To determine the levels of heterosis in F1 hybrids, four current pea (Pisum sativum L.) cultivars from southern Australia were used as female parents and crossed with 18 introduced genotypes. The 22 parents, 72 F1 hybrids and, depending on the environment, either 54 or all 72 F2 families were grown in replicated plots in four environments. Grain yield, total dry matter, harvest index, branches per plant, pods per plant, seeds per pod, hundred seed weight, plant height, onset of flowering and flowering periods were evaluated. For both the F1 and F2 generation, heterosis was determined as the superiority over the mid-parent and also over the better parent. In addition, the superiority over the best commercial cultivar was calculated. Most hybrids were higher yielding than their mid-parent but were less stable in yield across environments. Four F1 hybrids were significantly higher yielding than the best parent, by up to 26%. There were significant correlations between F1 hybrid and mid-parent value for plant height, pods per plant and hundred seed weight but not for yield. Overall, grain yield heterosis was mainly due to more pods per plant in the hybrids. The level of heterosis for yield in a poor yielding environment was higher than that in a high yielding one. Both additive and non-additive gene effects were important in the expression of all studied traits. The average level of heterosis for grain yield and total dry matter in the F2 population was half of that in F1 hybrids. The low level of inbreeding depression from the F1 to the F2 generation suggested that epistatic gene action also contributed to the expression of grain yield. Some F2 populations maintained the high yield levels of the corresponding F1 hybrids.
Rapid advancement in imaging technology generates an enormous amount of heterogeneous medical data for disease diagnosis and rehabilitation process. Radiologists may require related clinical cases from medical archives for analysis and disease diagnosis. It is challenging to retrieve the associated clinical cases automatically, efficiently and accurately from the substantial medical image archive due to diversity in diseases and imaging modalities. We proposed an efficient and accurate approach for medical image modality classification that can used for retrieval of clinical cases from large medical repositories. The proposed approach is developed using transfer learning concept with pre-trained ResNet50 Deep learning model for optimized features extraction followed by linear discriminant analysis classification (TLRN-LDA). Extensive experiments are performed on challenging standard benchmark ImageCLEF-2012 dataset of 31 classes. The developed approach yields improved average classification accuracy of 87.91%, which is higher up-to 10% compared to the state-of-the-art approaches on the same dataset. Moreover, hand-crafted features are extracted for comparison. Performance of TLRN-LDA system demonstrates the effectiveness over state-of-the-art systems. The developed approach may be deployed to diagnostic centers to assist the practitioners for accurate and efficient clinical case retrieval and disease diagnosis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.