Radiomics relies on the extraction of a wide variety of quantitative image-based features to provide decision support. Magnetic resonance imaging (MRI) contributes to the personalization of patient ...care but suffers from being highly dependent on acquisition and reconstruction parameters. Today, there are no guidelines regarding the optimal pre-processing of MR images in the context of radiomics, which is crucial for the generalization of published image-based signatures. This study aims to assess the impact of three different intensity normalization methods (Nyul, WhiteStripe, Z-Score) typically used in MRI together with two methods for intensity discretization (fixed bin size and fixed bin number). The impact of these methods was evaluated on first- and second-order radiomics features extracted from brain MRI, establishing a unified methodology for future radiomics studies. Two independent MRI datasets were used. The first one (DATASET1) included 20 institutional patients with WHO grade II and III gliomas who underwent post-contrast 3D axial T1-weighted (T1w-gd) and axial T2-weighted fluid attenuation inversion recovery (T2w-flair) sequences on two different MR devices (1.5 T and 3.0 T) with a 1-month delay. Jensen-Shannon divergence was used to compare pairs of intensity histograms before and after normalization. The stability of first-order and second-order features across the two acquisitions was analysed using the concordance correlation coefficient and the intra-class correlation coefficient. The second dataset (DATASET2) was extracted from the public TCIA database and included 108 patients with WHO grade II and III gliomas and 135 patients with WHO grade IV glioblastomas. The impact of normalization and discretization methods was evaluated based on a tumour grade classification task (balanced accuracy measurement) using five well-established machine learning algorithms. Intensity normalization highly improved the robustness of first-order features and the performances of subsequent classification models. For the T1w-gd sequence, the mean balanced accuracy for tumour grade classification was increased from 0.67 (95% CI 0.61-0.73) to 0.82 (95% CI 0.79-0.84, P = .006), 0.79 (95% CI 0.76-0.82, P = .021) and 0.82 (95% CI 0.80-0.85, P = .005), respectively, using the Nyul, WhiteStripe and Z-Score normalization methods compared to no normalization. The relative discretization makes unnecessary the use of intensity normalization for the second-order radiomics features. Even if the bin number for the discretization had a small impact on classification performances, a good compromise was obtained using the 32 bins considering both T1w-gd and T2w-flair sequences. No significant improvements in classification performances were observed using feature selection. A standardized pre-processing pipeline is proposed for the use of radiomics in MRI of brain tumours. For models based on first- and second-order features, we recommend normalizing images with the Z-Score method and adopting an absolute discretization approach. For second-order feature-based signatures, relative discretization can be used without prior normalization. In both cases, 32 bins for discretization are recommended. This study may pave the way for the multicentric development and validation of MR-based radiomics biomarkers.
We report the clinical features of 3 patients in France who had parotitis (inflammation of the parotid salivary glands) as a clinical manifestation of confirmed coronavirus disease. Results from ...magnetic resonance imaging support the occurrence of intraparotid lymphadenitis, leading to a parotitis-like clinical picture.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, ODKLJ, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
BACKGROUND AND PURPOSE—Arterial wall enhancement on vessel wall MRI was described in intracranial inflammatory arterial disease. We hypothesized that circumferential aneurysmal wall enhancement ...(CAWE) could be an indirect marker of aneurysmal wall inflammation and, therefore, would be more frequent in unstable (ruptured, symptomatic, or undergoing morphological modification) than in stable (incidental and nonevolving) intracranial aneurysms.
METHODS—We prospectively performed vessel wall MRI in patients with stable or unstable intracranial aneurysms. Two readers independently had to determine whether a CAWE was present.
RESULTS—We included 87 patients harboring 108 aneurysms. Interreader and intrareader agreement for CAWE was excellent (κ=0.85; 95% confidence interval, 0.75–0.95 and κ=0.90; 95% confidence interval, 0.83–0.98, respectively). A CAWE was significantly more frequently seen in unstable than in stable aneurysms (27/31, 87% versus 22/77, 28.5%, respectively; P<0.0001). Multivariate logistic regression, including CAWE, size, location, multiplicity of aneurysms, and daily aspirin intake, revealed that CAWE was the only independent factor associated with unstable status (odds ratio, 9.20; 95% confidence interval, 2.92–29.0; P=0.0002).
