Glutathione ligand self-assembly enables luminescence from Au15 nanoclusters for highly sensitive and selective monitoring of blood Pb(II) ions

Lead Pb(II) ions is a cumulative toxicant that impacts several biological systems and poses severe harm to young children. Accurate Pb(II) ions monitoring is thus of paramount importance. Here, we present the synthesis and application of glutathione-capped Au15 nanoclusters (Au15(SG)13) as a luminescence probe for the accurate and selective monitoring of blood Pb(II). The introduction of Pb(II) ions triggers orderly self-assembly of Au15 nanoclusters, resulting in the formation of rigid shell around Au nuclei. This limits the localized vibration of the glutathione ligands and their interaction with water molecules, greatly reducing non-radiative energy loss, and thereby enhancing the photoluminescence signal. Consequently, Au15(SG)13 nanoclusters exhibit high sensitivity for Pb(II) detection. The detection signal displays a linear relationship with Pb(II) over a wide detection range (0-800 μg/L), demonstrating a substantial sensitivity of 35.29 μg/L. Moreover, the developed nanoclusters show superior selectivity for Pb(II) ions, distinguishing them from other prevalent heavy metals. This work pave the way for the development of advanced Pb(II) sensors with high sensitivity and selectivity.

Quality of Answers of Generative Large Language Models Versus Peer Users for Interpreting Laboratory Test Results for Lay Patients: Evaluation Study

BACKGROUND

Although patients have easy access to their electronic health records and laboratory test result data through patient portals, laboratory test results are often confusing and hard to understand. Many patients turn to web-based forums or question-and-answer (Q&A) sites to seek advice from their peers. The quality of answers from social Q&A sites on health-related questions varies significantly, and not all responses are accurate or reliable. Large language models (LLMs) such as ChatGPT have opened a promising avenue for patients to have their questions answered.

OBJECTIVE

We aimed to assess the feasibility of using LLMs to generate relevant, accurate, helpful, and unharmful responses to laboratory test-related questions asked by patients and identify potential issues that can be mitigated using augmentation approaches.

METHODS

We collected laboratory test result-related Q&A data from Yahoo! Answers and selected 53 Q&A pairs for this study. Using the LangChain framework and ChatGPT web portal, we generated responses to the 53 questions from 5 LLMs: GPT-4, GPT-3.5, LLaMA 2, MedAlpaca, and ORCA_mini. We assessed the similarity of their answers using standard Q&A similarity-based evaluation metrics, including Recall-Oriented Understudy for Gisting Evaluation, Bilingual Evaluation Understudy, Metric for Evaluation of Translation With Explicit Ordering, and Bidirectional Encoder Representations from Transformers Score. We used an LLM-based evaluator to judge whether a target model had higher quality in terms of relevance, correctness, helpfulness, and safety than the baseline model. We performed a manual evaluation with medical experts for all the responses to 7 selected questions on the same 4 aspects.

RESULTS

Regarding the similarity of the responses from 4 LLMs; the GPT-4 output was used as the reference answer, the responses from GPT-3.5 were the most similar, followed by those from LLaMA 2, ORCA_mini, and MedAlpaca. Human answers from Yahoo data were scored the lowest and, thus, as the least similar to GPT-4-generated answers. The results of the win rate and medical expert evaluation both showed that GPT-4's responses achieved better scores than all the other LLM responses and human responses on all 4 aspects (relevance, correctness, helpfulness, and safety). LLM responses occasionally also suffered from lack of interpretation in one's medical context, incorrect statements, and lack of references.

CONCLUSIONS

By evaluating LLMs in generating responses to patients' laboratory test result-related questions, we found that, compared to other 4 LLMs and human answers from a Q&A website, GPT-4's responses were more accurate, helpful, relevant, and safer. There were cases in which GPT-4 responses were inaccurate and not individualized. We identified a number of ways to improve the quality of LLM responses, including prompt engineering, prompt augmentation, retrieval-augmented generation, and response evaluation.

[Clinical efficacy of induction chemoimmunotherapy for locally advanced hypopharyngeal carcinoma: a prospective phase Ⅱ study]

Objective: To evaluate the objective response rate (ORR) of induction chemoimmunotherapy with camrelizumab plus TPF (docetaxel, cisplatin, and capecitabine) for locally advanced hypopharyngeal squamous cell carcinoma (LA HSCC) and potential predictive factors for ORR. Methods: A single-center, prospective, phase 2 and single-arm trial was conducted for evaluating antitumor activity of camrelizumab+TPF(docetaxel+cisplatin+capecitabine) for LA HSCC between May 21, 2021 and April 15, 2023, patients admitted to the Eye & ENT Hospital affiliated with Fudan University. The primary endpoint was ORR, and enrolled patients with LA HSCC at T3-4N0-3M0 received induction chemoimmunotherapy for three cycles: camrelizumab 200 mg day 1, docetaxel 75 mg/m2 day 1, cisplatin 25 mg/m2 days 1-3, and capecitabine 800 mg/m2 days 1-14. Patients were assigned to radioimmunotherapy when they had complete response or partial response (PR)>70% (Group A), or assigned to surgery plus adjuvant radiotherapy/chemoradiotherapy when they had PR≤70% (Group B), and the responses were defined by using tumor volume evaluation system. Tumor diameter was also used to assess the treatment responses by Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Use SPSS 23.0 software was used to analyze the data. Results: A total of 51 patients were enrolled who underwent the induced chemoimmunotherapy for three cycles, and all were males, aged 35-69 years old. After three cycles of induction immunochemotherapy, 42 (82.4%) patients existed in Group A (complete response or PR>70%) and 9 patients (17.6%) in Group B (PR≤70%), the ORR was 82.4%. The primary endpoint achieved expected main research objectives. Compared to the patients of Group A, the patients of Group B showed the higher T stage and the larger volume of primary tumor before induced immunochemotherapy, and also had the less regression of tumor volume after induced immunochemotherapy (all P<0.05). The optimal cutoff value of pre-treatment tumor volume for predicting ORR was 39 cm3. The T stage (OR=12.71, 95%CI: 1.4-112.5, P=0.022) and the volume (OR=7.1, 95%CI: 1.4-36.8, P=0.018) of primary tumor were the two main factors affecting ORR rate of induction chemoimmunotherapy. Conclusion: The induction chemoimmunotherapy with camrelizumab plus TPF shows an encouraging antitumor efficacy in LA HSCC.

PubTator 3.0: an AI-powered literature resource for unlocking biomedical knowledge

PubTator 3.0 (https://www.ncbi.nlm.nih.gov/research/pubtator3/) is a biomedical literature resource using state-of-the-art AI techniques to offer semantic and relation searches for key concepts like proteins, genetic variants, diseases and chemicals. It currently provides over one billion entity and relation annotations across approximately 36 million PubMed abstracts and 6 million full-text articles from the PMC open access subset, updated weekly. PubTator 3.0's online interface and API utilize these precomputed entity relations and synonyms to provide advanced search capabilities and enable large-scale analyses, streamlining many complex information needs. We showcase the retrieval quality of PubTator 3.0 using a series of entity pair queries, demonstrating that PubTator 3.0 retrieves a greater number of articles than either PubMed or Google Scholar, with higher precision in the top 20 results. We further show that integrating ChatGPT (GPT-4) with PubTator APIs dramatically improves the factuality and verifiability of its responses. In summary, PubTator 3.0 offers a comprehensive set of features and tools that allow researchers to navigate the ever-expanding wealth of biomedical literature, expediting research and unlocking valuable insights for scientific discovery.

