We study the problem Rendezvous for two autonomous mobile robots in asynchronous settings with persistent memory called light. It is well known that Rendezvous is impossible in a basic model when robots have no lights, even if the system is semi-synchronous. On the other hand, Rendezvous is possible if robots have lights of various types with a constant number of colors [10], [21]. With external-lights, robots can only observe the state of the lights of other robots. With internal-lights, robots can only observe their own light, and full-lights combine both. This paper focuses on robots with external-lights in asynchronous settings and considers a particular class of algorithms called L-algorithms, where an L-algorithm computes a destination based only on the current colors of observable lights. When considering L-algorithms, Rendezvous can be solved by robots with full-lights and 3 colors in general asynchronous settings (called ASYNC) and the number of colors is optimal under these assumptions. In contrast, there exist no L-algorithms in ASYNC with external-lights regardless of the number of colors [10]. This paper, extending the impossibility result, shows that there exists no L-algorithm in so-called LC-1-Bounded ASYNC with external-lights regardless of the number of colors, where LC-1-Bounded ASYNC is a proper subset of ASYNC in which no robot can execute more than 1 Look operation between the Look and its subsequent Compute operations of another robot. We also show that LC-1-Bounded ASYNC is the minimal subclass in which no L-algorithms with external-lights exist. That is, Rendezvous can be solved by L-algorithms using external-lights with a finite number of colors in LC-0-Bounded ASYNC (equivalently LC-atomic ASYNC). Furthermore, we show that the algorithms are optimal in the number of colors they use.

In the cases when public health authorities confirm patients with highly contagious diseases, they release summaries about patient locations and travel information to alert the public. Due to privacy concerns, these releases do not include the detailed data, and typically comprise the information about commercial facilities and public transportation used by the patients. This paper proposes a framework to represent the information in Resource Description Framework (RDF). The open data would then be used by residents for automated estimation of the potential risks of contacts, combined with the location information stored in their mobile phones.

We study the Rendezvous problem for two autonomous mobile robots in asynchronous settings with persistent memory called light. It is well known that Rendezvous is impossible in a basic model when robots have no lights, even if the system is semi-synchronous. On the other hand, Rendezvous is possible if robots have lights of various types with a constant number of colors. If robots can observe not only their own lights but also other robots' lights, their lights are called full-light. If robots can only observe the state of other robots' lights, the lights are called external-light. This paper focuses on robots with external-lights in asynchronous settings and a particular class of algorithms called L-algorithms, where an L-algorithm computes a destination based only on the current colors of observable lights. When considering L-algorithms, Rendezvous can be solved by robots with full-lights and three colors in general asynchronous settings (called ASYNC) and the number of colors is optimal under these assumptions. In contrast, there exist no L-algorithms in ASYNC with external-lights regardless of the number of colors. In this paper, extending the impossibility result, we show that there exist no L-algorithms in so-called LC-1-Bounded ASYNC with external-lights regardless of the number of colors, where LC-1-Bounded ASYNC is a proper subset of ASYNC and other robots can execute at most one Look operation between the Look operation of a robot and its subsequent Compute operation. We also show that LC-1-Bounded ASYNC is the minimal subclass in which no L-algorithms with external-lights exist. That is, Rendezvous can be solved by L-algorithms using external-lights with a finite number of colors in LC-0-Bounded ASYNC (equivalently LC-atomic ASYNC). Furthermore, we show that the algorithms are optimal in the number of colors they use.

Disease similarity is a useful measure that has potential application to various aspects of medicine. One such application is the mapping of diseases in a two-dimensional plane, which can be the foundation of a useful diagnostic reminder method called the “pivot and cluster strategy.” However, the mapping of diseases using a similarity measure has yet to be explored. This article investigates such a mapping, and quantifies its basic characteristics. We first collected mutual similarity data for 1,550 diseases using a machine learning approach. The calculated similarity data were then used to map the diseases using a “multidimensional scaling” algorithm. Quantitative analysis indicated that it is difficult to express all the diseases on the map and yet still show the similarity information between the items. Then, by restricting the input, the algorithm performed well in practice. To our knowledge, this is the first study to investigate the automated mapping of diseases on a plane for use in clinical practice.

The user interface of clinical decision support systems (CDSSs) has not yet been studied well. In a previous study, we investigated the presentation of diagnostic output and proposed the hierarchical representation of differential diagnosis lists that provides an effective way of interfacing with a CDSS. In this research, we propose the incorporation of relevance feedback to improve the user interface. Such feedback can be used in three ways: to sharpen the search query, to initiate a dialogue with the user with regard to the presence or absence of particular diagnostic information, and to improve the accuracy of the diagnostic algorithm. This research qualitatively evaluates our approach and demonstrates that relevance feedback has desirable characteristics as a user interface element for clinical decision support.

