Fruit freshness detection based on multi-task convolutional neural network.

Background: Fruit freshness detection by computer vision is essential for many agricultural applications, e.g., automatic harvesting and supply chain monitoring. This paper proposes to use the multi-task learning (MTL) paradigm to build a deep convolutional neural work for fruit freshness detection. Results: We design an MTL model that optimizes the freshness detection (T1) and fruit type classification (T2) tasks in parallel. The model uses a shared CNN (convolutional neural network) subnet and two FC (fully connected) task heads. The shared CNN acts as a feature extraction module and feeds the two task heads with common semantic features. Based on an open fruit image dataset, we conducted a comparative study of MTL and single-task learning (STL) paradigms. The STL models use the same CNN subnet with only one specific task head. In the MTL scenario, the T1 and T2 mean accuracies on the test set are 93.24% and 88.66%, respectively. Meanwhile, for STL, the two accuracies are 92.50% and 87.22%. Statistical tests report significant differences between MTL and STL on T1 and T2 test accuracies. We further investigated the extracted feature vectors (semantic embeddings) from the two STL models. The vectors have an averaged 0.7 cosine similarity on the entire dataset, with most values lying in the 0.6-0.8 range. This indicates a between-task correlation and justifies the effectiveness of the proposed MTL approach. Conclusion: This study proves that MTL exploits the mutual correlation between two or more relevant tasks and can maximally share their underlying feature extraction process. we envision this approach to be extended to other domains that involve multiple interconnected tasks.

Gi-protein-coupled ß 1-adrenergic receptor: re-understanding the selectivity of ß 1-adrenergic receptor to G protein.

ß 1-adrenergic receptor (ß 1-AR), a member in the family of G-protein-coupled receptors, is a transmembrane receptor of great significance in the heart. Physiologically, catecholamines activate ß 1-AR to initiate a positive chronotropic, inotropic, and dromotropic change. It is believed that ß 1-AR couples to Gs protein and transmits the signal through second messenger cAMP. However, increasing research shows that ß 1-AR can also bind with Gi protein in addition to Gs. When ß 1-AR-Gi is biasedly activated, cardioprotective effects are introduced by the activated cGMP-protein kinase G (PKG) pathway and the transactivation of epidermal growth factor receptor (EGFR) pathway. The discovery of ß 1-AR-Gi signaling makes us reconsider the selectivity of G protein with regard to ß 1-AR, which also provides new ideas for the treatment of heart diseases. This review summarizes the discovery of ß 1-AR-Gi pathway, including the evidence that supports ß 1-AR's capability to couple Gi, details of the transduction process and functions of the ß 1-AR-Gi signaling pathway.

Spectroscopic profiling-based geographic herb identification by neural network with random weights.

Daodi medicinal material plays an important role in traditional Chinese medicine (TCM). This study researches and validates the NNRW (neural network with random weights) model on spectroscopic profiling data for geographical origin identification. NNRW is a special neural network model that does not require an iterative training process. It has been proved effective in various resource-limited data-driven applications. However, whether NNRW works for spectroscopic profiling data remains to be explored. In this study, the Raman and UV (ultraviolet) profiling data of 160 radix astragali samples from four geographic regions are trained and evaluated by four classification models, i.e., NNRW, MLP (multi-layer perceptron), SVM (support vector machine), and DTC (decision tree classifier). Their validation accuracies are 96.3%, 98.0%, 98.4%, and 92.8% respectively. The training/fitting times are 0.372 ms (milli-seconds), 57.9 ms, 2.033 ms, and 3.351 ms, respectively. This study shows that NNRW has a significant training time cut while keeping a high prediction accuracy, and it is a promising solution to resource-limited edge computing applications.