Quality of Answers of Generative Large Language Models vs Peer Patients for Interpreting Lab Test Results for Lay Patients: Evaluation Study

Lab results are often confusing and hard to understand. Large language models (LLMs) such as ChatGPT have opened a promising avenue for patients to get their questions answered. We aim to assess the feasibility of using LLMs to generate relevant, accurate, helpful, and unharmful responses to lab test-related questions asked by patients and to identify potential issues that can be mitigated with augmentation approaches. We first collected lab test results related question and answer data from Yahoo! Answers and selected 53 QA pairs for this study. Using the LangChain framework and ChatGPT web portal, we generated responses to the 53 questions from four LLMs including GPT-4, Meta LLaMA 2, MedAlpaca, and ORCA_mini. We first assessed the similarity of their answers using standard QA similarity-based evaluation metrics including ROUGE, BLEU, METEOR, BERTScore. We also utilized an LLM-based evaluator to judge whether a target model has higher quality in terms of relevance, correctness, helpfulness, and safety than the baseline model. Finally, we performed a manual evaluation with medical experts for all the responses to seven selected questions on the same four aspects. The results of Win Rate and medical expert evaluation both showed that GPT-4's responses achieved better scores than all the other LLM responses and human responses on all four aspects (relevance, correctness, helpfulness, and safety). However, LLM responses occasionally also suffer from a lack of interpretation in one's medical context, incorrect statements, and lack of references. We find that compared to other three LLMs and human answer from the Q&A website, GPT-4's responses are more accurate, helpful, relevant, and safer. However, there are cases which GPT-4 responses are inaccurate and not individualized. We identified a number of ways to improve the quality of LLM responses.

PubTator 3.0: an AI-powered Literature Resource for Unlocking Biomedical Knowledge

PubTator 3.0 (https://www.ncbi.nlm.nih.gov/research/pubtator3/) is a biomedical literature resource using state-of-the-art AI techniques to offer semantic and relation searches for key concepts like proteins, genetic variants, diseases, and chemicals. It currently provides over one billion entity and relation annotations across approximately 36 million PubMed abstracts and 6 million full-text articles from the PMC open access subset, updated weekly. PubTator 3.0's online interface and API utilize these precomputed entity relations and synonyms to provide advanced search capabilities and enable large-scale analyses, streamlining many complex information needs. We showcase the retrieval quality of PubTator 3.0 using a series of entity pair queries, demonstrating that PubTator 3.0 retrieves a greater number of articles than either PubMed or Google Scholar, with higher precision in the top 20 results. We further show that integrating ChatGPT (GPT-4) with PubTator APIs dramatically improves the factuality and verifiability of its responses. In summary, PubTator 3.0 offers a comprehensive set of features and tools that allow researchers to navigate the ever-expanding wealth of biomedical literature, expediting research and unlocking valuable insights for scientific discovery.

Annotation and Information Extraction of Consumer-Friendly Health Articles for Enhancing Laboratory Test Reporting

Viewing laboratory test results is patients' most frequent activity when accessing patient portals, but lab results can be very confusing for patients. Previous research has explored various ways to present lab results, but few have attempted to provide tailored information support based on individual patient's medical context. In this study, we collected and annotated interpretations of textual lab result in 251 health articles about laboratory tests from AHealthyMe.com. Then we evaluated transformer-based language models including BioBERT, ClinicalBERT, RoBERTa, and PubMedBERT for recognizing key terms and their types. Using BioPortal's term search API, we mapped the annotated terms to concepts in major controlled terminologies. Results showed that PubMedBERT achieved the best F1 on both strict and lenient matching criteria. SNOMED CT had the best coverage of the terms, followed by LOINC and ICD-10-CM. This work lays the foundation for enhancing the presentation of lab results in patient portals by providing patients with contextualized interpretations of their lab results and individualized question prompts that they can, in turn, refer to during physician consults.

China Alzheimer's Disease and Neurodegenerative Disorder Research (CANDOR) -A Prospective Cohort Study for Alzheimer's Disease and Vascular Cognitive Impairment

BACKGROUND

Alzheimer's disease (AD) and vascular cognitive impairment (VCI) are the two main causes of dementia. AD and VCI share similar symptoms of cognitive decline and may be attributable to similar risk factors. Establishing a prospective cohort to compare VCI and AD would help to understand vascular risk factors related to dementia.

OBJECTIVES

China Alzheimer's disease and Neurodegenerative Disorder Research (CANDOR) study is a prospective multicenter cohort study. It aims to study the similarities and differences between AD and post stroke cognitive impairment (PSCI) in neuroimaging changes, disease progression, and multiple omics studies.

DESIGN

This is an ongoing study. From July 31, 2019, to August 1, 2022, we recruited 1449 participants with ages between 40 and 100 years. The cohort included three groups: AD group, PSCI group, and normal cognitive (NC) group. Data were collected in face-to-face interviews at baseline, and will be followed up every year for 4 years. The PSCI group had additional follow-ups at 3-month and 6-month after enrollment. Brain Magnetic Resonance Imaging (MRI) included high-resolution sequences for intracranial arteries. Cognitive assessments and follow-up information will be prospectively collected. Biological specimens including blood and urine at baseline were collected and tested.

PARTICIPANTS

The targeted sample size of PSCI group was 500, AD group with 600 and NC group with 2000. There were 1449 participants enrolled. Include 508 participants were in NC group, 387 in AD group and 554 in PSCI group.

MEASUREMENTS

Demographics, clinical parameters, and medical examinations were collected and performed. Cognitive assessment was performed to assess all cognitive domains including memory, language, executive function, and orientation function.

CONCLUSIONS

The CANDOR study is a prospective cohort study. Data from this cohort provide us an opportunity to investigate the contribution of vascular factors to dementia pathogenesis.

BioKG: a comprehensive, large-scale biomedical knowledge graph for AI-powered, data-driven biomedical research

To cope with the rapid growth of scientific publications and data in biomedical research, knowledge graphs (KGs) have emerged as a powerful data structure for integrating large volumes of heterogeneous data to facilitate accurate and efficient information retrieval and automated knowledge discovery (AKD). However, transforming unstructured content from scientific literature into KGs has remained a significant challenge, with previous methods unable to achieve human-level accuracy. In this study, we utilized an information extraction pipeline that won first place in the LitCoin NLP Challenge to construct a largescale KG using all PubMed abstracts. The quality of the large-scale information extraction rivals that of human expert annotations, signaling a new era of automatic, high-quality database construction from literature. Our extracted information markedly surpasses the amount of content in manually curated public databases. To enhance the KG's comprehensiveness, we integrated relation data from 40 public databases and relation information inferred from high-throughput genomics data. The comprehensive KG enabled rigorous performance evaluation of AKD, which was infeasible in previous studies. We designed an interpretable, probabilistic-based inference method to identify indirect causal relations and achieved unprecedented results for drug target identification and drug repurposing. Taking lung cancer as an example, we found that 40% of drug targets reported in literature could have been predicted by our algorithm about 15 years ago in a retrospective study, demonstrating that substantial acceleration in scientific discovery could be achieved through automated hypotheses generation and timely dissemination. A cloud-based platform (https://www.biokde.com) was developed for academic users to freely access this rich structured data and associated tools.

