Search

Sequential recommendation (SR) predicts the user's future preferences based on the sequence of interactions. Recently, some methods for SR have utilized contrastive learning to incorporate self-supervised signals into SR to alleviate the data sparsity problem. Despite these achievements, they overlook the fact that users' multi-behavior interactions in real-world scenarios (e.g., page view, favorite, add to cart, and purchase). Moreover, they disregard the temporal dependencies in users' preferences and their influence on attribute information, leading to models that struggle to accurately capture users' personalized preferences. Therefore, we propose a multi-behavior collaborative contrastive learning for sequential recommendation model. First, we introduce both user-side and item-side attribute information and design an attribute-weight-enhanced attention in multi-behavior interaction scenarios. It enhances the model's ability to capture user's multi-behavior preferences while considering the influence of attribute information. Second, in order to capture users' fine-grained temporal preferences. We divide the interaction sequences into different time scales based on the users' multi-behavior interaction timestamps. In addition, introduce temporal aware attention to generate temporal embeddings for different time scales and effectively integrate them with the user's multi-behavior embeddings. Finally, we design collaborative contrastive learning, which collaboratively captures users' multi-behavior personalized preferences from both temporal and attribute perspectives. This approach alleviates the issue of data sparsity. We conduct extensive experiments on two datasets to validate the effectiveness and superiority of our model. [ABSTRACT FROM AUTHOR]

In a typical wireless powered cognitive radio network, secondary transmitters (STs) perform energy harvesting when the channel is busy and perform active transmission using the harvested energy when the channel is idle. In order to further enhance the throughput of the network, we enable time-division-multiple-access-based backscatter communication (BackCom) when the channel is busy and non-orthogonal-multiple-access (NOMA)-based active transmission when the channel is idle. We aim to maximize the longterm average sum-throughput of all the STs by allocating each ST's BackCom time, energy harvesting time and transmit power for the NOMA-based active transmission. With consideration of limited battery capacity and time-varying channel, we formulate the problem as a Markov decision process. Both Q-learning and deep Q-learning (DQL) algorithms are proposed to solve the problem to obtain fully online policies. Simulation results show that the proposed DQL algorithm not only efficiently deals with the dynamics of the environment but also improves the average throughput up to 27.5% compared with Q-learning and up to 4 times compared with random policy. [ABSTRACT FROM AUTHOR]

Biopsy is the recommended standard for pathological diagnosis of liver carcinoma. However, this method usually requires sectioning and staining, and well-trained pathologists to interpret tissue images. Here, we utilize Raman spectroscopy to study human hepatic tissue samples, developing and validating a workflow for in vitro and intraoperative pathological diagnosis of liver cancer. We distinguish carcinoma tissues from adjacent non-tumour tissues in a rapid, non-disruptive, and label-free manner by using Raman spectroscopy combined with deep learning, which is validated by tissue metabolomics. This technique allows for detailed pathological identification of the cancer tissues, including subtype, differentiation grade, and tumour stage. 2D/3D Raman images of unprocessed human tissue slices with submicrometric resolution are also acquired based on visualization of molecular composition, which could assist in tumour boundary recognition and clinicopathologic diagnosis. Lastly, the potential for a portable handheld Raman system is illustrated during surgery for real-time intraoperative human liver cancer diagnosis. Biopsy is recommended for definitive diagnosis of liver carcinoma, however, this process requires staining and expert pathologists to confirm diagnosis. Here, the authors employ a portable Raman spectroscopy system combined with deep learning to detect carcinoma from normal tissue in real time. [ABSTRACT FROM AUTHOR]

Background: Capsular contracture (CC) is a complication of breast augmentation that frequently requires revision surgery. The axillary approach reduces the visibility of the postoperative scar. It is unclear whether the previous incision can be used to repair the deformity caused by CC. Methods: This study analyzed 21 patients (42 breasts) with grade III–IV CC during 2012–2017. The mean age of the patients was 32 years (range 23–48). Previous axillary scars were used to expose, and CCs were taken out completely or partially. Breast implants were removed. The dissection was performed with endoscopic assistance, using electrocautery under direct visualization. Results: The mean follow-up period was 13 months (range 6–24 months). The dissection plane was changed to dual plane. Thirty-five CCs were taken out completely. Thirty-eight breast implants taken out remained intact. None of the patients required additional surgery. Conclusion: Endoscopic-assisted treatment may be an effective technique for treating CC and avoiding the additional scar. Level of Evidence IV: This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266. [ABSTRACT FROM AUTHOR]