Generative approach for data augmentation for deep learning-based bone surface segmentation from ultrasound images.

PURPOSE: Precise localization of cystic bone lesions is crucial for osteolytic bone tumor surgery. Recently, there is a move toward ultrasound imaging over plain radiographs (X-rays) for intra-operative navigation due to the radiation-free and cost-effectiveness of the modality. In this process, the intra-operative bone model reconstructed from the segmented ultrasound image is registered with the pre-operative bone model. Deep learning approaches have recently shown remarkable success in bone surface segmentation from ultrasound images. However, to train deep learning models effectively with limited dataset size, data augmentation is essential. This study investigates the applicability of the generative approach for data augmentation as well as identifies standard data augmentation approaches for bone surface segmentation from ultrasound images. METHODS: The generative approach we used in our work is based on Pix2Pix image-to-image translation network. We have proposed a multiple-snapshot approach, which mitigates the uni-modal deterministic output issue in the Pix2Pix network without using any complex architecture and training process. We also identified standard data augmentation approaches necessary for ultrasound bone surface segmentation through experiments. RESULTS: We have evaluated our networks using 800 ultrasound images from trained regions (humerus bone) and 1200 ultrasound images from untrained regions (tibia and femur bones) using four different augmentation approaches. The results show that the generative augmentation approach has a positive impact on accuracy in both trained (+ 4.88%) and untrained regions (+ 25.84%) compared to using only standard augmentations. Moreover, compared to standard augmentation approaches, the addition of the generative augmentation approach also showed a similar trend in both trained (+ 8.74%) and untrained (+ 11.55%) regions. CONCLUSION: Generative approaches are very beneficial for data augmentation, where limited dataset size is prevalent, such as ultrasound bone segmentation. The proposed multiple-snapshot Pix2Pix approach has the potential to generate multimodal images, which enlarges the dataset considerably.

Cisplatin Relocalizes RNA Binding Protein HuR and Enhances the Oncolytic Activity of E4orf6 Deleted Adenovirus.

The combination of adenoviruses and chemotherapy agents is a novel approach for human cancer therapeutics. A meticulous analysis between adenovirus and chemotherapeutic agents can help to design an effective anticancer therapy. Human antigen R (HuR) is an RNA binding protein that binds to the AU-rich element (ARE) of specific mRNA and is involved in the export and stabilization of ARE-mRNA. Our recent report unveiled that the E4orf6 gene deleted oncolytic adenovirus (dl355) replicated for certain types of cancers where ARE-mRNA is stabilized. This study aimed to investigate whether a combined treatment of dl355 and Cis-diamminedichloroplatinum (CDDP) can have a synergistic cell-killing effect on cancer cells. We confirmed the effect of CDDP in nucleocytoplasmic HuR shuttling. In vitro and in vivo experiments showed the enhancement of cancer cell death by apoptosis induction and a significant reduction in tumor growth following combination treatment. These results suggested that combination therapy exerted a synergistic antitumor activity by upregulation of CDDP induced cytoplasmic HuR, which led to ARE mRNA stabilization and increased virus proliferation. Besides, the enhanced cell-killing effect was due to the activation of the intrinsic apoptotic pathway. Therefore, the combined treatment of CDDP and dl355 could represent a rational approach for cancer therapy.

Comparison of cake compositions, pepsin digestibility and amino acids concentration of proteins isolated from black mustard and yellow mustard cakes.

As a byproduct of oil production, black and yellow mustard cakes protein are considered as potential source of plant protein for feed applications to poultry, fish and swine industries. The protein contents in black and yellow mustard cakes were 38.17% and 28.80% and their pepsin digestibility was 80.33% and 77.43%, respectively. The proteins were extracted at different pH and maximum proteins (89.13% of 38.17% and 87.76% of 28.80% respectively) isolated from black and yellow mustard cakes at pH 12. The purity of isolated proteins of black and yellow mustard cakes was 89.83% and 91.12% respectively and their pepsin digestibility was 89.67% and 90.17% respectively which assigned the absence of antinutritional compounds. It was found that essential amino acids isoleucine, lysine, methionine, threonine and tryptophan and non essential amino acids arginine and tyrosine were present in greater concentration in black mustard cake protein whereas other amino acids were higher in yellow mustard cake protein.

Physico-chemical characteristics of gamma-irradiated gelatin.

This article reports the effects of gamma irradiation (dose ranges 0.1-10 kGy from 60Co source) on the characteristics of solid gelatin and the physico-mechanical, microstructural and bioactive properties of the scaffold prepared from irradiated gelatin solution. FTIR, intrinsic viscosity, bloom strength, thermal properties, SEM, tensile properties, water uptake ability and antimicrobial activities of non-irradiated and irradiated solid gelatin and its scaffolds were investigated. The detailed experimental results for the solid gelatin demonstrated that 1 kGy γ-irradiated samples showed higher intrinsic viscosity, enhanced thermal stability and bloom strength than other irradiated samples. Furthermore, the scaffold thus prepared from irradiated and non-irradiated gelatin also revealed that 1 kGy samples showed the highest tensile strength and modulus with good water resistivity than other irradiated and non-irradiated samples. In addition to the physico-mechanical properties, 1 kGy scaffolds have also exhibited the highest resistivity towards microbial growth that can have potentiality as scaffold in biomedical sector. The enhanced functional and bioactive properties at low irradiation doses (1 kGy) may occurred due to an initial breaking of hydrogen bonds of polypeptide chains in gelatin molecules that indicated by the shift of amide A, I and II peaks to higher wave numbers in FTIR. This enhancement resulted probably due to the domination of crosslinking over degradation at 1 kGy. It was also observed that 1 kGy γ-radiation-induced crosslinking has lowered the hydrophilicity by decreasing water uptake and mean pore diameter of the interconnected porous structures of gelatin.