Opportunities and challenges for ChatGPT and large language models in biomedicine and health

ChatGPT has drawn considerable attention from both the general public and domain experts with its remarkable text generation capabilities. This has subsequently led to the emergence of diverse applications in the field of biomedicine and health. In this work, we examine the diverse applications of large language models (LLMs), such as ChatGPT, in biomedicine and health. Specifically, we explore the areas of biomedical information retrieval, question answering, medical text summarization, information extraction and medical education and investigate whether LLMs possess the transformative power to revolutionize these tasks or whether the distinct complexities of biomedical domain presents unique challenges. Following an extensive literature survey, we find that significant advances have been made in the field of text generation tasks, surpassing the previous state-of-the-art methods. For other applications, the advances have been modest. Overall, LLMs have not yet revolutionized biomedicine, but recent rapid progress indicates that such methods hold great potential to provide valuable means for accelerating discovery and improving health. We also find that the use of LLMs, like ChatGPT, in the fields of biomedicine and health entails various risks and challenges, including fabricated information in its generated responses, as well as legal and privacy concerns associated with sensitive patient data. We believe this survey can provide a comprehensive and timely overview to biomedical researchers and healthcare practitioners on the opportunities and challenges associated with using ChatGPT and other LLMs for transforming biomedicine and health.

Opportunities and Challenges for ChatGPT and Large Language Models in Biomedicine and Health

ChatGPT has drawn considerable attention from both the general public and domain experts with its remarkable text generation capabilities. This has subsequently led to the emergence of diverse applications in the field of biomedicine and health. In this work, we examine the diverse applications of large language models (LLMs), such as ChatGPT, in biomedicine and health. Specifically we explore the areas of biomedical information retrieval, question answering, medical text summarization, information extraction, and medical education, and investigate whether LLMs possess the transformative power to revolutionize these tasks or whether the distinct complexities of biomedical domain presents unique challenges. Following an extensive literature survey, we find that significant advances have been made in the field of text generation tasks, surpassing the previous state-of-the-art methods. For other applications, the advances have been modest. Overall, LLMs have not yet revolutionized biomedicine, but recent rapid progress indicates that such methods hold great potential to provide valuable means for accelerating discovery and improving health. We also find that the use of LLMs, like ChatGPT, in the fields of biomedicine and health entails various risks and challenges, including fabricated information in its generated responses, as well as legal and privacy concerns associated with sensitive patient data. We believe this survey can provide a comprehensive and timely overview to biomedical researchers and healthcare practitioners on the opportunities and challenges associated with using ChatGPT and other LLMs for transforming biomedicine and health.

Loading IrOx Clusters on MnO2 Boosts Acidic Water Oxidation via Metal-Support Interaction

Noble metal-based electrocatalysts are crucial for efficient acidic water oxidation to develop green hydrogen energy. However, traditional noble metal catalysts loaded on inactive substrates show limited intrinsic catalytic activity, and their large sizes have compromised the atom efficiency of these noble metals. Herein, IrOx nanoclusters with sizes below 2 nm, displaying high atom-utilization efficiency of Ir species, were supported on a redox-active MnO2 nanosubstrate (IrOx/MnO2) with different phases (α-MnO2, δ-MnO2, and ε-MnO2) to explore the optimal combination. Electrochemical measurements showed that IrOx/ε-MnO2 had excellent OER performance with a low overpotential of 225 mV at 10 mA cm-2 in 0.5 M H2SO4, superior to its counterpart, IrOx/α-MnO2 (242 mV) and IrOx/δ-MnO2 (286 mV). Moreover, it also delivered robust stability with no obvious change in operating potential at 10 mA cm-2 during 50 h of continuous operation. Combining the XPS results and Bader charge analysis, we demonstrated that the strong metal-support interactions of IrOx/ε-MnO2 could effectively regulate the electronic structures of the active Ir atoms and stabilize IrOx nanoclusters on supports to suppress their detachment, resulting in significantly enhanced catalytic activity and stability for acidic OER. DFT calculations further supported that the enhanced catalytic OER performance of IrOx/ε-MnO2 could be ascribed to the appropriate strength of interactions between the active Ir sites and the reaction intermediates of the potential-determining step (*O and *OOH) regulated by the redox-active substrates.

A Multicenter Study on Deep Learning for Glioblastoma Auto-Segmentation with Prior Knowledge in Multimodal Imaging

PURPOSE/OBJECTIVE(S)

A precise radiotherapy plan is required to ensure accurate delineation of gross tumor volumes (GTV) and clinical target volumes (CTV1 and CTV2) of glioblastomas (GBMs). However, traditional manual delineation is labor intensive and highly dependent on oncologists' experience. To construct and evaluate a deep learning-based automatic delineation method using prior knowledge in multimodal medical imaging to automate precise GTV, CTV1 and CTV2 contouring in GBM patients.

MATERIALS/METHODS

We retrospectively collected the CT and MRI scans of 55 eligible patients with histologically proven high-grade glioma (HGG) from an institute, these scans were performed with non-enhanced CT (CT), contrast-enhanced T1-weighted (T1C) and T2-FLAIR (T2F) sequences. We proposed a two-stage automatic segmentation framework (PKMI-Net) for GTV, CTV1 and CTV2 based on deep learning using prior knowledge in multimodal medical imaging, and its segmentation performance was evaluated with dice similarity coefficient (DSC), 95% Harsdorff distance (HD95), average surface distance (ASD) and relative volume difference (RVD). To further investigate the generalizability of our method, we designed and conducted two evaluation strategies (Mix and Cross) on four multicenter datasets (including 55 patients, 37 patients, 21 patients and 35 patients).

RESULTS

The evaluation results with an 11-patient test set from the single institute were summarized in Table 1, the proposed method demonstrated the best accuracy in segmenting, respectively, GTV, CTV1 and CTV, achieving a DSC of 0.94, 0.95 and 0.92; HD95 of 2.07 mm, 1.18 mm and 3.80 mm; ASD of 0.69 mm, 0.39 mm and 1.13 mm and RVE of 5.50%, 3.97% and 9.68%. In the multicenter evaluation, the segmentation performance of our method implemented with the Cross strategy was comparable to that with the Mix strategy, demonstrating that our method had high and stable generalizability across multicenter datasets in automatically segmenting GTV, CTV1 and CTV2.

CONCLUSION

Our proposed method achieved promising results in automatically segmenting gliomas across various datasets, which could improve the quality and efficiency of glioblastoma radiotherapy.

Treatment of Thymoma and Thymic Carcinoma with Proton Beam Therapy: Outcomes from the Proton Collaborative Group Prospective Registry

PURPOSE/OBJECTIVE(S)

Given the generally long natural history of thymic malignancies, proton beam therapy (PBT) is advocated to minimize the risk of long-term toxicities to mediastinal organs. Adverse events (AE) and long-term clinical outcomes for this population have not been well-characterized.

