Multiple instance learning for eosinophil quantification of sinonasal histopathology images: A hierarchical determination on whole slide images.

KEY POINTS: We proposed a hierarchical framework including an unsupervised candidate image selection and a weakly supervised patch image detection based on multiple instance learning (MIL) to effectively estimate eosinophil quantities in tissue samples from whole slide images. MIL is an innovative approach that can help deal with the variability in cell distribution detection and enable automated eosinophil quantification from sinonasal histopathological images with a high degree of accuracy. The study lays the foundation for further research and development in the field of automated histopathological image analysis, and validation on more extensive and diverse datasets will contribute to real-world application.

ABC-Norm Regularization for Fine-Grained and Long-Tailed Image Classification.

Image classification for real-world applications often involves complicated data distributions such as fine-grained and long-tailed. To address the two challenging issues simultaneously, we propose a new regularization technique that yields an adversarial loss to strengthen the model learning. Specifically, for each training batch, we construct an adaptive batch prediction (ABP) matrix and establish its corresponding adaptive batch confusion norm (ABC-Norm). The ABP matrix is a composition of two parts, including an adaptive component to class-wise encode the imbalanced data distribution, and the other component to batch-wise assess the softmax predictions. The ABC-Norm leads to a norm-based regularization loss, which can be theoretically shown to be an upper bound for an objective function closely related to rank minimization. By coupling with the conventional cross-entropy loss, the ABC-Norm regularization could introduce adaptive classification confusion and thus trigger adversarial learning to improve the effectiveness of model learning. Different from most of state-of-the-art techniques in solving either fine-grained or long-tailed problems, our method is characterized with its simple and efficient design, and most distinctively, provides a unified solution. In the experiments, we compare ABC-Norm with relevant techniques and demonstrate its efficacy on several benchmark datasets, including (CUB-LT, iNaturalist2018); (CUB, CAR, AIR); and (ImageNet-LT), which respectively correspond to the real-world, fine-grained, and long-tailed scenarios.

Hierarchical synthesis of silver nanoparticles and wires by copolymer templates and visible light.

Self-assembled silver wires in micro-meter scale were obtained from aqueous silver nitrate solution in the presence of a comb-like copolymer as the sole organic component. The requisite copolymer was easily prepared by the grafting poly(oxyethylene)-monoamine (POE-amine) onto poly(styrene-co-maleic anhydride) (SMA). Upon storage at ambient temperature with exposure to daylight, the aqueous AgNO(3)/SMA-POE solution gradually underwent a color changed from transparent pale-yellow to dark-violet over a period of hours, and after several months a solid precipitate was deposited. The formation process was monitored by ultraviolet-visible spectrometer, particle size analysis, scanning electron microscope, and transmission electron microscope. Silver wires were hierarchically formed by progressive transformation from the initial appearance of silver nanoparticles (ca. 10nm in diameter), followed by the intermediate rectangles (0.6-1.0µm in width and 0.4µm in length) in solution and ultimately the precipitates in micro-scale of silver wires at 1.6-6.4µm in diameter and 100-370µm in length. The progressive formation of the precipitated silver wires was accelerated by the exposure of visible light as a photo-reducing energy source. The micron-scale wires have a silver content over 97.4wt.% and a sheet resistance of 5.5×10(1)Ω/square.

Temperature and pH-responsive properties of poly(styrene-co-maleic anhydride)-grafting poly(oxypropylene)-amines.

A series of temperature- and pH-responsive copolymers were prepared by grafting poly(oxypropylene)-diamines (POP-amine) onto poly(styrene-co-maleic anhydride) (SMA) copolymers. Owing to the presence of poly(oxypropylene)-segments in the pendants and their expressive hydrogen bonding properties, the copolymers exhibited lower critical solution temperature (LCST) behavior in water with a range of 11-49 degrees C depending on the POP-segmental weight and the SMA backbone structure. Furthermore, the co-existence of -NH(2) termini in the pendants and amidoacids in the linking sites in the structures rendered the copolymers also responsive to environmental pH changes. Structural variations in the SMA backbones, particularly the ratio of styrene/maleic anhydride monomers, POP lengths, and amine functionalities largely influenced the copolymer aggregation in the medium with different pH or thermal conditions. The responsiveness was correlated to the surface morphologies observed by tapping/phase mode atomic force microscopy (TM-AFM) on Mica film surface and also fluorescent properties in aqueous solution.