Parkinson Hastalığında Asimetrik Dejenerasyonu Karakterize Etmek için Nigrozom-1'in 3.0T Duyarlılık Ağırlıklı Görüntüleme Üzerine Radyomik Analizi

Araştırma özeti

Aşağıdaki metin kaynak makalenin lisansı doğrulanmış orijinal özetidir. cc by

Abstract Background Nigrosome-1 (N1) is one of the earliest regions affected in Parkinson's disease (PD) and can be visualized on 3.0-T susceptibility-weighted imaging (SWI) as the swallow tail sign (STS). This study investigated the associations of visual STS assessment and N1-derived radiomic features with clinical characteristics in PD. Methods In this retrospective study, 83 patients with PD who underwent 3.0-T SWI and clinical evaluation were included. Patients were categorized according to STS visualization as having total absence or non-total absence of dorsolateral nigral hyperintensity. The N1 region was manually segmented on SWI, and 107 radiomic features were extracted. Least bsolute shrinkage and selection operator (LASSO) regression was applied for radiomic feature selection. Based on the selected features, three machine learning classifiers—LASSO-logistic regression (LASSO LR), the random forest, and a linear support vector machine (linear SVM)—were constructed and compared. Results STS visualization was significantly associated with clinical severity indicators, including Hoehn-Yahr (H-Y) stage, MDS-UPDRS II score, rigidity score, and plasma iron level (P 

İlgili konular