MATERIALS/METHODS

The Proton Collaborative Group registry (NCT01255748), a multi-institutional prospective database of academic and community proton centers in the US, was queried for patients with thymomas and thymic carcinomas treated with PBT. Patients with recurrent/metastatic disease, non-thymic histology, received either prior or palliative radiotherapy (dose < 40 Gy RBE) were excluded. Overall survival (OS) and local control (LC) were estimated using Kaplan-Meier methods.

RESULTS

A total of 97 patients were identified in the PCG registry. After applying relevant exclusion criteria, 70 patients from 12 proton centers treated from 2011-2021 were included for analysis. Median follow-up length was 16 months. Median age was 58.5 years (IQR 46-63), and 60% were female. 81.4% had a diagnosis of thymoma, and 18.6% thymic carcinoma. 59 patients underwent surgical resection. 11 were treated with definitive PBT, of which 5 received concurrent chemotherapy. Median dose was 54 Gy RBE (range 41.4 - 70 Gy RBE), median number of fractions was 30 (range 21 - 38). 73.4% received pencil beam scanning and 23% uniform scanning PBT. Treatment was overall well-tolerated: a single patient developed grade 4 pneumonitis. Grade 3 AEs were seen in 3 patients - dyspnea, anorexia, and heart failure. Highest grade toxicity experienced was grade 2 for 47.1% and grade 1 for 42.9% of patients. 3-year overall survival (OS) was 82.6% for the entire cohort. 3-year OS was 94% for resected/adjuvant cohort and 35.6% in the non-surgical/definitive cohort. 3-year local control (LC) was 91.7% for the entire cohort. By surgery/margin status, 3-year LC was 96.8% in patients with close or negative margins (a single failure in a patient with close margins), whereas 3-year LC was 55.1% for patients with positive margins/unresectable disease.

CONCLUSION

Thymic malignancies treated with PBT appear to have favorable outcomes, especially in the adjuvant setting, in this cohort representing the largest series of such patients.

Combined Analysis of a Phase III Randomized Trial and Phase II Prospective Trial with Blind Control Matching of Patients Receiving Whole-Lung, Low-Dose Radiation for COVID-19: Full Results and Immunologic Correlates of the RESCUE 1-19 Trial

PURPOSE/OBJECTIVE(S)

Whole-lung, low-dose radiation (LD-RT) for COVID-19 requires randomization and biologic correlates to determine causality and mechanism.

MATERIALS/METHODS

A phase III trial randomized COVID-19 patients to physician's choice of drug therapy with or without LD-RT. Primary endpoint was intubation-free survival (IFS). The trial was designed with 80% power (two-sided log rank, alpha 0.05) to detect a hazard ratio of 0.3 after 12 intubation events. Estimating a 25% event rate, the planned sample size was 84 patients plus 25 to account for declining intubation rates and screen failures. Due to hospitalization declines and other barriers, the trial closed prematurely. Available randomized data were analyzed by intention-to-treat and combined with phase II results and immunologic correlates, using one-sided significance and an alpha of 0.1 to inform future trial design.

RESULTS

From Jun 2020-Jun 2022, 14 patients were randomized on a phase III trial. From Apr 2020-Dec 2020, 42 patients were enrolled on a phase II trial and blindly matched to 40 controls from contemporaneous trials. 96 total patients and 193 blood samples were available for analysis. Mean hospital duration with LD-RT was 12.9 vs 15.4 days in controls (p = 0.12). Oxygen flow rate >15 L/min (26% vs 38%, p = .27), high-flow oxygen >30 L/min (24% vs 38%, p = 0.18), non-invasive positive-pressure >60 L/min (9% vs 27%, p = 0.03), and mechanical ventilation (9% vs 24%, p = 0.05) reduced with LD-RT. Mean supplemented oxygen volume was 171,759 vs 547,626 liters in controls, with daily means of 10 vs 23 L/min (p = 0.03). Radiographs worsened in 43% vs 71% of controls (p = 0.03). Arterial blood gas mean P/F ratios improved 22% after LD-RT vs declined 8% in controls (p = 0.12). Mean days febrile were 1.8 vs 2.9 in controls (p = 0.10). Rate of myocardial injury was 47% vs 40% in controls (p = 0.77). Flow cytometry revealed 4-fold and 30-fold larger expansions, respectively, in CD8- and CD4-positive CD3+PD1+Ki67-high proliferating cytotoxic T-cells (300% vs 75% expansion, p = 0.07) and helper T-cells (200% expansion vs 6% contraction, p = 0.03) at day 7. In the randomized cohort, mean oxygen volume fell 75% with LD-RT to 78,336 vs 316,786 liters in controls (p = 0.13), mean flow rates were 5.1 vs 18.4 L/min (p = 0.13), radiographs worsened in 50% vs 100% (p = .17), P/F ratios improved 31% vs declined 68% in controls (p = 0.03), hospital duration was 8.9 vs 11.5 days (p = 0.22), and zero LD-RT patients vs one control intubated.

CONCLUSION

Combined analysis of a phase II/III randomized trial suggests that LD-RT prevents ventilation, reduces supplemental oxygen need, improves clinical course, and enhances immune response. LD-RT may have both immediate direct effects and delayed enhanced immunity in COVID-19. Larger multi-institutional trials are justified.

[Effects of rapid drug sensitivity testing for multidrug-resistant bacteria on the prognosis of patients with severe intra-abdominal infection]

Objective: To examine the clinical value of rapid detection of drug-resistant bacteria by immunochromatography and the effects of rapid detection on the prognosis of patients with severe intra-abdominal infection complicated by carbapenem-resistant Enterobacteriaceae (CRE) bloodstream infection. Methods: This was a retrospective cohort study. We analyzed clinical data of 73 patients with severe abdominal infections with sepsis or septic shock complicated by CRE bloodstream infection admitted to the general surgery department of Jinling Hospital between February 2022 and February 2023. Patients were divided into a colloidal gold immunochromatographic assay (GICA) group (17 patients) and conventional testing group (56 patients) based on whether a GICA for CRE had been performed on the patients' first blood culture sample during the diagnosis and treatment process. There were no statistically significant differences between the GICA and conventional testing groups in age ([55.9±17.3] vs. [47.6±16.4] years), sex ([16 men vs. one woman ] vs. [41 men vs. 15 women]), median Charlson comorbidity index (3.0[2.0,4.0] vs. 3.0[2.0, 4.8]), septic shock (10 vs. 39), or acute kidney injury (8 vs. 40) (all P>0.05). Both groups routinely underwent traditional bacterial identification and drug susceptibility testing. Additionally, patients in the GICA group were tested directly for positive blood cultures using a GICA carbapenemase test kit. The main outcomes were mortality rates on Days 28 and 90 after the first identification of CRE bloodstream infection in both groups. We also compared the microbial clearance rate, duration of hospitalization and intensive care unit stay, and time from onset of CRE bloodstream infection to initiation of targeted and appropriate antibiotics between the two groups. Results: The rate of microbial clearance of bloodstream infection was significantly greater in the GICA group than in the conventional testing group (15/17 vs. 34/56 [60.7%], χ2=4.476, P=0.034), whereas the 28-day mortality tended to be lower in the GICA than conventional testing group [5/17 vs. 44.6% [25/56], χ2=1.250, P=0.264). The 90-day mortality (8/17 vs. 53.6% [30/56], χ2=0.222, P=0.638), median duration of hospitalization (37.0 [18.0, 46.5] days vs. 45.5 [32.2, 64.8] days, Z=-1.867, P=0.062), and median duration of intensive care unit stay (18.0 [6.5, 35.0] days vs. 32.0 [5.0, 51.8] days, Z=-1.251, P=0.209). The median time between the onset of bloodstream infection and administration of antibiotics was 49.0 (38.0, 69.0) hours in the GICA group, which is significantly shorter than the 163.0 (111.8, 190.0) hours in the conventional testing group (Z=-5.731, P<0.001). The median time between the onset of bloodstream infection and administration of appropriate antibiotics was 40.0 (34.0, 80.0) hours in the GICA group, which is shorter than in the conventional testing group (68.0 [38.2, 118.8]) hours; however, this difference is not statistically significant (Z=-1.686, P=0.093). Conclusions: GICA can provide information on carbapenemase- producing pathogens faster than traditional drug sensitivity testing, enabling early administration of the optimal antibiotics. The strategy of 'carbapenemase detection first' for managing bacterial infection has the potential to improve prognosis of patients and reduce mortality rate.

