首页|New Findings from University of Wisconsin Update Understanding of Artificial Int elligence (Automated Deep Learning Artificial Intelligence Tool for Spleen Segme ntation On Ct: Defining Volume-based Thresholds for Splenomegaly)
New Findings from University of Wisconsin Update Understanding of Artificial Int elligence (Automated Deep Learning Artificial Intelligence Tool for Spleen Segme ntation On Ct: Defining Volume-based Thresholds for Splenomegaly)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Researchers detail new data in Artific ial Intelligence. According to news reporting from Madison, Wisconsin, by NewsRx journalists, research stated, "Splenomegaly historically has been assessed on i maging by use of potentially inaccurate linear measurements. Prior work tested a deep learning artificial intelligence (AI) tool that automatically segments the spleen to determine splenic volume." Financial support for this research came from National Institutes of Health (NIH ) - USA. The news correspondents obtained a quote from the research from the University o f Wisconsin, "The purpose of this study is to apply the deep learning AI tool in a large screening population to establish volume-based splenomegaly thresholds. This retrospective study included a primary (screening) sample of 8901 patients (4235 men, 4666 women; mean age, 56 +/- 10 [SD] years) who underwent CT colonoscopy (n = 7736) or renal donor CT (n = 1165) from April 2004 to January 2017 and a secondary sample of 104 patients (62 men, 42 w omen; mean age, 56 +/- 8 years) with end-stage liver disease who underwent contr ast-enhanced CT performed as part of evaluation for potential liver transplant f rom January 2011 to May 2013. The automated deep learning AI tool was used for s pleen segmentation, to determine splenic volumes. Two radiologists independently reviewed a subset of segmentations. Weight-based volume thresholds for splenome galy were derived using regression analysis. Performance of linear measurements was assessed. Frequency of splenomegaly in the secondary sample was determined u sing weight-based volumetric thresholds. In the primary sample, both observers c onfirmed splenectomy in 20 patients with an automated splenic volume of 0 mL; co nfirmed incomplete splenic coverage in 28 patients with a tool output error; and confirmed adequate segmentation in 21 patients with low volume (<50 mL), 49 patients with high volume (> 600 mL), and 20 0 additional randomly selected patients. In 8853 patients included in analysis o f splenic volumes (i.e., excluding a value of 0 mL or error values), the mean au tomated splenic volume was 216 +/- 100 [SD] mL. The weight-based volumetric threshold (expressed in milliliters) for splenom egaly was calculated as (3.01 x weight [expressed as kilogram s]) + 127; for weight greater than 125 kg, the splenomegaly t hreshold was constant (503 mL). Sensitivity and specificity for volume-defined s plenomegaly were 13% and 100%, respectively, at a tru e craniocaudal length of 13 cm, and 78% and 88% for a maximum 3D length of 13 cm. In the secondary sample, both observers identified segmentation failure in one patient. The mean automated splenic volume in the 1 03 remaining patients was 796 +/- 457 mL; 84% (87/103) of patients met the weight-based volume-defined splenomegaly threshold. We derived a weight -based volumetric threshold for splenomegaly using an automated AI-based tool. C LINICAL IMPACT."
MadisonWisconsinUnited StatesNorth and Central AmericaArtificial IntelligenceEmerging TechnologiesHealth and MedicineHemic and Immune SystemsLymphoid TissueMachine LearningSpleenSplenomegalyUniversity of Wisconsin
NETL
NSTL