Disease similarity is a useful measure to improve clinical decision support systems wherein it allows continuous presentation of similar diseases. In a previous study, we demonstrated that etiological and symptomatic information of diseases provide a reasonable approximation for the similarity of diseases. This study extends the previously proposed approach by incorporating the locational information of diseases, which may improve the performance against the baseline achieved only by the etiological and symptomatic features.First, this study attempted to acquire anatomical information from disease names in Japanese and evaluated the relation between those disease names by constructing a graph that represents the relations of anatomical concepts. Then, locational information was used as a supplemental input to the disease similarity calculation in the machine learning framework developed for the previous study. The baseline performance of the algorithm presented in the previous study was 0.907 using the normalized disease similarity measure (NDSM) which utilizes 80 annotated diseases as the gold standard. The proposed approach with the locational feature achieved a higher score: 0.921. As expected, in-depth microscopic analysis revealed that this approach benefits the diseases that have tight association with specific locations in the human body. The results suggest that location is a promising feature for computing disease similarity, which is worth further investigation by acquiring locational knowledge from disease profiles.

Clinical decision support systems (CDSSs) require disease knowledge bases that need enormous manual work by costly physicians to develop. To overcome this obstacle, we constructed a disease knowledge base by integrating and processing existing disease databases, Online Mendelian Inheritance in Man (OMIM) and Orphanet. In the present study, the refinement process of the integrated dataset, which was required for CDSS applications, is discussed. Only 43% of the OMIM entries and 50% of the Orphanet entries, in the unified knowledge base, are considered appropriate for CDSS use. The remaining entries include congenital anomalies, genetic variants, and other inappropriate data. The resulting list and additional data of rare diseases suitable for clinical decision support will be made public to facilitate further researches on CDSSs with rare disease coverage.

For clinical decision support systems, the disease knowledge base accounts for the majority of the total development cost, because it requires considerable effort by highly-paid domain experts. However, existing research on automated acquisition of medical knowledge has focused on accuracy, while mostly ignoring the cost issue. This case study investigates the cost breakdown of our simplified disease knowledge base, and discusses a way of reducing development cost of the core component for clinical decision support systems. To achieve broad disease coverage with limited cost, we adopted a hybridization approach, that is, combining a handmade knowledge base for essential diseases and an automatically generated knowledge base for rare diseases. Toward further cost reduction, the case study suggested to minimize the human intervention by medical experts through i) keeping the structure of the knowledge base simple, ii) establishing a public resource for laboratory examination results, and iii) lazy evaluation of data quality in the utilization phase. Although the resulting knowledge base may not be adequate for a definitive diagnosis, the approach could be suitable to build clinical decision support systems for differential diagnosis and for disease search engines.

Text mining of clinical findings has been employed to extract clinical information contained in “electronic medical records” without the need for labor intensive work by medical experts. However, the automated building of disease ontology necessitates knowledge acquisition of clinical findings documented in “medical literature” that requires an independent strategy. This study performs a preliminary analysis of clinical finding expressions in medical literature to enable the automated acquisition of disease knowledge. To this end, we selected descriptions of 20 diseases in a free-text format and annotated the texts to extract expressions of clinical findings. This resulted in 1368 expressions with varying lengths and syntactic features, and 161 annotator comments. The comments suggested that certain types of expressions, which were further classified into 10 categories. Also, in-depth analyses of their syntactic and semantic characteristics were performed, resulting in the following observations. First, expressions of clinical findings have certain patterns, syntactic and semantic, which can be exploited for appropriate knowledge acquisition. Second, clinical knowledge may guide the knowledge acquisition process in a top-down manner. Third, natural language processing of medical literature requires specific considerations compared with the processing of health records, namely, i) distinction of subjects, ii) handling of generalized knowledge, and iii) processing of expressions for examination results. This preliminary survey on the expressions in medical literature provides helpful insights for future corpus design.

Knowledge about the causal relationship between diseases and their laboratory findings is a key component for clinical decision support systems. For efficient acquisition of such knowledge, this paper attempted to represent the interpretation of laboratory results in a guidebook for laboratory examinations. A preliminary survey revealed the structure of the knowledge compiled in the guidebook, and found essential patterns in the cause-effect relationship. We then attempted to code the knowledge, utilizing a simple cause-effect relationship between exam results and their possible causes, expressed in a disease master table. For coding of the knowledge, a two-step approach was used: first the causing disease was looked up automatically in the disease master table, and then, manually. In the study, 84.5% of the knowledge in the guidebook was identified as a candidate for the coding in the simple cause-effect relationship, and 69.1% of the knowledge was successfully coded. Failure analysis suggested that further expressive power in the representation is gained only at the cost of considerable human intervention in the knowledge acquisition, and the cost for the utilization of the resulting data. Accordingly, for a certain type of application, which might prefer simplicity over accuracy or completeness of the information, the minimalist representation could be a reasonable choice.

In this paper, we propose a coaxial-fed circular dipole array in a borehole (CFCAB) for directional borehole radar and present a design method for the CFCAB. In the CFCAB, a circular dipole array is arranged around a conducting cylinder. When a wave is incident on the antenna, we may divide the signals received at the antenna into two components. One is caused by the waves arriving at the antenna directly, and the other is due to the waves scattered at the conducting cylinder. We proposed a criterion to quantify the influence of the conducting cylinder on the radar measurement. We show results of experiments in air and in granite to examine the validity of the criterion. Making use of the criterion, we designed and built the directional borehole radar system with the CFCAB working at around 200 MHz. Conducting the field experiments in granite, we confirmed that the CFCAB can estimate the directions of arrival waves on the time-frequency plane, as we expected with the criterion.

