Showing papers by "Simone Maurea published in 2020"

MRI Radiomics for the Prediction of Fuhrman Grade in Clear Cell Renal Cell Carcinoma: a Machine Learning Exploratory Study

Abstract: The Fuhrman nuclear grade is a recognized prognostic factor for patients with clear cell renal cell carcinoma (CCRCC) and its pre-treatment evaluation significantly affects decision-making in terms of management. In this study, we aimed to assess the feasibility of a combined approach of radiomics and machine learning based on MR images for a non-invasive prediction of Fuhrman grade, specifically differentiation of high- from low-grade tumor and grade assessment. Images acquired on a 3-Tesla scanner (T2-weighted and post-contrast) from 32 patients (20 with low-grade and 12 with high-grade tumor) were annotated to generate volumes of interest enclosing CCRCC lesions. After image resampling, normalization, and filtering, 2438 features were extracted. A two-step feature reduction process was used to between 1 and 7 features depending on the algorithm employed. A J48 decision tree alone and in combination with ensemble learning methods were used. In the differentiation between high- and low-grade tumors, all the ensemble methods achieved an accuracy greater than 90%. On the other end, the best results in terms of accuracy (84.4%) in the assessment of tumor grade were achieved by the random forest. These evidences support the hypothesis that a combined radiomic and machine learning approach based on MR images could represent a feasible tool for the prediction of Fuhrman grade in patients affected by CCRCC.

FDG-PET/CT imaging during the Covid-19 emergency: a southern Italian perspective.

Abstract: Purpose To assess the impact of the Covid-19 pandemic on FDG-PET/CT work volume and to evaluate the occurrence of abnormal imaging findings suspicious or potentially diagnostic for interstitial pneumonia by Covid-19 infection in south Italy Methods We retrospectively reviewed the number and the findings of FDG-PET/CT studies acquired between February and April 2020 during the Covid-19 pandemic at the University of Napoli Federico II The number and the findings of FDG-PET/CT studies acquired in the corresponding period of 2019 were also assessed for direct comparison Results The number of FDG-PET/CT studies performed during the pandemic (n = 299) and in the corresponding period of 2019 (n = 335) were comparable The percentage of abnormal FDG-PET/CT findings, suspicious for interstitial pneumonia by Covid-19 infection, was significantly higher during the pandemic (9%) compared with that found in the corresponding period of 2019 (4%) (χ2 545, P = 002) No significant differences were observed in the distribution of Covid-19 reporting and data system (CO-RADS) classification and in the maximum standardized uptake value between the pandemic (26 ± 22) and the corresponding period of 2019 (32 ± 14) Of note, patients with abnormal imaging findings during the pandemic time had clinical data and/or laboratory tests negative for Covid-19 infection Conclusion Despite the restrictive medical measures for the emergency, the number of FDG-PET/CT studies was unchanged during the pandemic compared with the previous year Our findings also indicate that Covid-19 infection was contained in our series of patients from southern Italy

Ultrasound, shear-wave elastography, and magnetic resonance imaging in native liver survivor patients with biliary atresia after Kasai portoenterostomy: correlation with medical outcome after treatment.

Abstract: BackgroundBiliary atresia (BA) is a rare obliterative cholangiopathy and Kasai portoenterostomy (KP) represents its first-line treatment; clinical and laboratory parameters together with abdominal ...

Resolution Resampling of Ultrasound Images in Placenta Previa Patients: Influence on Radiomics Data Reliability and Usefulness for Machine Learning

Abstract: Placenta previa (PP) and Placenta Accreta Spectrum (PAS) are obstetric pathologies whose early detection is fundamental for an appropriate patient management. In this paper, ultrasonography (US) is performed on 53 patients and from the images a texture analysis feature extraction is performed through PyRadiomics. The US images were acquired with 3 different resampling resolutions: 1 × 1, 2 × 2 and 3 × 3. The features extracted from the images at each resolution were used to investigate which one is the best to make the correct diagnosis by employing machine learning techniques. Knime analytics platform was employed to implement decision tree, k nearest neighbor and naive Bayes. Synthetic minority oversampling technique was used to balance the dataset and some evaluation metrics were computed after a leave one out cross-validation. Averaging all the metrics among all the algorithms, 1 × 1 resolution achieved the best mean accuracy (75.97%), sensitivity (83.33%), specificity (68.50%) and Area Under the Curve Receiver Operating Characteristics (0.81). Moreover, k nearest neighbor was the algorithm with the highest metrics (greater than 80%). Despite using also artificial data to balance the dataset (less than 30% of total analysed sample), this study provides researchers with the idea that employing a 1 × 1 resolution could be the best option when analysing images with machine learning algorithms on texture analysis features US-derived.

Comparison Between Multidetector CT and High-Field 3T MR Imaging in Diagnostic and Tumour Extension Evaluation of Patients with Cholangiocarcinoma.

Abstract: To compare multidetector computed tomography (MDCT) and magnetic resonance imaging (MRI) with cholangiopancreatography (MRCP) findings in the diagnostic evaluation of patients with cholangiocarcinoma (CCA) to establish tumour resectability. Thirty patients (22 M, 8 F) with pathologically proven CCA by post-surgical specimens (n = 20), core biopsy (n = 6) or cytology (n = 4) underwent both MDCT and MRI with MRCP. CCA lesions were classified on the basis of anatomical locations in intra-hepatic (iCCA), peri-hilar (pCCA) and distal (dCCA) tumours. Morphological tumour pattern, lesion size, biliary dilatation, tumour contrast enhancement type, lymph node involvement and vascular infiltration were directly compared between MDCT and MRI with MRCP. As a rule, a tumour resectability judgement for each patient was formulated by both imaging techniques comparing imaging results with direct surgical assessment (n = 20) or interventional procedures (n = 10). In terms of anatomical location, 14 iCCA, 8 pCCA and 8 dCCA were observed; both imaging techniques were concordant about the identification and morphological characterization of tumour lesions and in the evaluation of tumour features (lesion size, contrast enhancement pattern, capsular retraction, biliary dilatation, lymph node involvement and vascular infiltration) as well as in assessing lesion resectability; an excellent agreement (k = 1) for the assessment of all the parameters included in imaging analysis was observed. The comparative concordant results of our study suggest that MRI with MRCP represents a valid alternative to MDCT for the diagnostic evaluation of patients with CCA to establish tumour resectability providing multiplanar scanning of high-contrast imaging quality; MDCT should be preferred in uncooperative patients, in the presence of biliary stents or when MRI is absolutely contraindicated for incompatible medical devices.