Prediction of Outcomes After Heart Transplantation in Pediatric Patients Using National Registry Data: Evaluation of Machine Learning Approaches

BACKGROUND

The prediction of posttransplant health outcomes for pediatric heart transplantation is critical for risk stratification and high-quality posttransplant care.

OBJECTIVE

The purpose of this study was to examine the use of machine learning (ML) models to predict rejection and mortality for pediatric heart transplant recipients.

METHODS

Various ML models were used to predict rejection and mortality at 1, 3, and 5 years after transplantation in pediatric heart transplant recipients using United Network for Organ Sharing data from 1987 to 2019. The variables used for predicting posttransplant outcomes included donor and recipient as well as medical and social factors. We evaluated 7 ML models-extreme gradient boosting (XGBoost), logistic regression, support vector machine, random forest (RF), stochastic gradient descent, multilayer perceptron, and adaptive boosting (AdaBoost)-as well as a deep learning model with 2 hidden layers with 100 neurons and a rectified linear unit (ReLU) activation function followed by batch normalization for each and a classification head with a softmax activation function. We used 10-fold cross-validation to evaluate model performance. Shapley additive explanations (SHAP) values were calculated to estimate the importance of each variable for prediction.

RESULTS

RF and AdaBoost models were the best-performing algorithms for different prediction windows across outcomes. RF outperformed other ML algorithms in predicting 5 of the 6 outcomes (area under the receiver operating characteristic curve [AUROC] 0.664 and 0.706 for 1-year and 3-year rejection, respectively, and AUROC 0.697, 0.758, and 0.763 for 1-year, 3-year, and 5-year mortality, respectively). AdaBoost achieved the best performance for prediction of 5-year rejection (AUROC 0.705).

CONCLUSIONS

This study demonstrates the comparative utility of ML approaches for modeling posttransplant health outcomes using registry data. ML approaches can identify unique risk factors and their complex relationship with outcomes, thereby identifying patients considered to be at risk and informing the transplant community about the potential of these innovative approaches to improve pediatric care after heart transplantation. Future studies are required to translate the information derived from prediction models to optimize counseling, clinical care, and decision-making within pediatric organ transplant centers.

Towards Interpretable Multimodal Predictive Models for Early Mortality Prediction of Hemorrhagic Stroke Patients

The increasing death rate over the past eight years due to stroke has prompted clinicians to look for data-driven decision aids. Recently, deep-learning-based prediction models trained with fine-grained electronic health record (EHR) data have shown superior promise for health outcome prediction. However, the use of EHR-based deep learning models for hemorrhagic stroke outcome prediction has not been extensively explored. This paper proposes an ensemble deep learning framework to predict early mortality among ICU patients with hemorrhagic stroke. The proposed ensemble model achieved an accuracy of 83%, which was higher than the fusion model and other baseline models (logistic regression, decision tree, random forest, and XGBoost). Moreover, we used SHAP values for interpretation of the ensemble model to identify important features for the prediction. In addition, this paper follows the MINIMAR (MINimum Information for Medical AI Reporting) standard, presenting an important step towards building trust among the AI system and clinicians.

Risk prediction of second primary malignancies in primary colorectal neuroendocrine neoplasms patients: a population-based study

PURPOSE

In this study, we aimed to identify risk factors for developing second primary malignancies (SPMs) in colorectal neuroendocrine neoplasms (NENs) patients and develop a competing-risk nomogram to predict SPMs' probabilities quantitatively.

METHODS

Patients with colorectal NENs were retrospectively collected from the Surveillance, Epidemiology, and End Results (SEER) database during 2000-2013. Potential risk factors for SPMs' occurrence in colorectal NENs' patients were identified by the Fine and Gray's proportional sub-distribution hazards model. Then, a competing-risk nomogram was constructed to quantify SPMs' probabilities. The discriminative abilities and calibrations of this competing-risk nomogram were assessed by the area under the receiver-operating characteristic (ROC) curves (AUC) and calibration curves.

RESULTS

We identified 11,017 colorectal NENs' patients, and randomly divided them into training (n = 7711 patients) and validation (n = 3306 patients) cohorts. In the whole cohort, 12.4% patients (n = 1369) had developed SPMs during the maximum follow-up of approximately 19 years (median 8.9 years). Sex, age, race, primary tumor location, and chemotherapy were identified as risk factors for SPMs' occurrence in colorectal NENs' patients. Such factors were selected to develop a competing-risk nomogram and showed excellent predictive ability for SPMs' occurrence (the 3-, 5-, and 10-year AUC values were 0.631, 0.632, and 0.629 in the training cohort and 0.665, 0.639, 0.624 in the validation cohort, respectively).

CONCLUSIONS

This research identified risk factors for SPMs' occurrence in colorectal NENs' patients. Competing-risk nomogram was constructed and proved to have good performance.

Using Twitter Data Analysis to Understand the Perceptions, Awareness, and Barriers to the Wide Use of Pre-Exposure Prophylaxis in the United States

User-generated social media posts such as tweets can provide insights about the public's perception, cognitive, and behavioral responses to health-related issues. Pre-Exposure Prophylaxis (PrEP) is one of the most effective ways to reduce the risk of HIV infection. However, its utilization is low in the US, especially among populations disproportionately affected by HIV such as the age group of under 24 years old. It is therefore important to understand the barriers to the wider use of PrEP in the US using social media posts. In this study, we collected tweets from Twitter about PrEP in the past 4 years to identify such barriers by first identifying tweets about personal discussions, and then performing textual analysis using word analysis, UMLS semantic type analysis, and topic modeling. We found that the public often discussed advocacy, risks/benefits, access, pricing, insurance coverage, legislation, stigma, health education, and prevention of HIV. This result is consistent with the literature and can help identify strategies for promoting the use of PrEP, especially among young adults.

Motivation to Engage in Aging Research: Are There Typologies and Predictors?

BACKGROUND AND OBJECTIVES

Study recruitment and retention of older adults in research studies is a major challenge. Enhancing understanding of individual differences in motivations to participate, and predictors of motivators, can serve the dual aims of facilitating the recruitment and retention of older adults, benefiting study validity, economy, and power.

