Cardiothoratic ratio (CTR) estimated from chest radiographs is a marker indicative of cardiomegaly, the presence of which is in the criteria for heart failure diagnosis. Existing methods for automatic assessment of CTR are driven by Deep Learning-based segmentation. However, these techniques produce only point estimates of CTR but clinical decision making typically assumes the uncertainty. In this paper, we propose a novel method for chest X-ray segmentation and CTR assessment in an automatic manner. In contrast to the previous art, we, for the first time, propose to estimate CTR with uncertainty bounds. Our method is based on Deep Convolutional Neural Network with Feature Pyramid Network (FPN) decoder. We propose two modifications of FPN: replace the batch normalization with instance normalization and inject the dropout which allows to obtain the Monte-Carlo estimates of the segmentation maps at test time. Finally, using the predicted segmentation mask samples, we estimate CTR with uncertainty. In our experiments we demonstrate that the proposed method generalizes well to three different test sets. Finally, we make the annotations produced by two radiologists for all our datasets publicly available.

Objective MEN1 is associated with an increased risk of developing tumors in different endocrine organs. Neuroendocrine tumors of the thymus (TNETs) are very rare but often have an aggressive nature. We evaluated patients with MEN1 and TNET in three university hospitals in Finland. Design/Methods We evaluated patient records of 183 MEN1-patients from three university hospitals between the years 1985–2019 with TNETs. Thymus tumor specimens were classified according to the new WHO 2021 classification of TNET. We collected data on treatments and outcomes of these patients. Results There were six patients (3.3%) with MEN1 and TNET. Five of them had the same common gene mutation occurring in Finland. They originated from common ancestors encompassing two pairs of brothers from sequential generations. The mean age at presentation of TNET was 44.7 ± 11.9 years. TNET was classified as atypical carcinoid (AC) in five out of six patients. One patient had a largely necrotic main tumor with very few mitoses and another nodule with 25 mitoses per 2 mm2, qualifying for the 2021 WHO diagnosis of large cell neuroendocrine carcinoma (LCNEC). In our patients, the 5-year survival of the TNET patients was 62.5% and 10-year survival 31.3%. Conclusion In this study, TNETs were observed in one large MEN1 founder pedigree, where an anticipation-like earlier disease onset was observed in the most recent generation. TNET in MEN1 patients is an aggressive disease. The prognosis can be better by systematic screening. We also show that LCNEC can be associated with TNET in MEN1 patients.

No abstract