Background: In patients with soft-tissue sarcomas, tumor grading constitutes a decisive factor to determine the best treatment decision. Tumor grading is obtained by pathological work-up after focal biopsies. Deep learning (DL)-based imaging analysis may pose an alternative way to characterize STS tissue. In this work, we sought to non-invasively differentiate tumor grading into low-grade (G1) and high-grade (G2/G3) STS using DL techniques based on MR-imaging. Methods: Contrast-enhanced T1-weighted fat-saturated (T1FSGd) MRI sequences and fat-saturated T2-weighted (T2FS) sequences were collected from two independent retrospective cohorts (training: 148 patients, testing: 158 patients). Tumor grading was determined following the French Federation of Cancer Centers Sarcoma Group in pre-therapeutic biopsies. DL models were developed using transfer learning based on the DenseNet 161 architecture. Results: The T1FSGd and T2FS-based DL models achieved area under the receiver operator characteristic curve (AUC) values of 0.75 and 0.76 on the test cohort, respectively. T1FSGd achieved the best F1-score of all models (0.90). The T2FS-based DL model was able to significantly risk-stratify for overall survival. Attention maps revealed relevant features within the tumor volume and in border regions. Conclusions: MRI-based DL models are capable of predicting tumor grading with good reproducibility in external validation.

Background Ultrasound is an essential imaging tool for evaluating physiological and pathological fetal or maternal conditions during pregnancy. Published data is limited with respect to the application of CEUS during pregnancy. CEUS has already been safely applied for assessing uteroplacental blood flow, cesarean scar pregnancy and invasive placenta percreta. CT and MRI scans during pregnancy must be thoroughly evaluated due to harmful ionizing radiation and cerebral gadolinium deposition, respectively. Purpose The aim of the present retrospective single-center study is to assess the diagnostic performance and safety of CEUS during pregnancy to evaluate hepatic lesions of unknown entity. Methods 6 pregnant patients who underwent CEUS between 2005 and 2014 (mean age: 32 years; mean weeks of pregnancy: 28 weeks) were included in this study. The applied contrast agent was a second-generation blood-pool agent (SonoVue^®, Bracco, Milan, Italy). CEUS examinations were performed and interpreted by a single radiologist with experience since 2000 (EFSUMB Level 3). Results CEUS was safely performed on all included pregnant women without the occurrence of adverse fetal or maternal events. In the context of the present study, CEUS helped to safely differentiate hepatic metastases, focal nodular hyperplasia, atypical hemangioma, hepatic arteriovenous malformation and cystic echinococcosis. In two patients CEUS determined immediate therapy. Conclusion Although not clinically approved in obstetrics so far, CEUS is a safe imaging modality which, in addition to B-mode and color Doppler ultrasonography, may be applied during pregnancy for further medical indications and to provide helpful information. Hintergrund Ultraschall ist ein unentbehrliches bildgebendes Verfahren, um den physiologischen und pathologischen Zustand bei Mutter und Fötus in der Schwangerschaft zu beurteilen. Die Publikationen bezüglich des Einsatzes von CEUS in der Schwangerschaft sind begrenzt. Eine sichere Anwendung von CEUS erfolgte bereits, um den uteroplazentaren Blutfluss, die „Cesearen Scar Pregnancy“ und die invasive Plazenta percreta zu beurteilen. CT- und MRT-Aufnahmen in der Schwangerschaft müssen aufgrund der schädlichen ionisierenden Strahlung und wegen der Anreicherung von Gadolinium im Gehirn sorgfältig abgeschätzt werden. Ziel Ziel dieser retrospektiven Single-Center-Studie ist die Bewertung der diagnostischen Leistung und Sicherheit von CEUS in der Schwangerschaft im Hinblick auf die Beurteilung von Leberläsionen unbekannter Entität. Methoden In diese Studie wurden 6 Schwangere (Durchschnittsalter: 32 Jahre; mittlere SSW: 28 Wochen), bei denen von 2005 bis 2014 CEUS durchgeführt wurde, eingeschlossen. Das verabreichte Kontrastmittel war ein Blutpool-Kontrastmittel der zweiten Generation (SonoVue^®, Bracco, Mailand, Italien). Die CEUS-Untersuchungen wurden von einem seit dem Jahr 2000 erfahrenen Radiologen (EFSUMB Level 3) durchgeführt und interpretiert. Ergebnisse CEUS wurde bei allen schwangeren Frauen sicher durchgeführt, ohne dass unerwünschte fetale oder maternale Nebenwirkungen auftraten. Im Rahmen der vorliegenden Studie diente CEUS der sicheren Differenzierung von Lebermetastasen, fokaler nodulärer Hyperplasie, atypischem Hämangiom, arteriovenöser Malformation und zystischer Echinokokkose. Bei 2 Patientinnen bewirkte CEUS eine sofortige Therapie. Schlussfolgerung Obwohl CEUS in der Geburtshilfe bisher nicht klinisch zugelassen ist, handelt es sich um eine sichere Bildgebungsmethode, die in der Schwangerschaft bei weiteren medizinischen Indikationen zusätzlich zur nativen B-Mode Sonographie und Farbdoppler-Sonographie nützliche Informationen liefern kann. ^* Co-Letztautorenschaft.

