Joon-Ho Lim

Finding the evidence for protein-protein interactions from PubMed abstracts

MOTIVATION Protein-protein interactions play critical roles in biological processes, and many biologists try to find or to predict crucial information concerning these interactions. Before verifying interactions in biological laboratory work, validating them from previous research is necessary. Although many efforts have been made to create databases that store verified information in a structured form, much interaction information still remains as unstructured text. As the amount of new publications has increased rapidly, a large amount of research has sought to extract interactions from the text automatically. However, there remain various difficulties associated with the process of applying automatically generated results into manually annotated databases. For interactions that are not found in manually stored databases, researchers attempt to search for abstracts or full papers. RESULTS As a result of a search for two proteins, PubMed frequently returns hundreds of abstracts. In this paper, a method is introduced that validates protein-protein interactions from PubMed abstracts. A query is generated from two given proteins automatically and abstracts are then collected from PubMed. Following this, target proteins and their synonyms are recognized and their interaction information is extracted from the collection. It was found that 67.37% of the interactions from DIP-PPI corpus were found from the PubMed abstracts and 87.37% of interactions were found from the given full texts. AVAILABILITY Contact authors.

ISO/IEEE 11073 PHD standardization of legacy healthcare devices for home healthcare services

As transformed to aging society, many personal devices for healthcare have recently been developed. According to the increase of individual development of health devices, the ISO/IEEE 11073 PHD standard is established for gathering health measurement data. But many existing healthcare devices are not standardized. In this paper, we propose implementation model of standardization for legacy healthcare devices which do not follow 11073 PHD standard. We develop universal PHD Adapter (UPA) and UPA interface board for ISO/IEEE 11073 PHD standardization of legacy (non-PHD) device for home healthcare service. Using the system, legacy devices communicate with 11073 PHD manager by 11073 PHD standard protocols.

Home healthcare settop-box for Senior Chronic Care using ISO/IEEE 11073 PHD standard

As the number of seniors with chronic disease increases, the need of home healthcare settop-box is increased to manage their chronic disease in their home environment. Using the home healthcare settop-box, the patients can regularly check their health data, and finally, it can lead the decrease of medical expenses. For the home healthcare settop-box, the most important factor is the standard compatibility, which can interoperate with standard devices of any other companies. In this paper, we propose a home healthcare settop-box using ISO/IEEE 11073 PHD standard. It collects health data according to the PHD standard, and provides a chronic-care service based on the collected data. The proposed settop-box is connected with 3 devices of weigh scale, blood pressure monitor, and glucose meter, and tested at 10 homes for a month. Lastly, the proposed settop-box can be used for various healthcare services such as Google Health and Telemedicine Services using a healthcare platform server.

Daily activity recognition system for the elderly using pressure sensors

As transformed to aging society rapidly, the number of old persons who live alone is drastically increased. Because these old people may have disorder of bodily function, and be suffering from geriatric disease, needs of a health assistance system to make them healthier are strongly increased. In this paper, we propose a daily activity recognition system for an old person using pressure sensors. The target daily activities are MEAL, SLEEP, EXCRETION, GO-OUT, and REST. The proposed system installs pressure sensors to furniture and floors in home, and recognizes daily activities based on the object usage information. By using the proposed system, we can provide a warning sign for unhealthy cases such as skipping meals. And, the families who live in remote place can check that their parent takes a healthy daily living.

ISO/IEEE 11073 PHD standardization of weighting scale using Nintendo's Wii Balance board™ for healthcare services

As transformed to aging society, many personal devices for healthcare have recently been developed. According to the increase of individual development of health devices, the ISO/IEEE 11073 PHD standard is established for gathering health measurement data. But many existing healthcare devices are not standardized. In this paper, we propose implementation model of standardization for existing healthcare devices which do not follow 11073 PHD standard. We use Nintendos Wii Balance board as weighting scale and develop virtual weight scale agent for 11073 PHD standardization. It communicates 11073 gateway (manager) with 11073 PHD standard protocols.

