A Ranking Based Approach for Robust Object Discovery from Images of Mixed Classes

2017 ◽  
pp. 71-83 ◽  
Author(s):  
Min Ge ◽  
Chenyi Zhuang ◽  
Qiang Ma
Keyword(s):  
Object Discovery ◽  
Mixed Classes

scholarly journals Advances in Membranous Nephropathy

2021 ◽  
Vol 10 (4) ◽  
pp. 607
Author(s):  
Pierre Ronco ◽  
Emmanuelle Plaisier ◽  
Hanna Debiec
Keyword(s):  
Membranous Nephropathy ◽  
Calcineurin Inhibitor ◽  
Sequential Therapy ◽  
Randomized Controlled ◽  
Therapeutic Algorithms ◽  
Long Time ◽  
Substantial Progress ◽  
Thrombotic Complications ◽  
Mixed Classes ◽  
End Stage

Membranous nephropathy (MN) is a rare auto-immune disease where the glomerulus is targeted by circulating auto-antibodies mostly against podocyte antigens, which results in the formation of electron-dense immune complexes, activation of complement and massive proteinuria. MN is the most common cause of nephrotic syndrome in adults leading to severe thrombotic complications and kidney failure. This review is focused on the recent therapeutic and pathophysiological advances that occurred in the last two years. For a long time, we were lacking a head-to-head comparison between cyclophosphamide considered as the gold standard therapy and other medications, notably rituximab. Substantial progress has been achieved owing to three randomized controlled trials. MENTOR (Membranous Nephropathy Trial of Rituximab) and STARMEN (Sequential Therapy with Tacrolimus and Rituximab in Primary Membranous Nephropathy) conclusively established that calcineurin inhibitor-based regimens are slower to result in an immunologic response than rituximab or cyclophosphamide, achieve fewer complete clinical remissions, and are less likely to maintainremission. Rituximab Versus Steroids and Cyclophosphamide in the Treatment of Idiopathic Membranous Nephropathy (RI-CYCLO) suggested that competition between cyclophosphamide and rituximab remains open. Given the technological leap combining laser microdissection of glomeruli and mass spectrometry of solubilized digested proteins, four “new antigens” were discovered including NELL-1 and Semaphorin 3B in so-called primary MN, and exostosins 1 and 2 and NCAM 1 in lupus MN. NELL-1 is associated with about 8% of primary MN and is characterized by segmental immune deposits and frequent association with cancer (30%). Semaphorin 3B-associated MN usually occurs in children, often below the age of two years, where it is the main antigen, representing about 16% of non-lupus MN in childhood. Exostosins 1/2 and NCAM 1 are associated with 30% and 6% of lupus MN, respectively. Exostosins 1/2 (EXT1/2) staining is associated with a low rate of end-stage kidney disease (ESKD) even in mixed classes III/IV+V. These findings already lead to revisiting the diagnostic and therapeutic algorithms toward more personalized medicine.

scholarly journals Exploiting domain knowledge for Object Discovery

2013 ◽  
Author(s):  
Alvaro Collet ◽  
Bo Xiong ◽  
Corina Gurau ◽  
Martial Hebert ◽  
Siddhartha S. Srinivasa
Keyword(s):  
Domain Knowledge ◽  
Object Discovery

scholarly journals Unsupervised Learning from Videos for Object Discovery in Single Images

Symmetry ◽  
2020 ◽  
Vol 13 (1) ◽  
pp. 38
Author(s):  
Dong Zhao ◽  
Baoqing Ding ◽  
Yulin Wu ◽  
Lei Chen ◽  
Hongchao Zhou
Keyword(s):  
Input Image ◽  
Frame Structure ◽  
Object Discovery ◽  
Spatial Transformations ◽  
Learning Tasks ◽  
Video Frames ◽  
Segmentation Methods ◽  
Feed Forward Network ◽  
Single Input ◽  
Inter Frame

This paper proposes a method for discovering the primary objects in single images by learning from videos in a purely unsupervised manner—the learning process is based on videos, but the generated network is able to discover objects from a single input image. The rough idea is that an image typically consists of multiple object instances (like the foreground and background) that have spatial transformations across video frames and they can be sparsely represented. By exploring the sparsity representation of a video with a neural network, one may learn the features of each object instance without any labels, which can be used to discover, recognize, or distinguish object instances from a single image. In this paper, we consider a relatively simple scenario, where each image roughly consists of a foreground and a background. Our proposed method is based on encoder-decoder structures to sparsely represent the foreground, background, and segmentation mask, which further reconstruct the original images. We apply the feed-forward network trained from videos for object discovery in single images, which is different from the previous co-segmentation methods that require videos or collections of images as the input for inference. The experimental results on various object segmentation benchmarks demonstrate that the proposed method extracts primary objects accurately and robustly, which suggests that unsupervised image learning tasks can benefit from the sparsity of images and the inter-frame structure of videos.

Joint object discovery and segmentation with image-wise reconstruction error

2016 ◽  
Author(s):  
Shuhei Tarashima ◽  
Jingjing Pan ◽  
Go Irie ◽  
Takayuki Kurozumi ◽  
Tetsuya Kinebuchi
Keyword(s):  
Reconstruction Error ◽  
Object Discovery

scholarly journals Human lipoproteins comprise at least 12 different classes that are 1 lognormally distributed.

2021 ◽  
Author(s):  
Tomokazu Konishi ◽  
Risako Fujiwara ◽  
Tadaaki Saito ◽  
Tadaaki Saito ◽  
Nozomi Satou ◽  
...  
Keyword(s):  
Human Serum ◽  
Separation Method ◽  
High Gravity ◽  
Serum Samples ◽  
Rapid Separation ◽  
Human Serum Samples ◽  
Multiple Factors ◽  
Mixed Classes ◽  
Time Required ◽  
Enzymatic Methods

Lipoproteins in medical samples have been measured by enzymatic methods that coincide with conventional ultracentrifugation. However, the high gravity and time required for ultracentrifugation can cause sample degradation. This study presents the results of HPLC, a gentler and rapid separation method, for 55 human serum samples. The elution patterns were analysed parametrically, and the attribute of each class was confirmed biochemically. Human samples contained 12 classes of lipoproteins, each of which may consist primarily of proteins. There are three classes of VLDLs. The level of each class was distributed lognormally, and the standard amount and the 95% range were estimated. Enzymatic methods measure the levels of several mixed classes. This lognormal character suggests that the levels are controlled by the synergy of multiple factors.

Sign in / Sign up

Export Citation Format

Share Document