Examples of Specific Causes Identified for AI Tool Failure.

Examples of Specific Causes Identified for AI Tool Failure. Axial images at L1 level, without (left) and with (right) segmentation overlay. Red indicates skeletal muscle, green indicates trabecular bone, yellow indicates visceral fat, and blue indicates subcutaneous fat. Segmented regions also include liver (beige) and spleen (orange), which were not evaluated as part of present analysis. (A) 78-year-old woman who underwent abdominopelvic CT at outside institution. Bone tool returned L1 vertebral body bone attenuation of -146 HU, outside of reference range. Thus, tool was deemed technical failure for bone tool. Failure was attributed to volume averaging of vacuum phenomenon within slice. (B) 64-year-old woman who underwent abdominopelvic CT at outside institution. Bone tool returned vertebral body bone attenuation of -10,000 HU (default value for segmentation failure detected by tool), outside of reference range. Thus, tool was deemed technical failure for bone tool. Failure was attributed to presence of spinal fusion hardware. 

 


February 22, 2023 — According to an accepted manuscript published in ARRS’ American Journal of Roentgenology (AJR), certain reasons for AI tool failure relating to technical factors may be largely preventable through proper acquisition and reconstruction protocols. 

“The automated AI body composition tools had high technical adequacy rates in a heterogeneous sample of external CT examinations, supporting the tools’ generalizability and potential for broad use,” concluded head researcher B. Dustin Pooler, MD, from the University of Wisconsin School of Medicine & Public Health in Madison. 

This AJR accepted manuscript included 8,949 patients (mean age, 55.5 years; 4,256 men, 4,693 women) who underwent abdominal CT—performed at different institutions on different scanners from different manufacturers—subsequently transferred to the local PACS for clinical purposes. Deploying three independent automated AI tools to assess body composition via bone attenuation, muscle amount and attenuation, as well as visceral and subcutaneous fat amounts, one axial series per examination was evaluated. 

Ultimately, three fully automated AI tools for measuring body composition (vertebral bone, body wall musculature, and visceral and subcutaneous abdominal fat) had technical adequacy rates of 97.8%- 99.1% on Pooler et al.’s sample of 11,699 external abdominal CT examinations—performed at 777 different external institutions using 82 different scanner models from 6 different manufacturers.  

Noting that reasons for failure also included factors inherent to patients that are more challenging to control, “explainability and an understanding of reasons for failure can help build trust in AI tools and increase acceptance among radiologists and other physicians,” the authors of this AJR accepted manuscript added

For more information: www.arrs.org 


Related Content

News | Enterprise Imaging

December 6, 2023 — NVIDIA launched a cloud service for medical imaging AI to further streamline and accelerate the ...

Time December 06, 2023
arrow
News | Ultrasound Imaging

December 6, 2023 — Clarius Mobile Health has been awarded a Technology Breakthrough Designation in the Ultrasound ...

Time December 06, 2023
arrow
News | Digital Pathology

December 6, 2023 — One of the largest independent pathology service providers in the United States, CellNetix Pathology ...

Time December 06, 2023
arrow
News | Artificial Intelligence

December 5, 2023 — Radiology is under severe economic pressure as reimbursement rates fall for many common scans. At the ...

Time December 05, 2023
arrow
News | RSNA

December 1, 2023 — Royal Philips, a global leader in health technology, put patients and healthcare providers center ...

Time December 01, 2023
arrow
News | RSNA

December 1, 2023 —Nuance Communications, Inc., a Microsoft Company, during the 2023 Radiological Society of North ...

Time December 01, 2023
arrow
News | RSNA

December 1, 2023 — In today's healthcare environment, the industry faces various pressing challenges. The magnetic ...

Time December 01, 2023
arrow
News | Radiology Imaging

November 30, 2023 — Through its leading imaging solutions and commitment to innovation, GE HealthCare remains at the ...

Time November 30, 2023
arrow
News | RSNA

November 30, 2023 — Hologic, Inc. showcased several recent developments in next-generation AI solutions at the 109th ...

Time November 30, 2023
arrow
News | RSNA

November 28, 2023 — Curtis P. Langlotz, MD, PhD, has taken the helm as 2023-2024 president of the Radiological Society ...

Time November 29, 2023
arrow
Subscribe Now