Artificial intelligence and radiomics enhance the positive predictive value of digital chest tomosynthesis for lung cancer detection within SOS clinical trial.

OBJECTIVE: To enhance the positive predictive value (PPV) of chest digital tomosynthesis (DTS) in the lung cancer detection with the analysis of radiomics features. METHOD: The investigation was carried out within the SOS clinical trial (NCT03645018) for lung cancer screening with DTS. Lung nodules were identified by visual analysis and then classified using the diameter and the radiological aspect of the nodule following lung-RADS. Haralick texture features were extracted from the segmented nodules. Both semantic variables and radiomics features were used to build a predictive model using logistic regression on a subset of variables selected with backward feature selection and using two machine learning: a Random Forest and a neural network with the whole subset of variables. The methods were applied to a train set and validated on a test set where diagnostic accuracy metrics were calculated. RESULTS: Binary visual analysis had a good sensitivity (0.95) but a low PPV (0.14). Lung-RADS classification increased the PPV (0.19) but with an unacceptable low sensitivity (0.65). Logistic regression showed a mildly increased PPV (0.29) but a lower sensitivity (0.20). Random Forest demonstrated a moderate PPV (0.40) but with a low sensitivity (0.30). Neural network demonstrated to be the best predictor with a high PPV (0.95) and a high sensitivity (0.90). CONCLUSIONS: The neural network demonstrated the best PPV. The use of visual analysis along with neural network could help radiologists to reduce the number of false positive in DTS. KEY POINTS: • We investigated several approaches to enhance the positive predictive value of chest digital tomosynthesis in the lung cancer detection. • Neural network demonstrated to be the best predictor with a nearly perfect PPV. • Neural network could help radiologists to reduce the number of false positive in DTS.

Clinical and magnetic resonance evolution of "infarct-like" myocarditis.

AIMS: To analyse the clinical and magnetic resonance evolution of myocarditis in patients with an "infarct-like" presentation pattern. METHODS: The study is a retrospective analysis of 52 patients with clinical diagnosis of "infarct-like" myocarditis confirmed by CMR as acute myocarditis according to Lake Louise criteria and 6 months follow-up. The CMR protocol included T2-weighted (oedema), early (hyperaemia) and late (fibrosis/necrosis) gadolinium enhancement sequences, according to Lake Louise criteria. Clinical and radiological follow-up by CMR was performed after a median time interval of 6 months (interquartile range 5-8). Quantitative outcomes were checked for normality and compared with the non-parametric Wilcoxon's test for matched data. RESULTS: At the clinical follow-up all patients were free of symptoms and reported no cardiac complications. The CMR follow-up evidenced a significant increase of the ejection fraction (from 53 ± 6 to 55 ± 4%, p = 0.03), a decrease of the ventricular mass [from 67.0 (58.8-79.0) to 61.0 (54.0-67.0), p < 0.0001] without significant modification of the cardiac volume index (p = 0.26). No patient had residual oedema or capillary leakage: the T2 ratio decreased from 3.94 (3.00-4.86) to 0.98 (0.75-1.17) with p < 0.0001 and the Early gadolinium enhancement (EGE) ratio from 5.7 (4.8-6.5) to 2.9 (2.4-3.2) with p < 0.0001. Late gadolinium enhancement (LGE) persisted over the course of the follow-up in 48/52 patients, but with a significant reduction in every patient (LGE % from 34.3 ± 9.1 to 19.4 ± 6.6%; p < 0.0001). CONCLUSION: Patients diagnosed with "infarct-like" myocarditis, according to both clinical and CMR examinations may look forward to a positive evolution with a good prognosis.

A rare case of adrenal ganglioneuroblastoma-intermixed in an adult and a review of literature.

Peripheral neuroblastic tumors are extremely rare in the adult with less just over 20 cases involving adrenal gland described in the literature. We reported herewith the case of a 22-year-old young male who presented with epigastric pain and diarrhea. Imaging studies documented a 3.5cm x 3cm x 4cm solid well-circumscribed right adrenal mass, of heterogeneous structure and with fine calcifications. The lesion turned negative at MIBG scintigraphy. A right robotic-assisted adrenalectomy was performed leading to complete excision of the lesion without complications. Histology was consistent with intermixed stroma-rich ganglioneuroblastoma. A wait-and-see strategy was considered adequate. Two years after diagnosis patient is alive disease-free. Although the definitive diagnosis of a peripheral neuroblastic tumor is obtained after histopathological analysis, CT, and MRI are helpful to further characterize masses and useful in pretreatment risk stratification. Clinicians should be aware of the possibility of GNB development in adult population and its malignant potential.