vances in targeted drug delivery applications. Several classes Protein-Polymer Nanoparticles for Delivery of nanomaterials, including gold shells, carbon nanotubes, of Therapeutics dendrimers, lipids, and polymers, are being investigated as matrices for developing multifunctional nanoparticles that not Ashutosh Chilkoti only carry the drug payload but also contain the targeting and imaging moieties. A myriad of specific cell-targeting ap-Center for Biologically Inspired Materials and Materials proaches are being used, which includes not only antibodies Systems and Department of Biomedical Engineering, but also small molecular ligands, peptides, and other biologics. Duke University, Durham, NC, USAWe have been investigating delivery of genes and small inter-I will, in my first example, describe a class of proteinfering RNAs (siRNAs) using the polycationic biodegradable polymers-artificial recombinant polypeptides-that spontapolymer, chitosan, and its derivatives. Examples of successful neously undergo self-assembly upon conjugation to the cancer applications of combined genomics and nanotechnology for chemotherapeutic doxorubicin (Dox) and other small hydrocancers, with a particular focus on delivery of polynucleotides phobic molecules. These chimeric polypeptides (CPs) consist to prostate, lung, and ovarian tumors, will be discussed. of a hydrophilic, biodegradable polypeptide segment that is attached to a short Cys-rich segment. Covalent modification of the Cys residues with a structurally diverse set of hydrophobic Cell-Mediated Drug Delivery to the Lungs small molecules, including Dox, leads to the spontaneous forby SNAP Methodology mation of nanoparticles for a range of CP compositions and molecular weights. The CP-Dox nanoparticles are ϳ40 nm in Don F. Cameron diameter, release drug at pH 5.0 (relevant to endolysosomal release), are taken up by cells, show subsequent localization Department of Pathology & Cell Biology, University of the drug to the nucleus, and are cytotoxic. Notably the CPof South Florida College of Medicine, Tampa, FL, USA Dox nanoparticles have a fourfold higher maximum tolerated dose than free drug and induce near complete tumor regression Chitosan nanoparticles are effectively used to deliver drugs in a murine model following a single dose. In the second exto the lungs by inhalation methodology for asthma and COPD. ample, I will discuss a new protein-polymer conjugate with However, delivery of drug-containing nanoparticles to the interesting pharmacological properties. I will describe two new deep lung in this way is poor (ϳ10-30%) and difficult to conand general routes to grow a PEG-like polymer, poly[oligo trol, especially when there are obstructive airway issues. To (ethylene glycol) methyl ether methacrylate] [poly(OEGMA)], overcome these limitations, isolated and prelabeled rat Sertoli with low polydispersity and high yield solely from the Ncells (SCs) were preloaded with labeled chitosan nanoparticles terminus or C-terminus of a protein by in situ atom-transfer that contained the ...

This paper introduces a method to use deep convolutional neural networks (CNNs) to automatically replace advertisement (AD) photo on social (or self-media) videos and provides the suitable evaluation method to compare different CNNs. An AD photo can replace a picture inside a video. However, if a human being occludes the replaced picture in the original video, the newly pasted AD photo will block the human occluded part. The deep learning algorithm is implemented to segment the human being from the video. The segmented human pixels are then pasted back to the occluded area, so that the AD photo replacement becomes natural and perfect appearance in the video. This process requires the predicted occlusion edge to be closed to the ground truth occlusion edge, so that the AD photo can be occluded naturally. Therefore, this research introduces a curve fitting method to measure the predicted occlusion edge’s error. By using this method, three CNN methods are applied and compared for the AD replacement. They are mask of regions convolutional neural network (Mask RCNN), a recurrent network for video object segmentation (ROVS) and DeeplabV3. The experimental results show the comparative segmentation accuracy of the different models and DeeplabV3 shows the best performance.