There is much epidemiological evidence that a diet rich in fruits and vegetables could lower the risk of certain cancers. The effect has been attributed, in part, to natural polyphenols. Besides, numerous studies have demonstrated that natural polyphenols could be used for the prevention and treatment of cancer. Potential mechanisms included antioxidant, anti-inflammation as well as the modulation of multiple molecular events involved in carcinogenesis. The current review summarized the anticancer efficacy of major polyphenol classes (flavonoids, phenolic acids, lignans and stilbenes) and discussed the potential mechanisms of action, which were based on epidemiological, in vitro, in vivo and clinical studies within the past five years.

Domain generalization aims to apply knowledge gained from multiple labeled source domains to unseen target domains. The main difficulty comes from the dataset bias: training data and test data have different distributions, and the training set contains heterogeneous samples from different distributions. Let X denote the features, and Y be the class labels. Existing domain generalization methods address the dataset bias problem by learning a domain-invariant representation h(X) that has the same marginal distribution P(h(X)) across multiple source domains. The functional relationship encoded in P(Y |X) is usually assumed to be stable across domains such that P(Y |h(X)) is also invariant. However, it is unclear whether this assumption holds in practical problems. In this paper, we consider the general situation where both P(X) and P(Y |X) can change across all domains. We propose to learn a feature representation which has domain-invariant class conditional distributions P(h(X)|Y ). With the conditional invariant representation, the invariance of the joint distribution P(h(X), Y ) can be guaranteed if the class prior P(Y ) does not change across training and test domains. Extensive experiments on both synthetic and real data demonstrate the effectiveness of the proposed method.

Abstract-Recent successes in learning-based image classification, however, heavily rely on the large number of annotated training samples, which may require considerable human efforts. In this paper, we propose a novel active learning framework, which is capable of building a competitive classifier with optimal feature representation via a limited amount of labeled training instances in an incremental learning manner. Our approach advances the existing active learning methods in two aspects. First, we incorporate deep convolutional neural networks into active learning. Through the properly designed framework, the feature representation and the classifier can be simultaneously updated with progressively annotated informative samples. Second, we present a cost-effective sample selection strategy to improve the classification performance with less manual annotations. Unlike traditional methods focusing on only the uncertain samples of low prediction confidence, we especially discover the large amount of high confidence samples from the unlabeled set for feature learning. Specifically, these high confidence samples are automatically selected and iteratively assigned pseudo-labels. We thus call our framework " Cost-Effective Active Learning" (CEAL) standing for the two advantages. Extensive experiments demonstrate that the proposed CEAL framework can achieve promising results on two challenging image classification datasets, i.e., face recognition on CACD database [1] and object categorization on Caltech-256 [2].

The epidemiological studies have indicated a possible oncostatic property of melatonin on different types of tumors. Besides, experimental studies have documented that melatonin could exert growth inhibition on some human tumor cells in vitro and in animal models. The underlying mechanisms include antioxidant activity, modulation of melatonin receptors MT1 and MT2, stimulation of apoptosis, regulation of pro-survival signaling and tumor metabolism, inhibition on angiogenesis, metastasis, and induction of epigenetic alteration. Melatonin could also be utilized as adjuvant of cancer therapies, through reinforcing the therapeutic effects and reducing the side effects of chemotherapies or radiation. Melatonin could be an excellent candidate for the prevention and treatment of several cancers, such as breast cancer, prostate cancer, gastric cancer and colorectal cancer. This review summarized the anticancer efficacy of melatonin, based on the results of epidemiological,experimental and clinical studies, and special attention was paid to the mechanisms of action.

Insomnia is a serious worldwide health threat, affecting nearly one third of the general population. Melatonin has been reported to improve sleep efficiency and it was found that eating melatonin-rich foods could assist sleep. During the last decades, melatonin has been widely identified and qualified in various foods from fungi to animals and plants. Eggs and fish are higher melatonin-containing food groups in animal foods, whereas in plant foods, nuts are with the highest content of melatonin. Some kinds of mushrooms, cereals and germinated legumes or seeds are also good dietary sources of melatonin. It has been proved that the melatonin concentration in human serum could significantly increase after the consumption of melatonin containing food. Furthermore, studies show that melatonin exhibits many bioactivities, such as antioxidant activity, anti-inflammatory characteristics, boosting immunity, anticancer activity, cardiovascular protection, anti-diabetic, anti-obese, neuroprotective and anti-aging activity. This review summaries the dietary sources and bioactivities of melatonin, with special attention paid to the mechanisms of action.