Search

Federated learning is a decentralized privacy-preserving mechanism that allows multiple clients to collaborate without exchanging their datasets. Instead, they jointly train a model by uploading their own gradients. However, recent research has shown that attackers can use clients’ gradients to reconstruct the original training data, compromising the security of federated learning. Thus, there has been an increasing number of studies aiming to protect gradients using different techniques. One common technique is secret sharing. However, it has been shown in previous research that when using secret sharing to protect gradient privacy, the original gradient cannot be reconstructed when one share is lost or a server is damaged, causing federated learning to be interrupted. In this paper, we propose an approach that involves using additive secret sharing for federated learning gradient aggregation, making it difficult for attackers to easily access clients’ original gradients. Additionally, our proposed method ensures that any server damage or loss of gradient shares are unlikely to impact the federated learning operation, within a certain probability. We also added a membership level system, allowing members of varying levels to ultimately obtain models with different accuracy levels.

Objective This research explores the performance of ChatGPT, compared to human doctors, in bilingual, Mandarin Chinese and English, medical specialty exam in Nuclear Medicine in Taiwan. Methods The study employed generative pre-trained transformer (GPT-4) and integrated chain-of-thoughts (COT) method to enhance performance by triggering and explaining the thinking process to answer the question in a coherent and logical manner. Questions from the Taiwanese Nuclear Medicine Specialty Exam served as the basis for testing. The research analyzed the correctness of AI responses in different sections of the exam and explored the influence of question length and language proportion on accuracy. Results AI, especially ChatGPT with COT, exhibited exceptional capabilities in theoretical knowledge, clinical medicine, and handling integrated questions, often surpassing, or matching human doctor performance. However, AI struggled with questions related to medical regulations. The analysis of question length showed that questions within the 109–163 words range yielded the highest accuracy. Moreover, an increase in the proportion of English words in questions improved both AI and human accuracy. Conclusions This research highlights the potential and challenges of AI in the medical field. ChatGPT demonstrates significant competence in various aspects of medical knowledge. However, areas like medical regulations require improvement. The study also suggests that AI may help in evaluating exam question difficulty and maintaining fairness in examinations. These findings shed light on AI role in the medical field, with potential applications in healthcare education, exam preparation, and multilingual environments. Ongoing AI advancements are expected to further enhance AI utility in the medical domain.

Enhancement of malignant cell immunogenicity to relieve immunosuppression of lung cancer microenvironment is essential in lung cancer treatment. In previous study, we have demonstrated that dihydroartemisinin (DHA), a kind of phytopharmaceutical, is effective in inhibiting lung cancer cells and boosting their immunogenicity, while the initial target of DHA's intracellular action is poorly understood. The present in‐depth analysis aims to reveal the influence of DHA on the highly expressed TOM70 in the mitochondrial membrane of lung cancer. The affinity of DHA and TOM70 was analyzed by microscale thermophoresis (MST), pronase stability, and thermal stability. The functions and underlying mechanism were investigated using western blots, qRT‐PCR, flow cytometry, and rescue experiments. TOM70 inhibition resulted in mtDNA damage and translocation to the cytoplasm from mitochondria due to the disruption of mitochondrial homeostasis. Further ex and in vivo findings also showed that the cGAS/STING/NLRP3 signaling pathway was activated by mtDNA and thereby malignant cells underwent pyroptosis, leading to enhanced immunogenicity of lung cancer cells in the presence of DHA. Nevertheless, DHA‐induced mtDNA translocation and cGAS/STING/NLRP3 mobilization were synchronously attenuated when TOM70 was replenished. Finally, DHA was demonstrated to possess potent anti‐lung cancer efficacy in vitro and in vivo. Taken together, these data confirm that TOM70 is an important target for DHA to disturb mitochondria homeostasis, which further activates STING and arouses pyroptosis to strengthen immunogenicity against lung cancer thereupon. The present study provides vital clues for phytomedicine‐mediated anti‐tumor therapy. [ABSTRACT FROM AUTHOR]

Prolamins, proteins derived from plants, have extensive applications in pharmaceutics and food science. Jiuzao is a byproduct of the Baijiu brewing industry, and is a great source of prolamin. Despite its importance, knowledge regarding the extraction techniques and the properties of prolamin derived from Baijiu Jiuzao (PBJ) remains limited. Reverse micelles (RMs) extraction offers an efficient and cost-effective method for purifying proteins. In the present study, prolamin was extracted from Baijiu Jiuzao using RMs extraction and subsequently characterized in terms of its secondary structure, morphology, and particle size distribution. Our findings indicate that the purified prolamin extracted using further RMs extraction possessed higher α-helix content (+13.25%), forming a large-scale protein network, and narrower particle size distributions compared to the crude prolamin obtained by NaOH-ethanol method. This research suggests that RMs extraction has potential applications in extracting prolamin from brewing industry byproducts, offering an environmentally friendly approach to Baijiu Jiuzao recycling. [ABSTRACT FROM AUTHOR]