CONCLUSIONS—CAWE was more frequently observed in unstable intracranial aneurysms and may be used as a surrogate of inflammatory activity in the aneurysmal wall.
BACKGROUND AND PURPOSE—Absence of arterial wall enhancement (AWE) of unruptured intracranial aneurysms (UIA) has shown promise at predicting which aneurysms will not rupture. We here tested the ...hypothesis that increased enhancement during follow-up (increased intensity, extension, or thickness or appearance of de novo enhancement), assessed using vessel wall magnetic resonance imaging, was associated with higher rates of subsequent growth.
METHODS—Patients with UIA were included between 2012 and 2018. Two readers independently rated AWE modification on 3T vessel wall magnetic resonance imaging, and morphological changes on time-of-flight magnetic resonance angiography during follow-up.
RESULTS—A total of 129 patients harboring 145 UIA (mean size 4.1 mm) met study criteria, of which 12 (8.3%) displayed morphological growth at 2 years. Of them, 8 demonstrated increased AWE during follow-up before or concurrently to morphological growth, and 4 had preexisting AWE that remained stable before growth. In the remaining 133 (nongrowing) UIAs, no AWE modifications were found. In multivariable analysis, increased AWE, not size, was associated with UIA growth (relative risk, 26.1 95% CI, 7.4–91.7, P<0.001). Sensitivity, specificity, positive predictive value, and negative predictive value for UIA growth of increased AWE during follow-up were, respectively, of 67%, 100%, 96%, and 100%.
CONCLUSIONS—Increased AWE during follow-up of conservatively managed UIAs predicts aneurysm growth over a 2-year period. This may impact UIA management towards closer monitoring or preventive treatment. Replication in a different setting is warranted.
Artificial intelligence (AI)-based tools are gradually blending into the clinical neuroradiology practice. Due to increasing complexity and diversity of such AI tools, it is not always obvious for ...the clinical neuroradiologist to capture the technical specifications of these applications, notably as commercial tools very rarely provide full details. The clinical neuroradiologist is thus confronted with the increasing dilemma to base clinical decisions on the output of AI tools without knowing in detail what is happening inside the “black box” of those AI applications. This dilemma is aggravated by the fact that currently, no established and generally accepted rules exist concerning best clinical practice and scientific and clinical validation nor for the medico-legal consequences in cases of wrong diagnoses. The current review article provides a practical checklist of essential points, intended to aid the user to identify and double-check necessary aspects, although we are aware that not all this information may be readily available at this stage, even for certified and commercially available AI tools. Furthermore, we therefore suggest that the developers of AI applications provide this information.
Functional-based resection under awake conditions had been associated with a nonnegligible rate of intraoperative and postoperative epileptic seizures. The authors assessed the incidence of ...intraoperative and early postoperative epileptic seizures after functional-based resection under awake conditions.
The authors prospectively assessed intraoperative and postoperative seizures (within 1 month) together with clinical, imaging, surgical, histopathological, and follow-up data for 202 consecutive diffuse glioma adult patients who underwent a functional-based resection under awake conditions.
Intraoperative seizures occurred in 3.5% of patients during cortical stimulation; all resolved without any procedure being discontinued. No predictor of intraoperative seizures was identified. Early postoperative seizures occurred in 7.9% of patients at a mean of 5.1 ± 2.9 days. They increased the duration of hospital stay (p = 0.018), did not impact the 6-month (median 95 vs 100, p = 0.740) or the 2-year (median 100 vs 100, p = 0.243) postoperative Karnofsky Performance Status score and did not impact the 6-month (100% vs 91.4%, p = 0.252) or the 2-year (91.7 vs 89.4%, p = 0.857) postoperative seizure control. The time to treatment of at least 3 months (adjusted OR aOR 4.76 95% CI 1.38-16.36, p = 0.013), frontal lobe involvement (aOR 4.88 95% CI 1.25-19.03, p = 0.023), current intensity for intraoperative mapping of at least 3 mA (aOR 4.11 95% CI 1.17-14.49, p = 0.028), and supratotal resection (aOR 6.24 95% CI 1.43-27.29, p = 0.015) were independently associated with early postoperative seizures.
