RESUMEN
Chest CT is a useful initial exam in patients with coronavirus disease 2019 (COVID-19) for assessing lung damage. AI-powered predictive models could be useful to better allocate resources in the midst of the pandemic. Our aim was to build a deep-learning (DL) model for COVID-19 outcome prediction inclusive of 3D chest CT images acquired at hospital admission. This retrospective multicentric study included 1051 patients (mean age 69, SD = 15) who presented to the emergency department of three different institutions between 20th March 2020 and 20th January 2021 with COVID-19 confirmed by real-time reverse transcriptase polymerase chain reaction (RT-PCR). Chest CT at hospital admission were evaluated by a 3D residual neural network algorithm. Training, internal validation, and external validation groups included 608, 153, and 290 patients, respectively. Images, clinical, and laboratory data were fed into different customizations of a dense neural network to choose the best performing architecture for the prediction of mortality, intubation, and intensive care unit (ICU) admission. The AI model tested on CT and clinical features displayed accuracy, sensitivity, specificity, and ROC-AUC, respectively, of 91.7%, 90.5%, 92.4%, and 95% for the prediction of patient's mortality; 91.3%, 91.5%, 89.8%, and 95% for intubation; and 89.6%, 90.2%, 86.5%, and 94% for ICU admission (internal validation) in the testing cohort. The performance was lower in the validation cohort for mortality (71.7%, 55.6%, 74.8%, 72%), intubation (72.6%, 74.7%, 45.7%, 64%), and ICU admission (74.7%, 77%, 46%, 70%) prediction. The addition of the available laboratory data led to an increase in sensitivity for patient's mortality (66%) and specificity for intubation and ICU admission (50%, 52%, respectively), while the other metrics maintained similar performance results. We present a deep-learning model to predict mortality, ICU admittance, and intubation in COVID-19 patients. KEY POINTS: ⢠3D CT-based deep learning model predicted the internal validation set with high accuracy, sensibility and specificity (> 90%) mortality, ICU admittance, and intubation in COVID-19 patients. ⢠The model slightly increased prediction results when laboratory data were added to the analysis, despite data imbalance. However, the model accuracy dropped when CT images were not considered in the analysis, implying an important role of CT in predicting outcomes.
Asunto(s)
COVID-19 , Aprendizaje Profundo , Humanos , Anciano , COVID-19/diagnóstico por imagen , SARS-CoV-2 , Estudios Retrospectivos , Tomografía Computarizada por Rayos X/métodos , Unidades de Cuidados Intensivos , Intubación IntratraquealRESUMEN
PURPOSE: To investigate the diagnostic efficacy of MRI diagnostic algorithms with an ascending automatization, in distinguishing between high-grade glioma (HGG) and solitary brain metastases (SBM). METHODS: 36 patients with histologically proven HGG (n = 18) or SBM (n = 18), matched by size and location were enrolled from a database containing 655 patients. Four different diagnostic algorithms were performed serially to mimic the clinical setting where a radiologist would typically seek out further findings to reach a decision: pure qualitative, analytic qualitative (based on standardized evaluation of tumor features), semi-quantitative (based on perfusion and diffusion cutoffs included in the literature) and a quantitative data-driven algorithm of the perfusion and diffusion parameters. The diagnostic yields of the four algorithms were tested with ROC analysis and Kendall coefficient of concordance. RESULTS: Qualitative algorithm yielded sensitivity of 72.2%, specificity of 78.8%, and AUC of 0.75. Analytic qualitative algorithm distinguished HGG from SBM with a sensitivity of 100%, specificity of 77.7%, and an AUC of 0.889. The semi-quantitative algorithm yielded sensitivity of 94.4%, specificity of 83.3%, and AUC = 0.889. The data-driven algorithm yielded sensitivity = 94.4%, specificity = 100%, and AUC = 0.948. The concordance analysis between the four algorithms and the histologic findings showed moderate concordance for the first algorithm, (k = 0.501, P < 0.01), good concordance for the second (k = 0.798, P < 0.01), and third (k = 0.783, P < 0.01), and excellent concordance for fourth (k = 0.901, p < 0.0001). CONCLUSION: When differentiating HGG from SBM, an analytical qualitative algorithm outperformed qualitative algorithm, and obtained similar results compared to the semi-quantitative approach. However, the use of data-driven quantitative algorithm yielded an excellent differentiation.