Long non-coding RNAs (lncRNAs) are transcripts that contain more than 200 nucleotides. Despite their inability to code proteins, multiple studies have identified their important role in human cancer through different mechanisms. LncRNA lysyl oxidase like 1 antisense RNA 1 (LOXL1-AS1), a newly discovered lncRNA located on human chromosome 15q24.1, has recently been shown to be involved in the occurrence and progression of various malignancies, such as colorectal cancer, gastric cancer, hepatocellular carcinoma, prostate cancer, non-small cell lung cancer, ovarian cancer, cervical cancer, breast cancer, glioma, thymic carcinoma, pancreatic carcinoma. LOXL1-AS1 acts as competitive endogenous RNA (ceRNA) and via sponging various miRNAs, including miR-374b-5p, miR-21, miR-423-5p, miR-589-5p, miR-28-5p, miR-324-3p, miR-708-5p, miR-143-3p, miR-18b-5p, miR-761, miR-525-5p, miR-541-3p, miR-let-7a-5p, miR-3128, miR-3614-5p, miR-377-3p and miR-1224-5p to promote tumor cell proliferation, invasion, migration, apoptosis, cell cycle, and epithelial–mesenchymal transformation (EMT). In addition, LOXL1-AS1 is involved in the regulation of P13K/AKT and MAPK signaling pathways. This article reviews the current understanding of the biological function and clinical significance of LOXL1-AS1 in human cancers. These findings suggest that LOXL1-AS1 may be both a reliable biomarker and a potential therapeutic target for cancers. [ABSTRACT FROM AUTHOR]

Interferon-inducible double-stranded RNA-dependent protein kinase (PKR) is one of the key antiviral arms in the innate immune system. The activated PKR performs its antiviral function by inhibiting protein translation and inducing apoptosis. In our previous study, we identified grass carp TARBP2 as an inhibitor of PKR activity, thereby suppressing cell apoptosis. This study aimed to explore the effects of grass carp TARBP2 on PKR activity and cell apoptosis. Grass carp TARBP2 comprises two N-terminal dsRBDs and a C-terminal C4 domain. Subcellular localization analysis conducted in CIK cells revealed that TARBP2-FL (full-length TARBP2), TARBP2-Δ1 (lack of the first dsRBD), and TARBP2-Δ2 (lack of the second dsRBD) are predominantly located in the cytoplasm, while TARBP2-Δ3 (lack of the two dsRBDs) is distributed both in the nucleus and cytoplasm. Colocalization and immunoprecipitation assays confirmed the interaction of TARBP2-FL, TARBP2-Δ1, and TARBP2-Δ2 with PKR, while TARBP2-Δ3 showed no binding. Furthermore, our findings suggested that the inhibitory effect of TARBP2-Δ1 or TARBP2-Δ2 on the PKR-eIF2α pathway is depressed compared to TARBP2-FL. In cell apoptosis assays, it was observed that TARBP2-FL inhibits PKR-mediated cell apoptosis. TARBP2-Δ1 or TARBP2-Δ2 exhibits decreased inhibition to PKR-mediated cell apoptosis, whereas TARBP2-Δ3 nearly completely loses this inhibitory effect. These findings highlight the critical importance of two dsRBDs of TARBP2 in interaction with PKR, as well as in the inhibition of PKR activity, resulting in the suppression of cell apoptosis triggered by prolonged PKR activation., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Pancreatic cutaneous fistula is a complex condition, making it challenging to achieve favorable outcomes with conservative medical treatment. Surgical interventions often entail surgical risks and postoperative complications. Here, we present a case involving endoscopically guided stent placement between the stomach and the fistula. By internalizing the fistula, patients can potentially remove the external drainage tube, offering a novel endoscopic treatment approach for such cases.

Background: The present study aimed to investigate the drug-drug interaction and initial dosage optimization of aripiprazole in patients with schizophrenia based on population pharmacokinetics., Research Design and Methods: A total of 119 patients with schizophrenia treated with aripiprazole were included to build an aripiprazole population pharmacokinetic model using nonlinear mixed effects., Results: The weight and concomitant medication of fluoxetine influenced aripiprazole clearance. Under the same weight, the aripiprazole clearance rates were 0.714:1 in patients with or without fluoxetine, respectively. In addition, without fluoxetine, for the once-daily aripiprazole regimen, dosages of 0.3 and 0.2 mg kg -1 day -1 were recommended for patients with schizophrenia weighing 40-95 and 95-120 kg, respectively, while for the twice-daily aripiprazole regimen, 0.3 mg kg -1 day -1 was recommended for those weighing 40-120 kg. With fluoxetine, for the once-daily aripiprazole regimen, a dosage of 0.2 mg kg -1 day -1 was recommended for patients with schizophrenia weighing 40-120 kg, while for the twice-daily aripiprazole regimen, 0.3 and 0.2 mg kg -1 day -1 were recommended for those weighing 40-60 and 60-120 kg, respectively., Conclusion: This is the first investigation of the effects of fluoxetine on aripiprazole via drug-drug interaction. The optimal aripiprazole initial dosage is recommended in patients with schizophrenia., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Zhang, Jiang, Hu, Zhang, Han, Chen, Bulubu, Dong, Shi, He, Yu, Chen and Wang.)