Regional Selectivity of Neuromelanin Changes in the Substantia Nigra in Atypical Parkinsonism.

BACKGROUND: Neurodegeneration in the substantia nigra pars compacta (SNc) in parkinsonian syndromes may affect the nigral territories differently. OBJECTIVE: The objective of this study was to investigate the regional selectivity of neurodegenerative changes in the SNc in patients with Parkinson's disease (PD) and atypical parkinsonism using neuromelanin-sensitive magnetic resonance imaging (MRI). METHODS: A total of 22 healthy controls (HC), 38 patients with PD, 22 patients with progressive supranuclear palsy (PSP), 20 patients with multiple system atrophy (MSA, 13 with the parkinsonian variant, 7 with the cerebellar variant), 7 patients with dementia with Lewy body (DLB), and 4 patients with corticobasal syndrome were analyzed. volume and signal-to-noise ratio (SNR) values of the SNc were derived from neuromelanin-sensitive MRI in the whole SNc. Analysis of signal changes was performed in the sensorimotor, associative, and limbic territories of the SNc. RESULTS: SNc volume and corrected volume were significantly reduced in PD, PSP, and MSA versus HC. Patients with PSP had lower volume, corrected volume, SNR, and contrast-to-noise ratio than HC and patients with PD and MSA. Patients with PSP had greater SNR reduction in the associative region than HC and patients with PD and MSA. Patients with PD had reduced SNR in the sensorimotor territory, unlike patients with PSP. Patients with MSA did not differ from patients with PD. CONCLUSIONS: This study provides the first MRI comparison of the topography of neuromelanin changes in parkinsonism. The spatial pattern of changes differed between PSP and synucleinopathies. These nigral topographical differences are consistent with the topography of the extranigral involvement in parkinsonian syndromes. © 2022 International Parkinson and Movement Disorder Society.

Comparative performances of machine learning algorithms in radiomics and impacting factors.

There are no current recommendations on which machine learning (ML) algorithms should be used in radiomics. The objective was to compare performances of ML algorithms in radiomics when applied to different clinical questions to determine whether some strategies could give the best and most stable performances regardless of datasets. This study compares the performances of nine feature selection algorithms combined with fourteen binary classification algorithms on ten datasets. These datasets included radiomics features and clinical diagnosis for binary clinical classifications including COVID-19 pneumonia or sarcopenia on CT, head and neck, orbital or uterine lesions on MRI. For each dataset, a train-test split was created. Each of the 126 (9 × 14) combinations of feature selection algorithms and classification algorithms was trained and tuned using a ten-fold cross validation, then AUC was computed. This procedure was repeated three times per dataset. Best overall performances were obtained with JMI and JMIM as feature selection algorithms and random forest and linear regression models as classification algorithms. The choice of the classification algorithm was the factor explaining most of the performance variation (10% of total variance). The choice of the feature selection algorithm explained only 2% of variation, while the train-test split explained 9%.

Quantitative biomarkers allow the diagnosis of head and neck paraganglioma on multiparametric MRI.

PURPOSE: The aim of this study is to identify quantitative MR biomarkers in head and neck paragangliomas. METHODS: The study was approved by an institutional review board. A retrospective review of patients with head and neck paragangliomas (HNPGL) evaluated by time-resolved MRA sequences between 2009 and 2019 was performed. A control group investigated during the same period was analyzed, including nerve sheath tumors and metastatic lymph nodes from squamous cell carcinomas or undifferentiated nasopharyngeal cancer (UCNT). A gold standard was obtained for all cases. Semi-quantitative parameters of enhancement were extracted from time-intensity curves on time-resolved MRA sequences and diffusion weighted imaging/DWI was assessed for each lesion. RESULTS: Sixty head and neck paragangliomas (HNPGLs) were included from 50 patients. The control group consisted of 30 parapharyngeal space lesions (27 patients), which included nerve sheath tumors (n = 12) and metastatic lymph nodes (n = 18) from squamous cell carcinomas or UCNT. PGLs showed a shorter time-to-peak value compared to other groups, measured at 25.0 +/- 29 sec. The wash-in and wash-out ratios were also significantly higher for PGLs, respectively measured at 5.34 ± 2.99 (p < 0,001) and 1.24 ± 0.80 (p < 0.001). On DWI sequences, the mean ADC value for PGLs (1.17 ± 0.19 10^-3 mm2/s) was significantly different than the other tumor groups (p < 0.001). HNPGLs were clearly distinguishable from other tumors on classification with regression tree based on TTP and ADC values. These distinct group features were also consistent on principal component analysis. CONCLUSION: Our study identifies a multiparametric signature for disease subtyping, providing a strong impetus for switching from qualitative to quantitative analysis of deep soft-tissue tumors of the neck.

Facial Nerve Canal Paraganglioma.

We report the case of a 72-year-old woman presenting with a progressive left peripheral facial paralysis and a facial canal mass extending through the stylomastoid foramen along the posterior edge of the parotid gland. On MRI, the early and intense enhancement was highly suggestive of paraganglioma but could not rule out a nonossifying hemangioma. Laboratory analysis showed normal plasma metanephrines. On F-FDOPA PET/CT, the mass exhibited a typical paraganglioma feature with a marked tumor uptake. Our case demonstrates that F-FDOPA plays a vital role in this rare entity and can avoid any further confirmatory invasive procedure.