Related Experiment Video
Updated: Aug 6, 2026
11:18
Generation of Comprehensive Thoracic Oncology Database - Tool for Translational Research
Published on: January 22, 2011
Generating Patient Documents from Electronic Health Records Using Generative Artificial Intelligence: A Feasibility
Nobuaki Michihata1, Hiroshi Ishii2, Hideki Tsujimura3
1Cancer Prevention Center, Chiba Cancer Center Research Institute, Japan, Chiba.
Applied Clinical Informatics
|July 20, 2026
Summary
Generative artificial intelligence (AI) shows potential for drafting clinical documents in Japanese oncology settings. While not consistently meeting draft utility thresholds, AI demonstrated feasibility and improved referral document generation from electronic health record (EHR) data.
Area of Science:
- Oncology
- Medical Informatics
- Artificial Intelligence
Background:
- Clinical documentation significantly burdens healthcare professionals, impacting patient care.
- Generative artificial intelligence (AI) offers a potential solution for automating tasks like drafting discharge summaries and referral documents.
- Evaluating AI feasibility in non-Western language oncology settings with real-world electronic health record (EHR) data is crucial but underexplored.
Purpose of the Study:
- To assess the feasibility of using an enterprise AI system for drafting clinical documents in a Japanese cancer hospital.
- To evaluate the practical utility of AI-generated discharge summaries and referral documents using real-world EHR data.
Main Methods:
- Analysis of de-identified EHR data from 61 adult cancer patients at Chiba Cancer Center.
- Limited data input included physician notes and nursing records; structured data and reports were excluded.
- An enterprise AI system (GaiXer) generated 31 discharge summaries and 30 referral documents, followed by a refinement cycle based on evaluator feedback.
Main Results:
- AI-generated documents scored approximately 60-70 out of 100.
- Nine of 31 discharge summaries and five of 30 referral documents (after refinement) met the exploratory threshold of ≥80/100.
- Referral document scores significantly improved after refinement (p=0.006), while discharge summary scores did not.
Conclusions:
- AI shows feasibility for drafting clinical documents in Japanese oncology using real-world EHR data within a secure environment.
- Generated documents did not consistently achieve draft-level utility, indicating areas for improvement.
- Future research should focus on larger datasets, multi-institutional studies, and evaluating clinician acceptance and workflow impact.
Related Concept Videos
01:29Issues And Trends In Healthcare Delivery System
The issues and trends in healthcare delivery are constantly changing. The COVID-19 pandemic is one recent issue that wreaked havoc on healthcare systems, causing a shortage of healthcare workers, high demand for medicines and supplies, and increased medical expenditure due to a lack of insurance. Other issues include rising healthcare costs and care fragmentation.
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...
01:30Methods of Documentation VII: EMR
Electronic Medical Records (EMRs) primarily center around electronically documenting patients' health information within a single healthcare organization or practice. They contain essential clinical data related to a patient's medical history, diagnoses, medications, treatment plans, lab results, and other pertinent information relevant to the specific encounter or episode of care. EMRs are designed to streamline documentation and workflow processes within individual healthcare settings,...