edges is increase by 2.83 and the number of nodes is increase by 2.27 compared to before 2010). Our preliminary results show that the new visualization method enables a clinician to analyze a medication strategy changes after the new medication released.

Advanced Biomedical Engineering, 2022

Appropriate evaluation of the intraoperative state of a surgical team is essential for the improv... more Appropriate evaluation of the intraoperative state of a surgical team is essential for the improvement of teamwork and hence a safe surgical environment. Traditional methods to evaluate intraoperative team states such as interview and self-check questionnaire on each surgical team member often require human efforts, which are time-consuming and can be biased by individual recall. One effective solution is to analyze the surgical video and track the important team activities, such as whether the members are complying with the surgical procedure or are being distracted by unexpected events. However, due to the complexity of the situations in an operating room, identifying the team activities without any human effort remains challenging. In this work, we propose a novel approach that automatically recognizes and quanti es intraoperative activities from surgery videos. As a rst step, we focus on recognizing two activities that especially involve multiple individuals: (a) passing of clean-packaged surgery instruments which is a representative interaction between the surgical technologists such as the circulating nurse and scrub nurse, and (b) group attention that may be attracted by unexpected events. We record surgical videos as input, and apply pose estimation and particle lters to extract individual s face orientation, body orientation, and arm raise. These results coupled with individual IDs are then sent to an estimation model that provides the probability of each target activity. Simultaneously, a person model is generated and bound to each individual, which describes all the involved activities along the timeline. We tested our method using videos of simulated activities. The results showed that the system was able to recognize instrument passing and group attention with F1 = 0.95 and F1 = 0.66, respectively. We also implemented a system with an interface that automatically annotated intraoperative activities along the video timeline, and invited feedback from surgical technologists. The results suggest that the quanti ed and visualized activities can help improve understanding of the intraoperative state of the surgical team.

Advanced Biomedical Engineering, 2022

Designing a deep neural network model that integrates clinical images with other electronic medic... more Designing a deep neural network model that integrates clinical images with other electronic medical records entails various preprocessing operations. Preprocessing of clinical images often requires trimming of parts of the lesions shown in the images, whereas preprocessing of other electronic medical records requires vectorization of these records; for example, patient age is often converted into a categorical vector of 10-year intervals. Although these preprocessing operations are critical to the performance of the classi cation model, there is no guarantee that the preprocessing step chosen is appropriate for model training. The ability to integrate these preprocessing operations into a deep neural network model and to train the model, including the preprocessing operations, can help design a multi-modal medical classi cation model. This study proposes integration layers of preprocessing, both for clinical images and electronic medical records, in deep neural network models. Preprocessing of clinical images is realized by a vision transformer layer that selectively adopts the parts of the images requiring attention. The preprocessing of other medical electrical records is performed by adopting full-connection layers and normalizing these layers. These proposed preprocessing-integrated layers were veri ed using a posttreatment visual acuity prediction task in ophthalmology as a case study. This prediction task requires clinical images as well as patient pro le data corresponding to each patient s posttreatment logMAR visual acuity. The performance of a heuristically designed prediction model was compared with the performance of the prediction model that includes the proposed preprocessing integration layers. The mean square errors between predicted and correct results were 0.051 for the heuristic model and 0.054 for the proposed model. Experimental results showed that the proposed model utilizing preprocessing integration layers achieved nearly the same performance as the heuristically designed model.

Lecture Notes in Computer Science, 2019

We present the design process and evaluation of a pervasive, location-based mobile game created t... more We present the design process and evaluation of a pervasive, location-based mobile game created to act as an experiment system and allow evaluation of how different design elements can influence player behaviour, using social interaction as a study case. A feasibility study with a group of community dwelling elderly volunteers from the city of Kyoto, Japan, was performed to evaluate the system. Results showed that the choice of theme and overall design of game was adequate, and that elderly people could understand the game rules and their goals while playing. Points of improvement included reducing the complexity of game controls and changing social interaction mechanics to account for situations when there are only a few players active or players are too far apart.

BACKGROUND Pervasive games aim to create more fun and engaging experiences by mixing elements fro... more BACKGROUND Pervasive games aim to create more fun and engaging experiences by mixing elements from the real world into the game world. Because they intermingle with players’ lives and naturally promote more casual gameplay, they could be a powerful strategy to stimulate physical activity among older adults. However, to use these games more effectively, it is necessary to understand how design elements of the game affect player behavior. OBJECTIVE The aim of this study was to evaluate how the presence of a specific design element, namely social interaction, would affect levels of physical activity. METHODS Participants were recruited offline and randomly assigned to control and intervention groups in a single-blind design. Over 4 weeks, two variations of the same pervasive game were compared: with social interaction (intervention group) and with no social interaction (control group). In both versions, players had to walk to physical locations and collect virtual cards, but the social...

Studies in health technology and informatics, 2020

Effective bed management is important for hospital management. Until now, bed allocation process ... more Effective bed management is important for hospital management. Until now, bed allocation process is generally controlled by administrative staffs in centralized manner but it is not always effective. In the present study, we proposed and evaluated new method for bed allocation applying market mechanism via token. Evaluation was performed with newly-developed game-type simulation. Nurse managers as research participants played it and answered for survey. The result showed that the proposed method can be useful with appropriate operational design.

Studies in health technology and informatics, 2020

Electronic Medical Record (EMR) systems are complex systems with interdependent features. Redesig... more Electronic Medical Record (EMR) systems are complex systems with interdependent features. Redesigning one feature of the system can create a cascade effect affecting the other features. By calculating the cascade effect, the designers can understand how each individual feature could be affected. This understanding allows them to maximize the positive effects and avoid negative consequences of their redesign activities. To understand the cascade effect, the designers can look at their computations' results; a task that becomes more difficult when the number of features grows. To reduce their task load, we propose a tool for visualizing the cascade effect of redesigning features in an EMR system. Our preliminary evaluation with six graduate students shows that visualizing the cascade effect reduces the task load and slightly improves their performance when analyzing the cascade effect. Ways for improving the tool include (i) showing the computation results within the visualization...

Studies in health technology and informatics, 2020

The goal of this research was to design a solution to detect non-reported incidents, especially s... more The goal of this research was to design a solution to detect non-reported incidents, especially severe incidents. To achieve this goal, we proposed a method to process electronic medical records and automatically extract clinical notes describing severe incidents. To evaluate the proposed method, we implemented a system and used the system. The system successfully detected a non-reported incident to the safety management department.