Software tool helps German radiologists find database errors

Article

German researchers have devised an automated system to rid their hospital database of administrative errors. The prototype detection tool is expected to boost efficiency by weeding out phantom patients and duplicate records.Simple database

German researchers have devised an automated system to rid their hospital database of administrative errors. The prototype detection tool is expected to boost efficiency by weeding out phantom patients and duplicate records.

Simple database inconsistencies can affect the level of service provided by digital radiology departments, according to Dr. Sven Nissen-Meyer at the Institute for Clinical Radiology, University Hospital Munich-Grosshadern, where the detection tool is undergoing a trial.

Spelling errors, mistyped numbers, and confusion between different forms of patient ID were making it difficult for radiologists to match images and reports to the correct patient record, he told delegates at last month's Computer Assisted Radiology and Surgery meeting in London. Data anomalies were also hampering efforts to retrieve previous exam details from the database archive, he added.

All patients admitted to the Munich hospital receive a unique 10-digit patient number (patient ID) and 10-digit admission or visit number (admission ID) from the central HIS. This information is transferred automatically to the RIS within four minutes of patient admission. Errors can arise if former patients are readmitted under a different name or if staff entering data manually into the RIS misspell names, inadvertently switch pairs of numbers, or confuse the two 10-digit identifiers, Nissen-Meyer said.

"So we have multiple patient records with the same patient ID number, and patient IDs with the wrong name, and vice versa," he said. "We also have multiple HIS admission records with the same admission ID."

Team members constructed a program capable of comparing each trio of patient identifiers (name, patient ID, admission ID) on the RIS-PACS with those on the HIS. Any inconsistencies were flagged, and the erroneous "triples" were saved to a separate database.

After running the program on more than 600,000 admission records and almost 300,000 patient records, they discovered that 6% of the patient IDs did not appear in the HIS database. A total of 2% of patient names and 8% of admission IDs were similarly invalid.

"This also meant that 1.1% of valid admission IDs were associated with invalid patient Ids, and 0.14% of correct patient IDs were incorrectly associated with an invalid admission ID," Nissen-Meyer said.

The error detection program is now in use on a daily basis at the Munich hospital, where it picks up an average of 50 to 60 mismatches per day. The next step should be the creation of a tool to correct obvious inconsistencies, such as cases where two of the patient identifiers in the RIS-PACS match those in the HIS, Nissen-Meyer said.

Recent Videos
Study: MRI-Based AI Enhances Detection of Seminal Vesicle Invasion in Prostate Cancer
What New Research Reveals About the Impact of AI and DBT Screening: An Interview with Manisha Bahl, MD
Can AI Assessment of Longitudinal MRI Scans Improve Prediction for Pediatric Glioma Recurrence?
A Closer Look at MRI-Guided Adaptive Radiotherapy for Monitoring and Treating Glioblastomas
Incorporating CT Colonography into Radiology Practice
What New Research Reveals About Computed Tomography and Radiation-Induced Cancer Risk
What New Interventional Radiology Research Reveals About Treatment for Breast Cancer Liver Metastases
New Mammography Studies Assess Image-Based AI Risk Models and Breast Arterial Calcification Detection
Can Deep Learning Provide a CT-Less Alternative for Attenuation Compensation with SPECT MPI?
Employing AI in Detecting Subdural Hematomas on Head CTs: An Interview with Jeremy Heit, MD, PhD
Related Content
© 2025 MJH Life Sciences

All rights reserved.