首页|Icahn School of Medicine at Mount Sinai Reports Findings inCOVID-19 (An assessm ent of PET and CMR radiomic featuresfor the detection of cardiac sarcoidosis)
Icahn School of Medicine at Mount Sinai Reports Findings inCOVID-19 (An assessm ent of PET and CMR radiomic featuresfor the detection of cardiac sarcoidosis)
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 DailyNews Daily News – New research on Coronavirus - COVID-19 is the subject of a report. According to newsreporting out of New York City, N ew York, by NewsRx editors, research stated, “Visual interpretationof PET and C MR may fail to identify cardiac sarcoidosis (CS) with high specificity. This stu dy aimedto evaluate the role of [F]FDG PE T and late gadolinium enhancement (LGE)-CMR radiomic features indifferentiating CS from another cause of myocardial inflammation, in this case patients with ca rdiac-relatedclinical symptoms following COVID-19.”Our news journalists obtained a quote from the research from the Icahn School of Medicine at MountSinai, “[F]FDG PET and LGE-CMR were treated separately in this work. There were 35 post-COVID-19(PC) a nd 40 CS datasets. Regions of interest were delineated manually around the entir e left ventricle forthe PET and LGE-CMR datasets. Radiomic features were then e xtracted. The ability of individual featuresto correctly identify image data as CS or PC was tested to predict the clinical classification of CS vs. PCusing M ann-Whitney -tests and logistic regression. Features were retained if the -value was <0.00053, theAUC was >0.5, an d the accuracy was >0.7. After applying the correlation test, uncorrelated features wereused as a signature (joint features) to train m achine learning classifiers. For LGE-CMR analysis, to furtherimprove the result s, different classifiers were used for individual features besides logistic regr ession, and theresults of individual features of each classifier were screened to create a signature that included all featuresthat followed the previously me ntioned criteria and used it them as input for machine learning classifiers.The Mann-Whitney -tests and logistic regression were trained on individual features to build a collectionof features. For [F] FDG PET analysis, the maximum target-to-background ratio ( ) showed a high areaunder the curve (AUC) and accuracy with small -values (<0.0 0053), but the signature performed better(AUC 0.98 and accuracy 0.91). For LGE- CMR analysis, the showed good results with small error bars(accuracy 0.75 and A UC 0.87). However, by applying a Support Vector Machine classifier to individualLGE-CMR features and creating a signature, a Random Forest classifier displayed better AUC and accuracy(0.91 and 0.84, respectively). Using radiomic features may prove useful in identifying individuals with CS.Some features showed promis ing results in differentiating between PC and CS.”
New York CityNew YorkUnited StatesNorth and CentralAmericaCOVID-19CardiologyCoronavirusCyborgsDermatolo gyEmerging TechnologiesHealth andMedicineLymphatic Diseases and Condition sLymphoproliferative Diseases and ConditionsLymphoproliferativeDisordersM achine LearningRNA VirusesSARS-CoV-2SarcoidosisSevere Acute RespiratorySyndrome Coronavirus 2ViralVirology
NETL
NSTL
