- AI
- Molecular Imaging
- CT
- X-Ray
- Ultrasound
- MRI
- Facility Management
- Mammography
System mines narrative reports for trends
A system that mines radiology reports for trends and findings was displayed during a scientific session Sunday.
A system that mines radiology reports for trends and findings was displayed during a scientific session Sunday.
The LEXIMER (for Lexicon Mediated Entropy Reduction) was making its second appearance at an RSNA meeting. This year, authors from Massachusetts General Hospital showed how the system could isolate reports with recommendations for additional action from those that had no recommendations and then identify patterns associated with those recommendations.
The study found, for example, that CT chest scans generated recommendations at rates ranging from 5% to 31%. Analyzing patterns over the course of seven years, the data miners found that recommendation rates for all CT scans grew from 11% in 1995 to 20% in 2002.
More than 3 million unstructured radiology reports were included in the database. The overall recommendation rate was 8.2%.
The study also pulled out of the seven years of reports the recommendation rates for other modalities:
- MR: 5% to 13%
- radiography: 4% to 7%
- fluoroscopy: 3% to 4%
- mammography: 5% to 12%
- nuclear medicine: 4% to 5%
- PET: 3% to 28%
- angiography: 11% to 18%
For individual radiologists, recommendation rates ranged from 3% to 15%.
Efforts now under way will expand data mining capabilities, said chief presenter Dr. Mannudeep K. Kalra. Researchers are looking for ways to merge clinical information, such as lab reports, with data mined from the radiology reports.
The system could also be made interactive. A report with a pattern contrary to the prevailing approach, for example, might be flagged for a second look by the radiologist, he said.
AI Facilitates Nearly 83 Percent Improvement in Turnaround Time for Fracture X-Rays
December 19th 2023In addition to offering a 98.5 percent sensitivity rate in diagnosing fractures on X-ray, an emerging artificial intelligence (AI) software reportedly helped reduce mean turnaround time on X-ray fracture diagnosis from 48 hours to 8.3 hours, according to new research presented at the Radiological Society of North America (RSNA) conference.
The Reading Room: Artificial Intelligence: What RSNA 2020 Offered, and What 2021 Could Bring
December 5th 2020Nina Kottler, M.D., chief medical officer of AI at Radiology Partners, discusses, during RSNA 2020, what new developments the annual meeting provided about these technologies, sessions to access, and what to expect in the coming year.
Can an Emerging PET Radiotracer Enhance Detection of Prostate Cancer Recurrence?
December 14th 2023The use of 68Ga-RM2 PET/MRI demonstrated a 35 percent higher sensitivity rate than MRI alone for the diagnosis of biochemical recurrence of prostate cancer, according to research recently presented at the Radiological Society of North America (RSNA) conference.
RSNA 2020: Addressing Healthcare Disparities and Access to Care
December 4th 2020Rich Heller, M.D., with Radiology Partners, and Lucy Spalluto, M.D., with Vanderbilt University School of Medicine, discuss the highlights of their RSNA 2020 session on health disparities, focusing on the underlying factors and challenges radiologists face to providing greater access to care.
Can AI Improve Detection of Extraprostatic Extension on MRI?
December 4th 2023Utilizing a deep learning-based AI algorithm to differentiate between diagnostic and non-diagnostic quality of prostate MRI facilitated a 10 percent higher specificity rate for diagnosing extraprostatic extension on multiparametric MRI, according to research presented at the recent RSNA conference.
Study: Regular Mammography Screening Reduces Breast Cancer Mortality Risk by More than 70 Percent
November 30th 2023Consistent adherence to the five most recent mammography screenings prior to a breast cancer diagnosis reduced breast cancer death risk by 72 percent in comparison to women who did not have the mammography screening, according to new research findings presented at the annual Radiological Society of North America (RSNA) conference.