Asunto(s)
Neoplasias Encefálicas , Glioma , Algoritmos , Neoplasias Encefálicas/secundario , Glioma/diagnóstico por imagen , Glioma/patología , Humanos , Imagen por Resonancia Magnética/métodos , Clasificación del Tumor , Curva ROC , Sensibilidad y EspecificidadRESUMEN
BACKGROUND: Given the recent concerns about gadolinium-based contrast agent safety, dose reduction strategies are being investigated. OBJECTIVE: To compare half-dose and standard full-dose gadoterate meglumine at 3-tesla (T) MRI in paediatric bone and soft-tissue diseases. MATERIALS AND METHODS: We prospectively enrolled 45 children (age range 2.7 months to 17.5 years, median age 8.7 years, 49 total anatomical segments) with bone and soft-tissue diseases (neoplastic, inflammatory/infectious, ischaemic and vascular) imaged at 3-T MRI. Two consecutive half-doses of gadoterate meglumine (0.05 mmol/kg body weight) were administered. Two sets of post-contrast T1-weighted images were obtained, one after the first half dose and the other after the second half dose. For qualitative analysis, three radiologists, masked to the gadolinium dose, compared the diagnostic quality of the images. For quantitative analysis, we compared signal-to-noise ratio and contrast-to-noise ratio at half and full doses. RESULTS: Signal-to-noise ratio and contrast-to-noise ratio did not vary significantly between the two groups. Qualitative analysis yielded excellent image quality in both post-contrast image datasets (Cohen κ=0.8). CONCLUSION: In paediatric bone and soft-tissue 3-T MRI, it is feasible to halve the standard dose of gadoterate meglumine without losing image quality.
Asunto(s)
Medios de Contraste/administración & dosificación , Imagen por Resonancia Magnética/métodos , Meglumina/administración & dosificación , Enfermedades Musculoesqueléticas/diagnóstico por imagen , Compuestos Organometálicos/administración & dosificación , Enfermedades Vasculares/diagnóstico por imagen , Adolescente , Niño , Preescolar , Medios de Contraste/efectos adversos , Femenino , Humanos , Lactante , Masculino , Meglumina/efectos adversos , Compuestos Organometálicos/efectos adversos , Estudios Prospectivos , Reproducibilidad de los Resultados , Sensibilidad y Especificidad , Relación Señal-RuidoRESUMEN
Background and purpose: Differentiating pediatric posterior fossa (PF) tumors such as medulloblastoma (MB), ependymoma (EP), and pilocytic astrocytoma (PA) remains relevant, because of important treatment and prognostic implications. Diffusion kurtosis imaging (DKI) has not yet been investigated for discrimination of pediatric PF tumors. Estimating diffusion values from whole-tumor-based (VOI) segmentations may improve diffusion measurement repeatability compared to conventional region-of-interest (ROI) approaches. Our purpose was to compare repeatability between ROI and VOI DKI-derived diffusion measurements and assess DKI accuracy in discriminating among pediatric PF tumors. Materials and methods: We retrospectively analyzed 34 children (M, F, mean age 7.48 years) with PF tumors who underwent preoperative examination on a 3 Tesla magnet, including DKI. For each patient, two neuroradiologists independently segmented the whole solid tumor, the ROI of the area of maximum tumor diameter, and a small 5 mm ROI. The automated analysis pipeline included inter-observer variability, statistical, and machine learning (ML) analyses. We evaluated inter-observer variability with coefficient of variation (COV) and Bland-Altman plots. We estimated DKI metrics accuracy in discriminating among tumor histology with MANOVA analysis. In