Background: To evaluate the diagnostic accuracy of quantitative perfusion parameters in contrast-enhanced ultrasound to differentiate malignant from benign liver lesions. Methods: In this retrospective study 134 patients with a total of 139 focal liver lesions were included who underwent contrast enhanced ultrasound (CEUS) between 2008 and 2018. All examinations were performed by a single radiologist with more than 15 years of experience using a second-generation blood pool contrast agent. The standard of reference was histopathology (n = 60), MRI or CT (n = 75) or long-term CEUS follow up (n = 4). For post processing regions of interests were drawn both inside of target lesions and the liver background. Time–intensity curves were fitted to the CEUS DICOM dataset and the rise time (RT) of contrast enhancement until peak enhancement, and a late-phase ratio (LPR) of signal intensities within the lesion and the background tissue, were calculated and compared between malignant and benign liver lesion using Student’s t-test. Quantitative parameters were evaluated with respect to their diagnostic accuracy using receiver operator characteristic curves. Both features were then combined in a logistic regression model and the cumulated accuracy was assessed. Results: RT of benign lesions (14.8 ± 13.8 s, p = 0.005), and in a subgroup analysis, particular hemangiomas (23.4 ± 16.2 s, p < 0.001) differed significantly to malignant lesions (9.3 ± 3.8 s). The LPR was significantly different between benign (1.59 ± 1.59, p < 0.001) and malignant lesions (0.38 ± 0.23). Logistic regression analysis with RT and LPR combined showed a high diagnostic accuracy of quantitative CEUS parameters with areas under the curve of 0.923 (benign vs. malignant) and 0.929 (hemangioma vs. malignant. Conclusions: Quantified CEUS parameters are helpful to differentiate malignant from benign liver lesions, in particular in case of atypical hemangiomas.

BACKGROUND: Frequently focal splenic lesions (FSL) - cysts, hemangioma, hamartoma, metastases or infarction amongst others – are incidentally found within the scope of the sonographic examination of the abdomen. By using native B-mode and Color Doppl

Background and objectives: Native ultrasound is the most common imaging modality in obstetrics. The use of contrast-enhanced ultrasound (CEUS) during pregnancy has not been officially approved by leading societies for obstetrics and ultrasound. The present study aims to monitor the safety and diagnostic performance of CEUS for assessing abdominal issues in five pregnant women. Materials and Methods: Five pregnant patients who underwent a total of 11 CEUS examinations between June 2020 and October 2020 were included (mean age: 34 years; mean time of pregnancy: 21 weeks). All CEUS scans were interpreted by one experienced consultant radiologist (EFSUMB Level 3). Results: Upon contrast application, no maternal nor fetal adverse effects were observed. Moreover, no fetal contrast enhancement was observed in any patient. CEUS helped to diagnose renal angiomyolipoma, pyelonephritis, necrotic uterine fibroid, gallbladder polyp, and superior mesenteric vein thrombosis. Conclusions: In our study, off-label use of CEUS showed an excellent safety profile allowing the avoidance of ionizing radiation exposure as well as contrast agents in case of CT or use of gadolinium-based contrast agents in case of MRI. CEUS is a promising diagnostic instrument for facilitating clinical decision-making and improving the management of pregnant women.