RESEARCH DESIGN AND METHODS

Older adults (N = 472) past and potential participants were surveyed about motivations to participate in research, demographic, and individual difference measures (e.g., health status, cognitive difficulties). Latent class and clustering analyses explored motivation typologies, followed by regression models predicting individual motivators and typologies.

RESULTS

Older adults endorsed a diversity of research motivations, some of which could be predicted by individual difference measures (e.g., older participants were more motivated by the desire to learn new technology, participants without a college education were more motivated by financial compensation, and participants with greater self-reported cognitive problems were more likely to participate to gain cognitive benefit). Clustering analysis revealed 4 motivation typologies: brain health advocates, research helpers, fun seekers, and multiple motivation enthusiasts. Cognitive difficulties, age, employment status, and previous participation predicted membership in these categories.

DISCUSSION AND IMPLICATIONS

Results provide an understanding of different participant motivations beyond differences between younger and older adults and begin to identify different classes of older adults motivated to participate in research studies. Results can provide guidance for targeted recruitment and retention strategies based on individual differences in stated or predicted motivations.

Deep learning-based predictions of older adults' adherence to cognitive training to support training efficacy

As the population ages, the number of older adults experiencing mild cognitive impairment (MCI), Alzheimer's disease, and other forms of dementia will increase dramatically over the next few decades. Unfortunately, cognitive changes associated with these conditions threaten independence and quality of life. To address this, researchers have developed promising cognitive training interventions to help prevent or reverse cognitive decline and cognitive impairment. However, the promise of these interventions will not be realized unless older adults regularly engage with them over the long term, and like many health behaviors, adherence to cognitive training interventions can often be poor. To maximize training benefits, it would be useful to be able to predict when adherence lapses for each individual, so that support systems can be personalized to bolster adherence and intervention engagement at optimal time points. The current research uses data from a technology-based cognitive intervention study to recognize patterns in participants' adherence levels and predict their future adherence to the training program. We leveraged the feature learning capabilities of deep neural networks to predict patterns of adherence for a given participant, based on their past behavior. A separate, personalized model was trained for each participant to capture individualistic features of adherence. We posed the adherence prediction as a binary classification problem and exploited multivariate time series analysis using an adaptive window size for model training. Further, data augmentation techniques were used to overcome the challenge of limited training data and enhance the size of the dataset. To the best of our knowledge, this is the first research effort to use advanced machine learning techniques to predict older adults' daily adherence to cognitive training programs. Experimental evaluations corroborated the promise and potential of deep learning models for adherence prediction, which furnished highest mean F-scores of 75.5, 75.5, and 74.6% for the Convolution Neural Network (CNN), Long Short-Term Memory (LSTM) network, and CNN-LSTM models respectively.

Proteome Analysis of Temporomandibular Joint with Disc Displacement

Disc displacement without reduction is a common disorder of the temporomandibular joint, causing clinical symptoms and sometimes condylar degeneration. In some cases, bone regeneration is detected following disc-repositioning procedures. Until now, however, systems-wide knowledge of the protein levels for condylar outcome with disc position is still lacking. Here, we performed comprehensive expression profiling of synovial fluid from 109 patients with disc displacement without reduction using high-resolution data-independent acquisition mass spectrometry and characterized differences in 1,714 proteins. Based on magnetic resonance imaging, samples were divided into groups with versus without condylar absorption and subgroups with versus without new bone. For the proteomic analysis, 32 proteins in groups presented with statistical significance (>2-fold, P < 0.05). Pathways such as response to inorganic substances, blood coagulation, and estrogen signaling were significantly expressed in the group with bone absorption as compared with pathways such as regulation of body fluid levels, vesicle-mediated transport, and focal adhesion, which were enriched in the group without bone absorption. In subgroup analysis, 45 proteins of significant importance (>2-fold, P < 0.05) were associated with pathways including would healing, glycolysis and gluconeogenesis, and amino acid metabolism. Combined with clinical examination, molecules such as acetyl-CoA carboxylase beta (ACACB) and transforming growth factor beta 1 (TGFB1) were related to features such as visual analog scale and maximum interincisal opening (P < 0.05). In addition, 7 proteins were examined by Western blotting, including progesterone immunomodulatory binding factor 1 (PIBF1), histidine-rich glycoprotein (HRG), and protein kinase C and casein kinase substrate in neurons 2 (PACSIN2). In conclusion, this study provides the first proteome analysis of condylar absorption at disc displacement without reduction and postoperative new bone formation after disc reposition. Integrated with clinical data, this analysis provides an important insight into the proteomics of condylar modification at disc position.

[Nur77 promotes invasion and migration of gastric cancer cells through the NF-κB/IL-6 pathway]

OBJECTIVE

To analyze the association of Nur77 with overall survival of gastric cancer patients and investigate the role of Nur77 in invasion and migration of gastric cancer cells.

METHODS

Oncomine database was used to analyze the expression of Nur77 in gastric cancer and gastric mucosa tissues, and the distribution characteristics of Nur77 protein between gastric cancer and normal tissues were compared using Human Protein Atlas. GEPIA2 was used to analyze the relationship of Nur77 expression and the patients' survival. The expression of Nur77 in gastric cancer cell lines GES-1, AGS and MKN-45 were detected by Western blotting. The regulatory interactions between IL-6 and Nur77 were verified by transfecting the cells with specific Nur-77 siRNA and Nur-77-overexpressing plasmid. The changes in migration ability of the cells following Nur-77 knockdown were assessed with scratch assay. The effect of Nur-77 overexpression or IL-6 knockdown, or their combination, on migration and invasion of the gastric cancer cells were examined using Transwell assay. The effect of Nur77 expression level on NF-κB/IL-6 pathway activation was analyzed using Western blotting.

RESULTS

Oncomine database showed that gastric cancer tissues expressed a significantly higher level of Nur77 mRNA than normal tissues (P < 0.05). Nur77 expression was detected mostly in the nucleus, and a high Nur77 expression was associated with a poor survival outcome of the patients (P < 0.05). In gastric cancer cells, the high expression of Nur77 participated in the regulation of IL-6. Nur77 silencing significantly lowered the migration ability of the cells (P < 0.05), and IL-6 silencing significantly attenuated the enhanced migration caused by Nur77 overexpression (P < 0.05). Nur77 participates in the activation of NF-κB/IL-6 signaling pathway by regulating the expression of p-p65, p65, p-Stat3 and Stat3.

CONCLUSION

A high Nur77 expression is strongly correlated with a poor prognosis of gastric cancer patients. Nur77 promotes the invasion and migration of gastric cancer cells possibly by regulating the NF-κB/IL-6 signaling pathway.