order to account for class imbalances, we applied SMOTE to balance the dataset. Finally, we performed a Random Forest (RF) machine learning classification analysis based on all DKI metrics from the SMOTE dataset by partitioning 70/30 the training and testing cohort. Results: Tumor histology included medulloblastoma (15), pilocytic astrocytoma (14), and ependymoma (5). VOI-based measurements presented lower variability than ROI-based measurements across all DKI metrics and were used for the analysis. DKI-derived metrics could accurately discriminate between tumor subtypes (Pillai's trace: p < 0.001). SMOTE generated 11 synthetic observations (10 EP and 1 PA), resulting in a balanced dataset with 45 instances (34 original and 11 synthetic). ML analysis yielded an accuracy of 0.928, which correctly predicted all but one lesion in the testing set. Conclusions: VOI-based measurements presented improved repeatability compared to ROI-based measurements across all diffusion metrics. An ML classification algorithm resulted accurate in discriminating PF tumors on a SMOTE-generated dataset. ML techniques based on DKI-derived metrics are useful for the discrimination of pediatric PF tumors.
RESUMEN
Purpose: To develop a predictive grading model based on diffusion kurtosis imaging (DKI) metrics in children affected by gliomas, and to investigate the clinical impact of the predictive model by correlating with overall survival and progression-free survival. Materials and methods: 59 patients with a histological diagnosis of glioma were retrospectively studied (33 M, 26 F, median age 7.2 years). Patients were studied on a 3T scanner with a standardized MR protocol, including conventional and DKI sequences. Mean kurtosis (MK), axial kurtosis (AK), radial kurtosis (RK), fractional anisotropy (FA), and apparent diffusion coefficient (ADC) maps were obtained. Whole tumour volumes (VOIs) were segmented semi-automatically. Mean DKI values were calculated for each metric. The quantitative values from DKI-derived metrics were used to develop a predictive grading model to develop a probability prediction of a high-grade glioma (pHGG). Three models were tested: DTI-based, DKI-based, and combined (DTI and DKI). The grading accuracy of the resulting probabilities was tested with a receiver operating characteristics (ROC) analysis for each model. In order to account for dataset imbalances between pLGG and pHGG, we applied a random synthetic minority oversampling technique (SMOTE) analysis. Lastly, the most accurate model predictions were correlated with progression-free survival (PFS) and overall survival (OS) using the Kaplan−Meier method. Results: The cohort included 46 patients with pLGG and 13 patients with pHGG. The developed model predictions yielded an AUC of 0.859 (95%CI: 0.752−0.966) for the DTI model, of 0.939 (95%CI: 0.879−1) for the DKI model, and of 0.946 (95%CI: 0.890−1) for the combined model, including input from both DTI and DKI metrics, which resulted in the most accurate model. Sample estimation with the random SMOTE analysis yielded an AUC of 0.98 on the testing set. Model predictions from the combined model were significantly correlated with PFS (25.2 months for pHGG vs. 40.0 months for pLGG, p < 0.001) and OS (28.9 months for pHGG vs. 44.9 months for pLGG, p < 0.001). Conclusions: a DKI-based predictive model was highly accurate for pediatric glioma grading. The combined model, derived from both DTI and DKI metrics, proved that DKI-based model predictions of tumour grade were significantly correlated with progression-free survival and overall survival.
