Search

This paper presents a new 3D point cloud classification benchmark data set with over four billion manually labelled points, meant as input for data-hungry (deep) learning methods. We also discuss first submissions to the benchmark that use deep convolutional neural networks (CNNs) as a work horse, which already show remarkable performance improvements over state-of-the-art. CNNs have become the de-facto standard for many tasks in computer vision and machine learning like semantic segmentation or object detection in images, but have no yet led to a true breakthrough for 3D point cloud labelling tasks due to lack of training data. With the massive data set presented in this paper, we aim at closing this data gap to help unleash the full potential of deep learning methods for 3D labelling tasks. Our semantic3D.net data set consists of dense point clouds acquired with static terrestrial laser scanners. It contains 8 semantic classes and covers a wide range of urban outdoor scenes: churches, streets, railroad tracks, squares, villages, soccer fields and castles. We describe our labelling interface and show that our data set provides more dense and complete point clouds with much higher overall number of labelled points compared to those already available to the research community. We further provide baseline method descriptions and comparison between methods submitted to our online system. We hope semantic3D.net will pave the way for deep learning methods in 3D point cloud labelling to learn richer, more general 3D representations, and first submissions after only a few months indicate that this might indeed be the case.

We describe an effective and efficient method for point-wise semantic classification of 3D point clouds. The method can handle unstructured and inhomogeneous point clouds such as those derived from static terrestrial LiDAR or photogammetric reconstruction; and it is computationally efficient, making it possible to process point clouds with many millions of points in a matter of minutes. The key issue, both to cope with strong variations in point density and to bring down computation time, turns out to be careful handling of neighborhood relations. By choosing appropriate definitions of a point’s (multi-scale) neighborhood, we obtain a feature set that is both expressive and fast to compute. We evaluate our classification method both on benchmark data from a mobile mapping platform and on a variety of large, terrestrial laser scans with greatly varying point density. The proposed feature set outperforms the state of the art with respect to per-point classification accuracy, while at the same time being much faster to compute.

Purpose:  To evaluate imaging patterns of a COVID-19 infection of the lungs on chest radiographs and their value in discriminating this infection from other viral pneumonias., Materials and Methods:  All 321 patients who presented with respiratory impairment suspicious for COVID-19 infection between February 3 and May 8, 2020 and who received a chest radiograph were included in this analysis. Imaging findings were classified as typical for COVID-19 (bilateral, peripheral opacifications/consolidations), non-typical (findings consistent with lobar pneumonia), indeterminate (all other distribution patterns of opacifications/consolidations), or none (no opacifications/consolidations). The sensitivity, specificity, as well as positive and negative predictive value for the diagnostic value of the category "typical" were determined. Chi² test was used to compare the pattern distribution between the different types of pneumonia., Results:  Imaging patterns defined as typical for COVID-19 infections were documented in 35/111 (31.5 %) patients with confirmed COVID-19 infection but only in 4/210 (2 %) patients with any other kind of pneumonia, resulting in a sensitivity of 31.5 %, a specificity of 98.1 %, and a positive and negative predictive value of 89.7 % or 73 %, respectively. The sensitivity could be increased to 45.9 % when defining also unilateral, peripheral opacifications/consolidations with no relevant pathology contralaterally as consistent with a COVID-19 infection, while the specificity decreases slightly to 93.3 %. The pattern distribution between COVID-19 patients and those with other types of pneumonia differed significantly (p < 0.0001)., Conclusion:  Although the moderate sensitivity does not allow the meaningful use of chest radiographs as part of primary screening, the specific pattern of findings in a relevant proportion of those affected should be communicated quickly as additional information and trigger appropriate protective measures., Key Points:   · COVID-19 infections show specific X-ray image patterns in 1/3 of patients.. · Bilateral, peripheral opacities and/or consolidations are typical imaging patterns.. · Unilateral, peripheral opacities and/or consolidations should also raise suspicion of COVID-19 infection.., Citation Format: · Kasper J, Decker J, Wiesenreiter K et al. Typical Imaging Patterns in COVID-19 Infections of the Lung on Plain Chest Radiographs to Aid Early Triage. Fortschr Röntgenstr 2021; 193: 1189 - 1196., Competing Interests: The authors declare that they have no conflict of interest., (Thieme. All rights reserved.)