A Machine-Learning Based Approach for Predicting Older Adults' Adherence to Technology-Based Cognitive Training

Adequate adherence is a necessary condition for success with any intervention, including for computerized cognitive training designed to mitigate age-related cognitive decline. Tailored prompting systems offer promise for promoting adherence and facilitating intervention success. However, developing adherence support systems capable of just-in-time adaptive reminders requires understanding the factors that predict adherence, particularly an imminent adherence lapse. In this study we built machine learning models to predict participants' adherence at different levels (overall and weekly) using data collected from a previous cognitive training intervention. We then built machine learning models to predict adherence using a variety of baseline measures (demographic, attitudinal, and cognitive ability variables), as well as deep learning models to predict the next week's adherence using variables derived from training interactions in the previous week. Logistic regression models with selected baseline variables were able to predict overall adherence with moderate accuracy (AUROC: 0.71), while some recurrent neural network models were able to predict weekly adherence with high accuracy (AUROC: 0.84-0.86) based on daily interactions. Analysis of the post hoc explanation of machine learning models revealed that general self-efficacy, objective memory measures, and technology self-efficacy were most predictive of participants' overall adherence, while time of training, sessions played, and game outcomes were predictive of the next week's adherence. Machine-learning based approaches revealed that both individual difference characteristics and previous intervention interactions provide useful information for predicting adherence, and these insights can provide initial clues as to who to target with adherence support strategies and when to provide support. This information will inform the development of a technology-based, just-in-time adherence support systems.

Synergistic Effects in N,O-Comodified Carbon Nanotubes Boost Highly Selective Electrochemical Oxygen Reduction to H2 O2

Electrochemical 2-electron oxygen reduction reaction (ORR) is a promising route for renewable and on-site H2 O2 production. Oxygen-rich carbon nanotubes have been demonstrated their high selectivity (≈80%), yet tailoring the composition and structure of carbon nanotubes to further enhance the selectivity and widen working voltage range remains a challenge. Herein, combining formamide condensation coating and mild temperature calcination, a nitrogen and oxygen comodified carbon nanotubes (N,O-CNTs) electrocatalyst is synthesized, which shows excellent selective (>95%) H2 O2 selectivity in a wide voltage range (from 0 to 0.65 V versus reversible hydrogen electrode). It is significantly superior to the corresponding selectivity values of CNTs (≈50% in 0-0.65 V vs RHE) and O-CNTs (≈80% in 0.3-0.65 V vs RHE). Density functional theory calculations revealed that the C neighbouring to N is the active site. Introducing O-related species can strengthen the adsorption of intermediates *OOH, while N-doping can weaken the adsorption of in situ generated *O and optimize the *OOH adsorption energy, thus improving the 2-electron pathway. With optimized N,O-CNTs catalysts, a Janus electrode is designed by adjusting the asymmetric wettability to achieve H2 O2 productivity of 264.8 mol kgcat -1 h-1 .

Multi-label classification for biomedical literature: an overview of the BioCreative VII LitCovid Track for COVID-19 literature topic annotations

The coronavirus disease 2019 (COVID-19) pandemic has been severely impacting global society since December 2019. The related findings such as vaccine and drug development have been reported in biomedical literature-at a rate of about 10 000 articles on COVID-19 per month. Such rapid growth significantly challenges manual curation and interpretation. For instance, LitCovid is a literature database of COVID-19-related articles in PubMed, which has accumulated more than 200 000 articles with millions of accesses each month by users worldwide. One primary curation task is to assign up to eight topics (e.g. Diagnosis and Treatment) to the articles in LitCovid. The annotated topics have been widely used for navigating the COVID literature, rapidly locating articles of interest and other downstream studies. However, annotating the topics has been the bottleneck of manual curation. Despite the continuing advances in biomedical text-mining methods, few have been dedicated to topic annotations in COVID-19 literature. To close the gap, we organized the BioCreative LitCovid track to call for a community effort to tackle automated topic annotation for COVID-19 literature. The BioCreative LitCovid dataset-consisting of over 30 000 articles with manually reviewed topics-was created for training and testing. It is one of the largest multi-label classification datasets in biomedical scientific literature. Nineteen teams worldwide participated and made 80 submissions in total. Most teams used hybrid systems based on transformers. The highest performing submissions achieved 0.8875, 0.9181 and 0.9394 for macro-F1-score, micro-F1-score and instance-based F1-score, respectively. Notably, these scores are substantially higher (e.g. 12%, higher for macro F1-score) than the corresponding scores of the state-of-art multi-label classification method. The level of participation and results demonstrate a successful track and help close the gap between dataset curation and method development. The dataset is publicly available via https://ftp.ncbi.nlm.nih.gov/pub/lu/LitCovid/biocreative/ for benchmarking and further development. Database URL https://ftp.ncbi.nlm.nih.gov/pub/lu/LitCovid/biocreative/.

Pre-trained models, data augmentation, and ensemble learning for biomedical information extraction and document classification

Large volumes of publications are being produced in biomedical sciences nowadays with ever-increasing speed. To deal with the large amount of unstructured text data, effective natural language processing (NLP) methods need to be developed for various tasks such as document classification and information extraction. BioCreative Challenge was established to evaluate the effectiveness of information extraction methods in biomedical domain and facilitate their development as a community-wide effort. In this paper, we summarize our work and what we have learned from the latest round, BioCreative Challenge VII, where we participated in all five tracks. Overall, we found three key components for achieving high performance across a variety of NLP tasks: (1) pre-trained NLP models; (2) data augmentation strategies and (3) ensemble modelling. These three strategies need to be tailored towards the specific tasks at hands to achieve high-performing baseline models, which are usually good enough for practical applications. When further combined with task-specific methods, additional improvements (usually rather small) can be achieved, which might be critical for winning competitions.

Database URL: https://doi.org/10.1093/database/baac066

Iridium Doped Pyrochlore Ruthenates for Efficient and Durable Electrocatalytic Oxygen Evolution in Acidic Media

Developing highly active, durable, and cost-effective electrocatalysts for the oxygen evolution reaction (OER) is of prime importance in proton exchange membrane (PEM) water electrolysis techniques. Ru-based catalysts have high activities but always suffer from severe fading and dissolution issues, which cannot satisfy the stability demand of PEM. Herein, a series of iridium-doped yttrium ruthenates pyrochlore catalysts is developed, which exhibit better activity and much higher durability than commercial RuO₂ , IrO₂ , and most of the reported Ru or Ir-based OER electrocatalysts. Typically, the representative Y₂ Ru_1.2 Ir_0.8 O₇ OER catalyst demands a low overpotential of 220 mV to achieve 10 mA cm^-2 , which is much lower than that of RuO₂ (300 mV) and IrO₂ (350 mV). In addition, the catalyst does not show obvious performance decay or structural degradation over a 2000 h stability test. EXAFS and XPS co-prove the reduced valence state of ruthenium and iridium in pyrochlore contributes to the improved activity and stability. Density functional theory reveals that the potential-determining steps barrier of OOH* formation is greatly depressed through the synergy effect of Ir and Ru sites by balancing the d band center and oxygen intermediates binding ability.

