BS-Net: Learning COVID-19 pneumonia severity on a large chest X-ray dataset.

In this work we design an end-to-end deep learning architecture for predicting, on Chest X-rays images (CXR), a multi-regional score conveying the degree of lung compromise in COVID-19 patients. Such semi-quantitative scoring system, namely Brixia score, is applied in serial monitoring of such patients, showing significant prognostic value, in one of the hospitals that experienced one of the highest pandemic peaks in Italy. To solve such a challenging visual task, we adopt a weakly supervised learning strategy structured to handle different tasks (segmentation, spatial alignment, and score estimation) trained with a "from-the-part-to-the-whole" procedure involving different datasets. In particular, we exploit a clinical dataset of almost 5,000 CXR annotated images collected in the same hospital. Our BS-Net demonstrates self-attentive behavior and a high degree of accuracy in all processing stages. Through inter-rater agreement tests and a gold standard comparison, we show that our solution outperforms single human annotators in rating accuracy and consistency, thus supporting the possibility of using this tool in contexts of computer-assisted monitoring. Highly resolved (super-pixel level) explainability maps are also generated, with an original technique, to visually help the understanding of the network activity on the lung areas. We also consider other scores proposed in literature and provide a comparison with a recently proposed non-specific approach. We eventually test the performance robustness of our model on an assorted public COVID-19 dataset, for which we also provide Brixia score annotations, observing good direct generalization and fine-tuning capabilities that highlight the portability of BS-Net in other clinical settings. The CXR dataset along with the source code and the trained model are publicly released for research purposes.

Myocarditis: imaging up to date.

Myocarditis is an inflammatory disease of the heart muscle, diagnosed by histological, immunological, and immunohistochemical criteria. Endomyocardial biopsy represents the diagnostic gold standard for its diagnosis but is infrequently used. Due to its noninvasive ability to detect the presence of myocardial edema, hyperemia and necrosis/fibrosis, Cardiac MR imaging is routinely used in the clinical practice for the diagnosis of acute myocarditis. Recently pixel-wise mapping of T1 and T2 relaxation time have been introduced into the clinical Cardiac MR protocol increasing its accuracy. Our paper will review the role of MR imaging in the diagnosis of acute myocarditis.

Adrenocortical Carcinoma and CT Assessment of Therapy Response: The Value of Combining Multiple Criteria.

We evaluated tumor response at Computed Tomography (CT) according to three radiologic criteria: RECIST 1.1, CHOI and tumor volume in 34 patients with metastatic adrenocortical carcinoma (ACC) submitted to standard chemotherapy. These three criteria agreed in defining partial response, stable or progressive disease in 24 patients (70.5%). Partial response (PR) was observed in 29.4%, 29.4% and 41.2% of patients according to RECIST 1.1, CHOI and tumor volume, respectively. It was associated with a favorable prognosis, regardless of the criterion adopted. The concordance of all the 3 criteria in defining the disease response identified 8 patients (23.5%) which displayed a very good prognosis: median progression free survival (PFS) and overall survival (OS) 14.9 and 37.7 months, respectively. Seven patients (20.6%) with PR assessed by one or two criteria, however, still had a better prognosis than non-responding patients, both in terms of PFS: median 12.3 versus 9.9 months and OS: 21 versus 12.2, respectively. In conclusions, the CT assessment of disease response of ACC patients to chemotherapy with 3 different criteria is feasible and allows the identification of a patient subset with a more favorable outcome. PR with at least one criterion can be useful to early identify patients that deserve continuing the therapy.

Technical Success and Long-Term Results with Excluder/C3 Endoprosthesis in Narrow Aortic Bifurcations: First Italian Multicentre Experience.

BACKGROUND: This study aims at evaluating technical success and long-term results using Gore Excluder/C3 endoprosthesis in patients with narrow aortic bifurcation (NAB; inner aortic diameter < 18 mm). METHODS: Clinical and anatomical data were collected retrospectively from patients treated in 2 high-volume Italian vascular centers between 2005 and 2017. A total of 1325 endovascular aneurysm repair procedures were performed, of which 195 involved Excluder/C3 Gore endoprosthesis. One hundred forty-one patients had a regular aortic bifurcation (RAB; maximum inner diameter ≥ 18 mm), whereas 54 presented with NAB (<18 mm). Technical success and procedural time were considered as primary outcomes. Secondary outcomes were perioperative complications, long-term graft-related complications and reintervention rates. RESULTS: Demographic data and risk factors were similar in the 2 groups. The mean diameter of aortic bifurcation was 29.8 ± 10.4 mm in RAB versus 17.1 ± 0.9 mm in NAB. Technical success was 100% in both groups. Incidence of intraoperative kinking/stenosis of limb graft was significantly higher in NAB (40.7% vs. 12.8%; P < 0.001), which was treated by means of kissing balloon technique (KiBaTe) and selective stenting. Time of procedure was similar. Post-operative complications rate was similar in both the groups (9.5% in RAB versus 4.2% in NAB, P = 0.180). Mean follow-up period was 40.0 months (range, 1-130). No significant difference was registered in long-term graft-related complications between RAB and NAB (38.3% vs. 38.8%; P = 0.939). Reintervention rate after 70 months was 21.8% vs. 24.6% in RAB and NAB, respectively (log rank = 0.517). Primary patency at 6, 12, and 48 months was 99.3% vs. 100%, 99.3% vs. 100%, and 98.5% vs. 97.9%, respectively (log rank = 0.497). Assisted primary patency was 98.6% vs. 96.0%, 97.8% vs. 96.0%, and 97.8% vs. 96.0%, respectively (log rank = 0.789). Secondary patency was 100% in both the groups. CONCLUSIONS: Endovascular aneurysm repair of NAB using Gore Excluder/C3 graft can be considered safe and effective in early and late follow-up. Intraoperative graft kinking is frequent in NAB and KiBaTe is recommended to prevent occlusive complications.