Medical History and Initial Presentation: A 35-year-old patient with a previous history of persistent episodic fever, sore throat, myalgia, and cephalgia presented for evaluation of pancytopenia. He had no recent travel history, except for a stay in Italy 1 year prior to admission and in Spain several years in the past., Diagnostic Workup: Laboratory evaluation confirmed pancytopenia, agranulocytosis, and elevated infection parameters without indicative serological results en par with lymphadenitis colli. Computed tomography scanning revealed cervical lymphadenopathy, hepatosplenomegaly, and colitis with occult perforation of the sigmoid colon. Bone marrow biopsy showed an infiltration of polyclonal plasma cells. Lymph node biopsy was compatible with necrotizing lymphadenitis., Diagnosis: Polymerase chain reaction analysis of a lymph node specimen confirmed the presence of Leishmania species, thereby enabling the diagnosis of visceral Leishmania., Therapy Course: Treatment with liposomal amphotericin B was initiated. Both fever and lymphadenopathy quickly resolved., Conclusion: VL is a clinically pleiotropic, severe disease with fatal outcome if left untreated. It often presents with distinct similarities to hematologic malignancies. Exacerbation can occasionally occur as fulminant macrophage activation syndrome. Disease incidence is globally increasing and has not peaked as yet. A complex interplay between pathogen and the immune system is the key pathophysiological mechanism.

In the presence of matter, there is no fundamental limit preventing confinement of visible light even down to atomic scales. Achieving such confinement and the corresponding resonant intensity enhancement inevitably requires simultaneous control over atomic-scale details of material structures and over the optical modes that such structures support. By means of self-assembly we have obtained side-by-side aligned gold nanorod dimers with robust atomically defined gaps reaching below 0.5 nm. The existence of atomically confined light fields in these gaps is demonstrated by observing extreme Coulomb splitting of corresponding symmetric and antisymmetric dimer eigenmodes of more than 800 meV in white-light scattering experiments. Our results open new perspectives for atomically resolved spectroscopic imaging, deeply nonlinear optics, ultrasensing, cavity optomechanics, as well as for the realization of novel quantum-optical devices.

Objective: The diagnostic value of unenhanced helical computed tomography (CT) for the evaluation of acute flank pain is investigated in a prospective study., Patients and Methods: In 125 patients aged 18-86 years, we performed unenhanced helical CT in addition to abdominal plain film, abdominal ultrasound and urinalysis as a diagnostic measure for acute flank pain. Ureteral calculi were confirmed or, respectively, excluded by retrograde ureteropyelography in 80 cases. In the other cases, diagnosis was verified by clinical course and/or stone asservation., Results: In 91 of 125 patients the flank pain was caused by a ureteral calculus. In 67 of 91 patients with urolithiasis, stones could be collected for analysis. Helical CT was able to precisely identify 90 ureteral calculi. Abdominal plain films led to 8 false-positive and 48 false-negative findings. Thus, sensitivity of plain radiography, ultrasound and urinalysis was 47, 11 and 84% with a specificity 76, 97 and 32%, respectively., Conclusions: Unenhanced helical CT reaches a distinctively increased diagnostic value (sensitivity 99%, specificity 97%) in the evaluation of acute flank pain as compared to plain radiography, ultrasound and urinalysis.

Preoperative embolisation of renal carcinomas has several pros and cons for the patient. The negative aspects can be summarized as 'post-infarction' syndrome'. Radiologically, intrarenal gas formation is always evident. As sterile breakdown products of tumor cell necrosis these have to be interpreted as regular postinterventional findings and not as indicators for infection or even sepsis. Tumor embolisation as a means to reduce surgical difficulties in large hypervascularized renal carcinomas and also as a palliative measure in marked macrohematuria and/or tumor-induced flank pain is thus very conceivable. The best time for the tumor-nephrectomy is the day of embolisation or the first postinterventional day. This makes the nephrectomy easier and prevents the postinfarction syndrome for the patient.