Refined Voting and Scene Feature Fusion for 3D Object Detection in Point Clouds.

An essential task for 3D visual world understanding is 3D object detection in lidar point clouds. To predict directly bounding box parameters from point clouds, existing voting-based methods use Hough voting to obtain the centroid of each object. However, it may be difficult for the inaccurately voted centers to regress boxes accurately, leading to the generation of redundant bounding boxes. For objects in indoor scenes, there are several co-occurrence patterns for objects in indoor scenes. Concurrently, semantic relations between object layouts and scenes can be used as prior context to guide object detection. We propose a simple, yet effective network, RSFF-Net, which adds refined voting and scene feature fusion for indoor 3D object detection. The RSFF-Net consists of three modules: geometric function, refined voting, and scene constraint. First, a geometric function module is used to capture the geometric features of the nearest object of the voted points. Then, the coarse votes are revoted by a refined voting module, which is based on the fused feature between the coarse votes and geometric features. Finally, a scene constraint module is used to add the association information between candidate objects and scenes. RSFF-Net achieves competitive results on indoor 3D object detection benchmarks: ScanNet V2 and SUN RGB-D.

Modeling Habitat Suitability of Migratory Birds from Remote Sensing Images Using Convolutional Neural Networks.

With the application of various data acquisition devices, a large number of animal movement data can be used to label presence data in remote sensing images and predict species distribution. In this paper, a two-stage classification approach for combining movement data and moderate-resolution remote sensing images was proposed. First, we introduced a new density-based clustering method to identify stopovers from migratory birds’ movement data and generated classification samples based on the clustering result. We split the remote sensing images into 16 × 16 patches and labeled them as positive samples if they have overlap with stopovers. Second, a multi-convolution neural network model is proposed for extracting the features from temperature data and remote sensing images, respectively. Then a Support Vector Machines (SVM) model was used to combine the features together and predict classification results eventually. The experimental analysis was carried out on public Landsat 5 TM images and a GPS dataset was collected on 29 birds over three years. The results indicated that our proposed method outperforms the existing baseline methods and was able to achieve good performance in habitat suitability prediction.

Spheroid-type of AL amyloid deposition associated with colonic adenocarcinoma: A case report with literature review.

We report a colonic adenocarcinoma associated with diffuse submucosal deposition of a peculiar spheroid-type amyloid identified in the colon, terminal ileum, and appendix. A 65-year-old woman with past medical histories of hypertension, and chronic obstructive pulmonary disease, presented to the emergency room with cramping abdominal pain and nausea. A computed tomography (CT) scan of abdomen showed right colonic volvulus. Emergency right hemicolectomy was performed. The specimen showed colonic adenocarcinoma with focal submucosal invasion (pT1) arising from a villotubular adenoma. A diffuse submucosal spheroid-type amyloid deposition (resembling corpora amylacea-like structures with Liesegang ring formation) was identified in the colon, ileum, and appendix. Electron microscopy examination of this unusual spheroidal-type material further confirmed the presence of amyloid fibrils. Analysis by liquid chromatography-mass spectrometry detected AL (lambda) type amyloidosis in this specimen. Tests for monoclonal gammopathy were not performed because patient consent was not obtained. In tissue section evaluation, however, no plasma cell neoplasm was identified. Cases with isolated AL amyloid deposition in the gastrointestinal tract have been reported rarely, and there is no case report of colonic adenocarcinoma associated with primary amyloid deposition in the English literature.

Diagnostic challenges of metastatic spindle cell melanoma on fine-needle aspiration specimens.

BACKGROUND: Spindle cell melanoma is a morphologic variant of melanoma that can be difficult to diagnose on specimens obtained via fine-needle aspiration (FNA). Published cytology studies concerning this entity were based for the most part on small series. In the current study, a large series of metastatic spindle cell melanoma is described and the diagnostic pitfalls present in FNA samples addressed. METHODS: The authors retrospectively reviewed the cytologic features of 81 metastatic spindle cell melanoma specimens obtained from 67 patients. Corresponding primary tumors or metastatic tumors taken elsewhere from the same patient were also evaluated. RESULTS: The cytologic smears were mostly cellular and comprised of predominantly spindle tumor cells that frequently formed cohesive fascicles or whorls intermingled with scattered epithelioid tumor cells. The classic cytologic characteristics of conventional melanoma (predominantly dyshesive cellular distribution, cytoplasmic melanin pigments, intranuclear pseudoinclusions, macronucleoli, and binucleation or multinucleation) were noted infrequently or, if present, were more readily found in coexisting epithelioid cells. Remarkably, 9% of the cases failed to demonstrate any of the above classic characteristics. In addition, spindle cells demonstrated a wide range of cytologic atypia, from deceptively bland cells resembling reactive fibroblasts to those indistinguishable from pleomorphic high-grade sarcomatous neoplasms. When the morphologic features were compared with those of the primary tumor or metastatic melanoma taken elsewhere from the same patient, cell type discrepancy was found in 20% of the cases in that the previous counterparts demonstrated the epithelioid cell type. Spindle cells also tended to lose immunoexpression of melanoma markers. CONCLUSIONS: Spindle cell melanoma infrequently demonstrates the diagnostic cytologic features and immunoreactivity of conventional melanoma. Varying degrees of cytologic atypia and possible cell type differences from the primary counterpart or metastatic melanoma occurring elsewhere are additional sources of diagnostic challenges, especially in the metastatic setting. Familiarity with cytologic features, combined with clinical and immunoperoxidase findings, is required to avoid misinterpretation.