Medical information systems carry sensitive data, which necessitates security and privacy. However, there is well-known trade-off between usability and security, and security can harm care providers and patients. This paper addresses the trade-off between security and usability in medical systems, and presents a case where higher security with better usability was achieved through close cooperation between computer scientists and care providers. The direct interaction might be costly for both, but appropriate tools could make the collaboration acceptable and practical.

Operating system extensions have been shown to be beneficial to implement custom kernel functionality. In most implementations, the extensions are made by an administrator with kernel loadable modules. An alternative approach is to provide a run-time system within the operating system itself that can execute user kernel extensions. In this paper, we describe such an approach,where a lightweight Java virtual machine is embedded within the kernel for flexible extension of kernel network I/O. For this purpose, we first implemented a compact Java Virtual Machine with a Just-In-Time compiler on the Intel IA32 instruction set architecture at the user space. Then, the virtual machine was embedded onto the FreeBSDoperating system kernel. We evaluate the system to validate the model, with systematic benchmarking.

Output of network file servers exhibits bursty traffic patterns, and this sometimes contends with control traffic. An example is contention between data traffic and write acknowledgments on a loaded server. In such a situation, we can improve the write performance by prioritizing write acknowledgments at the network interface of the server. To validate this scheme, we conducted an empirical study of the prioritization of write acknowledgments. Systematic experiments revealed that the proposed scheme can improve write latency and throughput of loaded servers without influencing read performance. The results suggested that the technique is widely applicable to systems where transactions are a mixture of read and write requests, and especially if the degree of concurrency is high.

There have been increasing demands for proper monitoring and control in end-host systems, mainly for security and QoS purposes. Nevertheless, existing technologies are insufficient as primitives for end-host security. For example, Berkeley packet filter (BPF), the most popular monitoring infrastructure for many Unix systems, is intended for packet capturing at physical interfaces, and thus, not appropriate for monitoring of applications, which is sometimes critical for system security. This paper presents a simple solution to the problem, utilizing hierarchical virtual network interface (VIF) mechanism. VIF is a new OS abstraction that can be hierarchically structured and attached to OS entities to control their network I/O. We extend VIFs to allow filtering and monitoring of their traffic, and show that it has desirable properties for end-host monitoring and control of traffic. We present our prototype implementation on FreeBSD, and evaluate it qualitatively and quantitatively. Demonstrated advantages include: i) ability to monitor terminating entities at arbitrary granularity, ii) a single consistent framework for both network security and network quality of service, iii) OS independence, iv) efficiency as a control primitive, v) compatibility with BPF interface and its applications, and vi) flexibility for future functional expansion.

In the recent past, with the advent of more powerful networks, computations have become more distributed in nature and control of network resources has become essential for operating systems (OS). Nevertheless, proposed primitives for network control at end-host OS are designed without an OS design perspective and have been in disagreement with existing OS constructs, causing a variety of problems. We propose a new OS service for network control, namely, hierarchical virtualization of network interface. The virtual network interface is hierarchically structured and attached to various OS constructs, such as threads, processes, and sockets, for the control of their network I/O. We show that our proposed mechanism provides the following properties: 1) flexible control granularity, 2) resource protection, 3) reasonable abstraction and application programming interface (API), and 4) various types of packet scheduling and control in a single framework, such as work-conserving and nonwork-conserving, in accordance with existing OS mechanisms. For a proof of concept, we present an implementation on a PC-Unix, using the file system abstraction, and carry out systematic profiling. The system exhibited the expected control behavior, that is, good responsiveness to the control commands while keeping the performance penalty small.

This paper tries to tackle the problem of providing retrofitting network QoS in clustered configurations. For this purpose, we designed a QoS manager which runs on each of the internal cluster nodes and controls network I/O of local interface cooperating with peer managers on other nodes towards a certain QoS policy. First, we show the design of control framework, contending that an end-host manager-based mechanism is a desirable approach, which utilizes an end-host oriented network control primitive, Netnice. Second, for flexibility of configuration, we propose object-oriented modeling of the QoS manager with event-handler based configuration mechanism, and show the design of an object-oriented configuration language that allow simple and flexible definition of QoS policies. Lastly, results from two simple experiments with a Web server cluster are analyzed.

In this paper, we present "netnice", a mechanism that allows processes to throttle their own network bandwidth consumption. As the name suggests, it is inspired by the Unix "nice" command in that it allows users and administrators to limit the network resources used by individual processes in order to avoid impacting the performance of other processes. In the paper, to address the problem of transient performance deterioration in local area network (LAN) environments, we propose a bandwidth limitation primitive that works within a host's kernel, called netnice. We also show several uses of netnice. Through experimentation in a small LAN of FreeBSD machines, we show how netnice allows for higher degree of controllability in bandwidth management.