This paper studies the effectiveness of accomplishing high-level tasks with a minimum of manual annotation and good feature representations for medical images. In medical image analysis, objects like cells are characterized by significant clinical features. Previously developed features like SIFT and HARR are unable to comprehensively represent such objects. Therefore, feature representation is especially important. In this paper, we study automatic extraction of feature representation through deep learning (DNN). Furthermore, detailed annotation of objects is often an ambiguous and challenging task. We use multiple instance learning (MIL) framework in classification training with deep learning features. Several interesting conclusions can be drawn from our work: (1) automatic feature learning outperforms manual feature; (2) the unsupervised approach can achieve performance that's close to fully supervised approach (93.56%) vs. (94.52%); and (3) the MIL performance of coarse label (96.30%) outweighs the supervised performance of fine label (95.40%) in supervised deep learning features.

The one-step hydrogenolysis of biomass-derived glycerol to propanols (1-propanol + 2-propanol), which are known as biopropanols, was investigated over different supported Pt-H 4 SiW 12 O 40 (HSiW) bi-functional catalysts in aqueous media. Among the catalysts/supports tested, Pt-HSiW supported over ZrO 2 converted glycerol to biopropanols with high selectivity and high yield (94.1%), while exhibiting long-term stability (160 h). In addition, this catalyst can be resistant to the impurities present in crude glycerol. The reaction pathway to propanols from glycerol is proposed to proceed mainly through 1,2-propanediol. With the strategy toward one-step hydrogenolysis of glycerol to biopropanols sustainably, the biomass can be readily transformed to biodiesel and biopropanols via glycerol, which will bring about the benign development of the biodiesel industry. † Electronic supplementary information (ESI) available. See

No abstract

No abstract