Retrieval technology of data on healthcare based on a formal language to represent a change in the frequency of an event

2018 Fiscal Year Final Research Report

• Project/Area Number: 15K00297
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Gunma University
• Principal Investigator: Takaki Osamu 群馬大学, 社会情報学部, 准教授 (30388011)
• Project Period (FY): 2015-04-01 – 2019-03-31
• Keywords: 頻度 / 時相論理 / 医療データ / 時系列データ / ケース・ファインディング

Outline of Final Research Achievements

This research project established a temporal logic that consists of a formal language and semantics to represent a change in the frequency of an event such as “surgery” or “administration of a medicine.” This logic is called the “frequency logic.” This temporal logic enables to determine if a given time series data satisfies the pattern of a frequency change of the event represented in this formal language. Moreover, a system based on the frequency logic has been developed to retrieve all parts of a given time series data that satisfy the pattern of a frequency change of the given event.

Free Research Field

医療情報学

Academic Significance and Societal Importance of the Research Achievements

頻度論理は，従来の時相論理では表現することが困難な，事象発生頻度の変化に関する命題を自然かつ比較的簡潔に表現することを可能にしている．また，特定の業務に関する履歴データ等の時系列データから，問題となりそうな事象の発生頻度パターンを満たす箇所を網羅的に洗い出すことによって，業務上の潜在的なリスクを洗い出し，業務上の安全性や効率性の改善を促進する働きを持っている．