Normalization of Gene/Protein Names in Biological Literatures using Vector-Space Model

As the number of biological literatures grows exponentially, needs for text mining system are increased. In text mining area, normalization is mapping gene/protein names to a database. It is necessary to combine extracted information from various literatures and to curate a database or an ontology using literatures. Previous normalization researches used direct comparison methods between a database and literatures, but it is weak to extremely variational gene/protein names in literatures. Therefore, in this paper, we propose a normalization method using vector-space model. For each gene/protein name, we rank identifiers using vector-space model, and find the most similar identifier with the name. Experimental result shows the proposed method has 70.7% f-measure.

Extracting protein-protein interactions in biomedical literature using an existing syntactic parser

We are developing an information extraction system for life science literature. We are currently focusing on PubMed abstracts and trying to extract named entities and their relationships, especially protein names and protein-protein interactions. We are adopting methods including natural language processing, machine learning, and text processing. But we are not developing a new tagging or parsing technique. Developing a new tagger or a new parser specialized in life science literature is a very complex job. And it is not easy to get a good result by tuning an existing parser or by training it without a sufficient corpus. These all are another research topics and we are trying to extract information, not to develop something to help the extracting job or else. In this paper, we introduce our method to use an existing full parser without training or tuning. After tagging sentences and extracting proteins, we make sentences simple by substituting some words like named entities, nouns into one word. Then parsing errors are reduced and parsing precision is increased by this sentence simplification. We parse the simplified sentences syntactically with an existing syntactic parser and extract protein-protein interactions from its results. We show the effects of sentence simplification and syntactic parsing.

Statistical analysis service of e-healthcare record on iPad system

As society is aging, many personal devices for healthcare have recently been developed. In response to the increase in the individual development of health devices, the many standard has been established for the gathering of health measurement data. And, a system which synthetically manages and analyzes personal health record gathered from devices is required. In this paper, we propose statistical analysis service of e-healthcare record which collect personal health record based on ISO/IEEE 11073 PHD standard. The gathered health record is transferred to database server through secure internet channel. In the database server, collecting data are stored in the medical record database and processed in statistical analysis in order to support decision making for doctor and researcher and use various healthcare services such as chronic disease management, aging independently, and etc.

BioProber: software system for biomedical relation discovery from PubMed.

The numbers of articles and journals that are published are increasing at a considerable rate, and the published information is growing continuously and fast. Because of this, researches to acquire knowledge automatically have been carried out in the areas of information retrieval, information extraction and text mining. Information retrieval approaches are good for specific topics that the number of related articles is small. But, if the number is bigger, searching skill and knowledge acquisition ability are useless. Though many efforts have been made to extract information from literature, many approaches have concentrated on specific entities, such as proteins, genes and their interactions, and much information is still remained in unstructured text. So, we have developed a system that discovers relations between various categories of biomedical entities. Our system collects abstracts from PubMed by queries representing a topic and visualizes relationship from the collection by automatic information extraction

BioProber2.0: a unified biomedical workbench with mining and probing literatures

The number of biomedical literatures is increasing at a considerable rate, and the information is growing continuously and fast as well. Accordingly, information retrieval is more and more important to support biomedical researches. However, it often retrieves too many literatures or too few literatures for the target gene/proteins or relations. And extremely various synonyms of the gene and protein names make information retrieval more difficult to support biomedical researches. To overcome these difficulties, we propose a unified biomedical workbench with mining and probing literatures. The proposed workbench is composed of searching/collecting, literature mining, relation probing, and statistics analysis. It provides searching and collecting literatures of Pubmed articles and USPTO patents. And, to extract biomedical relations, the collected literatures are mined using text mining techniques such as named entity recognition, gene/protein name normalization, and relation extraction. Users can probe their target relations using these extracted relation information, shown in form of relation network. Finally, the workbench provides statistics information of literature meta data such as authors, organizations, publication years and so on. That is, the proposed workbench provides unified literature-based functions from searching to probing, and including text mining and statistics analysis.