RESUMEN
We present a case demonstrating the performance of different radiographical imaging modalities in the diagnostic work-up of a patient with neurofibromatosis type 1 (NF1) and plexiform neurofibroma (PN). The newborn boy showed an expansive-infiltrative cervical and facial mass presented with macrocrania, craniofacial disfigurement, exophthalmos and glaucoma. A computer tomography (CT) and a magnetic resonance imaging (MRI) were performed. The CT was fundamental to evaluate the bone dysmorphisms and the MRI was crucial to estimate the mass extension. The biopsy of the lesion confirmed the suspicion of PN, thus allowing the diagnosis of NF1. PN is a variant of neurofibromas, a peripheral nerves sheath tumor typically associated with NF1. Even through currently available improved detection techniques, NF1 diagnosis at birth remains a challenge due to a lack of pathognomonic signs; therefore congenital PN are recognized in 20% of cases. This case highlights the importance of using different radiological methods both for the correct diagnosis and the follow-up of the patient with PN. Thanks to MRI evaluation, it was possible to identify earlier the progressive increasing size of the PN and the possible life threatening evolution in order to perform a tracheostomy to avoid airways compression.
RESUMEN
Background: Diffuse intrinsic pontine glioma (DIPG) has a dismal prognosis. Magnetic resonance imaging (MRI) remains the gold standard for non-invasive DIPG diagnosis. MRI features have been tested as surrogate biomarkers. We investigated the direct involvement of cranial nerve V (CN V) in DIPG at diagnosis and its utility as predictor of poor overall survival. Materials and Methods: We examined MRI scans of 35 consecutive patients with radiological diagnosis of DIPG. Direct involvement of CN V was assessed on the diagnostic scans. Differences in overall survival (OS) and time to progression (TTP) were analyzed for involvement of CN V, sex, age, tumor size, ring enhancement, and treatment regimen. Correlations between involvement of CN V and disease dissemination, magnet strength and slice thickness were analyzed. Statistical analyses included Kaplan-Meier curves, log-rank test and Spearman's Rho. Results: After excluding six long-term survivors, 29 patients were examined (15 M, 14 F). Four patients presented direct involvement of CN V. Histological data were available in 12 patients. Median OS was 11 months (range 3-23 months). Significant differences in OS were found for direct involvement of CN V (median OS: 7 months, 95% CI 1.1-12.9 months for involvement of CN V vs. 13 months, 95% CI 10.2-15.7 for lack of involvement of CN V, respectively, p < 0.049). Significant differences in TTP were found for the two treatment regimens (median TTP: 4 months, 95% CI 2.6-5.3 vs. 7 months, 95% CI 5.9-8.1, respectively, p < 0.027). No significant correlation was found between involvement of CN V and magnet strength or slice thickness (r = -0.201; p = NS). A trend toward positive correlation was found between direct involvement of CN V at diagnosis and dissemination of disease at follow-up (r = 0.347; p < 0.065). Conclusions: In our cohort, direct involvement of CN V correlated with poor prognosis. Based on our data, we suggest that in DIPG direct involvement of CN V should be routinely evaluated on diagnostic scans.
RESUMEN
Medulloblastoma is the most common malignant pediatric brain tumor. Medulloblastoma should not be viewed as a single disease, but as a heterogeneous mixture of various subgroups with distinct characteristics. Based on genomic profiles, four distinct molecular subgroups are identified: Wingless (WNT), Sonic Hedgehog (SHH), Group 3 and Group 4. Each of these subgroups are associated with specific genetic aberrations, typical age of onset as well as survival prognosis. Magnetic resonance imaging (MRI) is performed for all patients with brain tumors, and has a key role in the diagnosis, surgical guidance and follow up of patients with medulloblastoma. Several studies indicate MRI as a promising tool for early detection of medulloblastoma subgroups. The early identification of the subgroup can influence the extent of surgical resection, radiotherapy and chemotherapy targeted treatments. In this article, we review the state of the art in MRI-facilitated medulloblastoma subgrouping, with a summary of the main MRI features in medulloblastoma and a brief discussion on molecular characterization of medulloblastoma subgroups. The main focus of the article is MRI features that correlate with medulloblastoma subtypes, as well as features suggestive of molecular subgroups. Finally, we briefly discuss the latest trends in MRI studies and latest developments in molecular characterization.