Functional-based resection under awake conditions can be safely performed with a very low rate of intraoperative and early postoperative seizures and good 6-month and 2-year postoperative seizure outcomes. Intraoperatively, the use of the lowest current threshold producing reproducible responses is mandatory to reduce seizure occurrence intraoperatively and in the early postoperative period.
Purpose
The improving knowledge of interactions between meningiomas and progestin refines the management of this specific condition. We assessed the changes over time of the management of ...progestin-associated meningiomas.
Methods
We retrospectively studied consecutive adult patients who had at least one meningioma in the context of progestin intake (October 1995–October 2018) in a tertiary adult Neurosurgical Center.
Results
71 adult women with 125 progestin-associated meningiomas were included. The number of progestin-associated meningioma patients increased over time (0.5/year before 2008, 22.0/year after 2017). Progestin treatment was an approved indication in 27.0%. A mean of 1.7 ± 1.2 meningiomas were discovered per patient (median 1, range 1–6). Surgery was performed on 36 (28.8%) meningiomas and the histopathologic grading was WHO grade 1 in 61.1% and grade 2 in 38.9%. The conservative management of meningiomas increased over time (33.3% before 2008, 64.3% after 2017) and progestin treatment withdrawal increased over time (16.7% before 2008, 95.2% after 2017). Treatment withdrawal varied depending on the progestin derivative used (88.9% with cyproterone acetate, 84.6% with chlormadinone acetate, 28.6% with nomegestrol acetate, 66.7% with progestin derivative combination). The main reason for therapeutic management of meningiomas was the presence of clinical signs. Among the 54 meningiomas managed conservatively for which the progestin had been discontinued, MRI follow-up demonstrated a regression in 29.6%, a stability in 68.5%, and an ongoing growth in 1.9% of cases.
Conclusions
Conservative management, including progestin treatment discontinuation, has grown over time with promising results in terms of efficacy and safety.
Abstract
Background
We quantified the spontaneous imaging growth rate of oligodendrogliomas. We assessed whether (i) it discriminates between World Health Organization (WHO) grade II and grade III ...oligodendrogliomas, and (ii) grade III oligodendrogliomas with neo-angiogenesis are associated with more fast growth rates (≥8 mm/y).
Methods
This work employed a retrospective bicentric cohort study (2010–2016) of adult patients harboring a newly diagnosed supratentorial oligodendroglioma, isocitrate dehydrogenase (IDH) mutant and 1p/19q codeleted (WHO 2016 classification), with a minimum of 2 available MRIs before any treatment (minimum 6-week interval) to measure the spontaneous tumor growth rate.
Results
We included 108 patients (age 44.7 ± 14.1 y, 60 males). The tumor growth rate was higher in grade III oligodendrogliomas with neo-angiogenesis (n = 37, median 10.4 mm/y, mean 10.0 ± 6.9) than in grade III oligodendrogliomas with increased mitosis count only (cutoff ≥6 mitoses, n = 18, median 3.9 mm/y, mean 4.5 ± 3.2; P = 0.004), and higher than in grade II oligodendrogliomas (n = 53, median 2.3 mm/y, mean 2.8 ± 2.2; P < 0.001). There was increased prevalence of fast tumor growth rates in grade III oligodendrogliomas with neo-angiogenesis (54.1%) compared with grade III oligodendrogliomas with increased mitosis count only (11.1%; P < 0.001), and in grade II oligodendrogliomas (0.0%; P < 0.001). The tumor growth rate trends did not differ between centers (P = 0.121). Neo-angiogenesis (P < 0.001) and mitosis count at ≥9 (P = 0.013) were independently associated with tumor growth rates ≥8 mm/year. A tumor growth rate ≥8 mm/year was the only predictor independently associated with shorter progression-free survival (P = 0.041).
Conclusions
The spontaneous tumor growth rate recapitulates oligodendroglioma aggressiveness, permits identification of grade III oligodendrogliomas preoperatively when ≥8 mm/year, and questions the grading by mitosis count.
PDF
