FRONTEO's joint research paper on fall prediction system has been accepted by the international medical journal JMIR Medical Informatics

Home » corporate » News » 2020 » FRONTEO's joint research paper on fall prediction system has been accepted by the international medical journal JMIR Medical Informatics
2020.01.29 Press release

--To the press -

FRONTEO's joint research paper on fall prediction system has been accepted by the international medical journal JMIR Medical Informatics

 FRONTEO Inc. (Headquarters: Minato-ku, Tokyo, President: Masahiro Morimoto) conducted joint research at the Life Science AI Business Headquarters.Fall fall prediction systemHas been accepted by the international medical journal "JMIR (Journal of Medical Internet Research) Medical Informatics".

 

 JMIR Medical Informatics is one of the internationally reviewed open-access medical journals published by JMIR Publications, focusing on clinical computer science, health and healthcare big data, electronic medical records, and medical infrastructure. The impact factor (literature citation rate) of 1 in the journal evaluation analysis tool Journal Citation Reports is 2018, which is the top journal in 3.188 journals in the category "Medical Informatics".

 

The paper accepted this time is a summary of joint research by the following members.

Title: "Prediction of inpatient falls using natural language processing for nursing records on electronic medical records"

 

Responsible author: 
Hidefumi Uchiyama FRONTEO Inc. Life Science AI Business Headquarters Research and Development Department Scienty
     Fick Fellow, Doctor of Engineering1

 

Co-author:
Hayao Nakatani, Director of Chigasaki Shinbokuryo Hospital, Doctor of Medicine
Masatoshi Nakao Chief Nursing, Education Support Development Department, Nursing Department, NTT Medical Center Tokyo
Hiroyoshi Toyoshiba FRONTEO Inc. Life Science AI Business Headquarters Life Science AI CTO
      Director of Research and Development Department, Doctor of Mathematics
Yoshiyuki Ochiai, Honorary Director of NTT Medical Center Tokyo, Academic Advisor, Tokyo Health Care University, Doctor of Medicine

 

 In the study, we extracted the nursing records of 12 fallers and 335 non-fallers over a 408-month period and divided them into a learning data set and test data to be predicted, so that the learning data set could be used for falls. At the same time as building a model for prediction, we extracted morphological elements that affect the classification of fallers and non-fallers.In conclusion, the receiver operating characteristic (ROC) is used to determine the model predictions for the test data.

 

 The results of the study showed that the fall prediction by the Concept Encoder was good and showed high accuracy in the region below the ROC curve.In addition, the morphemes incorporated into the final model include words such as state of consciousness and mobility, and by combining natural language processing algorithms related to known risk factors for falls and Concept Encoder, the fall We have demonstrated that risk factors can be extracted more effectively.

 

 From this research, we believe that we were able to establish the usefulness of developing a new system that makes it possible to predict falls through analysis with FRONTEO's artificial intelligence Concept Encoder from text information such as nursing records.

 

 In 2015, FRONTEO started joint research with NTT Medical Center Tokyo on a fall prediction system. In September 2019, together with Eisai Co., Ltd., the fall / fall prediction system `` Coroban'' that predicts the fall / fall risk of inpatients in advance and displays an alert.®We are promoting the introduction toward the realization of medical safety in hospitals.

 

* 1: The title is at the time of writing.Currently belongs to Scientist, Pharmaceutical Research Department, Neopharma Japan Co., Ltd.

 

concept encoderabout URL: https://www.fronteo.com/products/conceptencoder/
Concept Encoder is an artificial intelligence (AI) developed by FRONTEO specifically for the healthcare industry.It was developed in 2018 with the aim of effectively analyzing and utilizing big data related to healthcare, which includes a large amount of free-form text data, based on evidence.We have introduced and realized statistical methods such as the significance test, which is indispensable for "evidence-based medicine (EBM)", which is a common understanding among healthcare professionals, in natural language analysis. Concept Encoder can also be co-analyzed with data other than text, and we are conducting research on co-analysis with "numerical data" such as gene expression information, vitals and various test values ​​accumulated in the healthcare field.Patent registration number: Patent No. XNUMX

 

FRONTEOabout URL: https://www.fronteo.com/
FRONTEO Inc. has developed its own artificial intelligence engine "KIBIT".®(Kibit) ”and“ concept encoder®(Concept encoder) ”is a data analysis company that supports information analysis such as big data.Established in August 2003 as a company that supports e-discovery (electronic discovery), which preserves, investigates, and analyzes electronic data necessary for international litigation, and digital forensic investigations.In-house developed data analysis platform "Lit i View"®(Lit Eye View) ”,“ Predictive Coding Predictive Coding®(Predictive Coding) ”technology is used to provide legal proceedings support to companies.The unique artificial intelligence-related technology cultivated and developed in this legal business learns "tacit knowledge" such as expert experience and intuition, and realizes prediction of future behavior from analysis of human thoughts.We have expanded into areas such as life science and business intelligence, and have achieved results in "work style reform" in addition to FinTech and RegTech. Listed on TSE Mothers on June 2007, 6 and NASDAQ on May 26, 2013.The capital is 5 thousand yen (as of March 16, 2,559,206). As of July 2019, 3, the company name has been changed from UBIC Co., Ltd. to the current company name.

 

<Inquiries about Life Science AI Business>
Life Science AI Business Department, FRONTEO, Inc.
TEL 03-5463-6330 FAX 03-5463-7578 Email: fhc_contact@fronteo.com

 

<Contact information for the press>
Public Relations Officer, FRONTEO Inc. Ikeuchi
TEL: 03-5463-6380 FAX: 03-5463-6345 Email: pr_contact@fronteo.com