Circular RNA circPFKP suppresses the proliferation and metastasis of gastric cancer cell via sponging miR-644 and regulating ADAMTSL5 expression

The treatment of gastric cancer (GC) is extremely challenging; however, the specific pathogenesis of GC remains unclear. Circular RNAs (CircRNAs) are non-coding RNAs that can regulate gene expression both transcriptionally and post-transcriptionally. However, little is known about the circRNAs that are important in the progression of GC. This study identified significantly dysregulated circRNAs by analyzing gastric cancer patients and normal control tissues. The target gene was predicted using online bioinformatics tools and verified using RNA pull-down and luciferase reporter assays. Quantitative real-time polymerase chain reaction (qRT-PCR) and western blotting were used to evaluate gene and protein expression. The malignant behavior of GC cells was determined using 3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide (MTT) assay, wound healing assay, Transwell invasion assay, and flow cytometry. CircPFKP is downregulated in GC tissues, and overexpression of circPFKP inhibits malignant behavior in GC cells. Bioinformatics predicted that circPFKP could bind to miR-644, and miR-644 could target disintegrin-like and metalloprotease domain-containing thrombospondin type 1 motif-like 5 (ADAMTSL5). Overexpression of circPFKP enhances the expression of ADAMTSL5 by decreasing the expression of miR-644 to suppress the growth of xenograft GC tumors in vivo and in vitro. In conclusion, the circPFKP/miR-644/ADAMTSL5 regulatory pathway inhibited the malignant progression of GC. These findings may extend our understanding of the effects of circRNAs on cancer development and provide novel targets for the diagnosis of GC.

Al3+ Dopants Induced Mg2+ Vacancies Stabilizing Single-Atom Cu Catalyst for Efficient Free-Radical Hydrophosphinylation of Alkenes

Utilizing heterogeneous catalysts to overcome obstacles for homogeneous reactions is fascinating but very challenging owing to the difficult fabrication of such catalysts based on the character of target reactions. Herein, we report a Al³⁺ doping strategy to construct single-atom Cu on MgO nanosheets (Cu₁/MgO(Al)) for boosting the free-radical hydrophosphinylation of alkenes. Al³⁺ dopants in MgO bring about abundant Mg²⁺ vacancies for stabilizing dense independent Cu atoms (6.3 wt %), while aggregated Cu nanoparticles are formed without Al³⁺ dopants (Cu/MgO). Cu₁/MgO(Al) exhibits preeminent activity and durability in the hydrophosphinylation of various alkenes with great anti-Markovnikov selectivity (99%). The turnover frequency (TOF) value reaches up to 1272 h^-1, exceeding those of Cu/MgO by ∼6-fold and of traditional homogeneous catalysts drastically. Further experimental and theoretical studies disclose that the prominent performance of Cu₁/MgO(Al) derives from the accelerated initiating step of phosphinoyl radical triggered by individual Cu atoms.

Temporal Subtyping of Alzheimer's Disease Using Medical Conditions Preceding Alzheimer's Disease Onset in Electronic Health Records

Subtyping of Alzheimer's disease (AD) can facilitate diagnosis, treatment, prognosis and disease management. It can also support the testing of new prevention and treatment strategies through clinical trials. In this study, we employed spectral clustering to cluster 29,922 AD patients in the OneFlorida Data Trust using their longitudinal EHR data of diagnosis and conditions into four subtypes. These subtypes exhibit different patterns of progression of other conditions prior to the first AD diagnosis. In addition, according to the results of various statistical tests, these subtypes are also significantly different with respect to demographics, mortality, and prescription medications after the AD diagnosis. This study could potentially facilitate early detection and personalized treatment of AD as well as data-driven generalizability assessment of clinical trials for AD.

Machine Learning Approaches to Understanding and Predicting Patterns of Adherence

In cognitive training of older adults, adherence is a major challenge, but appropriate just-in-time adaptive interventions can improve adherence. To understand adherence patterns and predictors of adherence lapses, we aggregated data from two previous trials (N &gt; 230) involving home-based cognitive interventions. This dataset, detailing 40,000 intervention interactions, contains information about intervention engagement and measures of objective and subjective cognitive performance, demographics, technology proficiency, and attitudes. Exploratory analyses were conducted to understand patterns and predictors of faltering adherence, using classification models, together with feature selection to remove redundant variables. Adherence behaviors in a week were predictive of quitting the following week. Game parameters such as the time of play were weak indicators of future playing patterns, whereas game success was a strong predictor of adherence. These and other useful observations will be incorporated in the design and development of the smart reminder system to be deployed in the APPT project.

Factors That Motivate Older Adults to Participate in Research: Typologies and Implications

A key challenge for scholars who study aging is identifying a pool of research volunteers willing to participate. Toolkits and strategies acknowledge the differences in recruitment needed for older adults relative to younger adults, but there is little information about variations among older adult research volunteers. Based on a community sample of older adults age 60+, this study evaluates differences across seven specific motivators across three broad categories: values/altruism, personal growth/improvement, and immediate gratification. We then identify and evaluate four typologies of older adult volunteers based on the combinations of motivations the older adults in our sample identify as important to participation in research studies. Based on these analyses, we describe how our results might inform recruitment and retention practices in aging studies. Further, we will discuss how these results will help shape our technology-based reminder system with a greater understanding of motivations.

Production and purification of 2-phenylethanol by Saccharomyces cerevisiae using tobacco waste extract as a substrate

2-phenylethanol (2-PE), which is extracted naturally from plant or biotechnology processing, is widely used in the food and cosmetics industries. Due to the high cost of 2-PE production, the valorization of waste carbon to produce 2-PE has gained increasing attention. Here, 2-PE was produced by Saccharomyces cerevisiae using tobacco waste extract (TWE) as the substrate. Considering the toxicity of nicotine and its inhibition of 2-PE, the tolerance of S. cerevisiae was first evaluated. The results suggested that the production of 2-PE by S. cerevisiae in TWEs could be carried out at 2·0 mg ml-1 nicotine concentrations and may be inhibited by 1·0 mg ml-1 2-PE. Thus, the compounds in the TWEs prepared at different temperatures were detected, and the results revealed that the TWEs prepared at 140°C contained 2·18 mg ml-1 of nicotine, had total sugar concentrations of 26·8 mg ml-1 and were suitable for 2-PE production. Due to feedback regulation, the 2-PE production was only 1·11 mg ml-1 , and the remaining glucose concentration remained at 13·78 mg ml-1 , which indicated insufficient glucose utilization. Then, in situ product recovery was further implemented to remove this inhibition; the glucose utilization (the remaining concentration decreased to 3·64 mg ml-1 ) increased, and the 2-PE production increased to 1·65 mg ml-1 . The 2-PE produced in the fermentation broth was first isolated by elution from the resin with 75% ethanol and then by removing the impurities with 2·5% activated charcoal, and pure 2-PE was identified by gas chromatography mass spectrometry. The results of this study suggest that TWE could be an alternative carbon source for 2-PE production. This could provide an outlet tobacco waste as well as reducing the price of natural 2-PE, although more strategies need to be explored to improve the production yield of 2-PE by using TWE.

Controllable synthesis and electrocatalytic applications of atomically precise gold nanoclusters

Nanoclusters are composed of metal atoms and ligands with sizes up to 2-3 nm. Due to their stability and unique structure, gold nanoclusters with precise atomic numbers have been widely studied. Until now, atomically precise gold nanoclusters have been synthesised by various methods. Common ones include the Brust-Schiffrin method and the size-focusing method. With more detailed research on gold nanoclusters, more novel methods have been adopted to synthesise atomically precise gold nanoclusters, such as anti-galvanic reduction, ligand-exchange reactions from metal nanoclusters, the seed growth method, and so on. Besides, the nanoclusters also have many unique properties in electrochemical catalyses, such as the ORR, OER, etc., which are helpful for the development of the energy and environment. In this review, the synthesis methods and electrochemical applications of atomically accurate gold nanoclusters in recent years are introduced.