Search

Abstract The objective of this study is to develop a multimodal neural network (MMNN) model that analyzes clinical variables and MRI images of a soft tissue sarcoma (STS) patient, to predict overall survival and risk of distant metastases. We compare the performance of this MMNN to models based on clinical variables alone, radiomics models, and an unimodal neural network. We include patients aged 18 or older with biopsy-proven STS who underwent primary resection between January 1st, 2005, and December 31st, 2020 with complete outcome data and a pre-treatment MRI with both a T1 post-contrast sequence and a T2 fat-sat sequence available. A total of 9380 MRI slices containing sarcomas from 287 patients are available. Our MMNN accepts the entire 3D sarcoma volume from T1 and T2 MRIs and clinical variables. Gradient blending allows the clinical and image sub-networks to optimally converge without overfitting. Heat maps were generated to visualize the salient image features. Our MMNN outperformed all other models in predicting overall survival and the risk of distant metastases. The C-Index of our MMNN for overall survival is 0.77 and the C-Index for risk of distant metastases is 0.70. The provided heat maps demonstrate areas of sarcomas deemed most salient for predictions. Our multimodal neural network with gradient blending improves predictions of overall survival and risk of distant metastases in patients with soft tissue sarcoma. Future work enabling accurate subtype-specific predictions will likely utilize similar end-to-end multimodal neural network architecture and require prospective curation of high-quality data, the inclusion of genomic data, and the involvement of multiple centers through federated learning.

Suboptimal knowledge of clinical pharmacology, therapeutics, and toxicology (CPT) and poor-quality prescribing are threats to patient safety. Our previous national survey of medical faculty identified limited confidence in medical student graduates’ ability to safely prescribe, as well as an interest in a national prescribing competence assessment. Given the in-person challenges posed by the restrictions related to the COVID-19 pandemic, we aimed to re-evaluate opinions and gauge the interest in e-learning resources and assessments. Using public sources, a sampling frame of medical school leaders from all 17 Canadian medical schools, including deans, vice-deans, and program directors for clerkship, residency, and e-learning, were invited to participate in a cross-sectional survey. Survey questions were finalized after several rounds of testing, and analyses were descriptive. Of 1448 invitations, 411 (28.4%) individuals reviewed the survey, and, among them, 278 (67.6%) completed at least one survey question, with representation from all schools. While more than 90% of respondents agreed that medical students should meet a minimum standard of prescribing competence, only 17 (7.9%) could vouch for their school meeting objectives in CPT, and many had significant concerns about their own or other schools’ recent graduate prescribing abilities. Given the lack of local CPT e-curricula resources, there was strong interest in a national online course and assessment in CPT. Our national survey results suggest an ongoing inadequacy of medical trainees’ prescribing competence, and also provide a strong endorsement for both a national online CPT course and assessment during medical school.

Abstract Importance Large language models (LLMs) like OpenAI’s ChatGPT are powerful generative systems that rapidly synthesize natural language responses. Research on LLMs has revealed their potential and pitfalls, especially in clinical settings. However, the evolving landscape of LLM research in medicine has left several gaps regarding their evaluation, application, and evidence base. Objective This scoping review aims to (1) summarize current research evidence on the accuracy and efficacy of LLMs in medical applications, (2) discuss the ethical, legal, logistical, and socioeconomic implications of LLM use in clinical settings, (3) explore barriers and facilitators to LLM implementation in healthcare, (4) propose a standardized evaluation framework for assessing LLMs’ clinical utility, and (5) identify evidence gaps and propose future research directions for LLMs in clinical applications. Evidence review We screened 4,036 records from MEDLINE, EMBASE, CINAHL, medRxiv, bioRxiv, and arXiv from January 2023 (inception of the search) to June 26, 2023 for English-language papers and analyzed findings from 55 worldwide studies. Quality of evidence was reported based on the Oxford Centre for Evidence-based Medicine recommendations. Findings Our results demonstrate that LLMs show promise in compiling patient notes, assisting patients in navigating the healthcare system, and to some extent, supporting clinical decision-making when combined with human oversight. However, their utilization is limited by biases in training data that may harm patients, the generation of inaccurate but convincing information, and ethical, legal, socioeconomic, and privacy concerns. We also identified a lack of standardized methods for evaluating LLMs’ effectiveness and feasibility. Conclusions and relevance This review thus highlights potential future directions and questions to address these limitations and to further explore LLMs’ potential in enhancing healthcare delivery.

Objectives To evaluate whether a focused, expert medication management intervention is feasible and potentially effective in preventing anticoagulation-related adverse events for patients transitioning from hospital to home.Design Randomised, parallel design.Setting Medical wards at six hospital sites in southern Ontario, Canada.Participants Adults 18 years of age or older being discharged to home on an oral anticoagulant (OAC) to be taken for at least 4 weeks.Interventions Clinical pharmacologist-led intervention, including a detailed discharge medication management plan, a circle of care handover and early postdischarge virtual check-up visits to 1 month with 3-month follow-up. The control group received the usual care.Outcomes measures Primary outcomes were study feasibility outcomes (recruitment, retention and cost per patient). Secondary outcomes included adverse anticoagulant safety events composite, quality of transitional care, quality of life, anticoagulant knowledge, satisfaction with care, problems with medications and health resource utilisation.Results Extensive periods of restriction of recruitment plus difficulties accessing patients at the time of discharge negatively impacted feasibility, especially cost per patient recruited. Of 845 patients screened, 167 were eligible and 56 were randomised. The mean age (±SD) was 71.2±12.5 years, 42.9% females, with two lost to follow-up. Intervention patients were more likely to rate their ability to manage their OAC as improved (17/27 (63.0%) vs 7/22 (31.8%), OR 3.6 (95% CI 1.1 to 12.0)) and their continuity of care as improved (21/27 (77.8%) vs 2/22 (9.1%), OR 35.0 (95% CI 6.3 to 194.2)). Fewer intervention patients were taking one or more inappropriate medications (7 (22.5%) vs 15 (60%), OR 0.19 (95% CI 0.06 to 0.62)).Conclusion This pilot randomised controlled trial suggests that a transitional care intervention at hospital discharge for older adults taking OACs was well received and potentially effective for some surrogate outcomes, but overly costly to proceed to a definitive large trial.Trial registration number NCT02777047.

Electrocardiography (ECG) has emerged as a ubiquitous diagnostic tool for the identification and characterization of diverse cardiovascular pathologies. Wearable health monitoring devices, equipped with on-device biomedical artificial intelligence (AI) processors, have revolutionized the acquisition, analysis, and interpretation of ECG data. However, these systems necessitate AI processors that exhibit flexible configuration, facilitate portability, and demonstrate optimal performance in terms of power consumption and latency for the realization of various functionalities. To address these challenges, this study proposes an instruction-driven convolutional neural network (CNN) processor. This processor incorporates three key features: (1) An instruction-driven CNN processor to support versatile ECG-based application. (2) A Processing element (PE) array design that simultaneously considers parallelism and data reuse. (3) An activation unit based on the CORDIC algorithm, supporting both Tanh and Sigmoid computations. The design has been implemented using 110 nm CMOS process technology, occupying a die area of 1.35 mm2 with 12.94 µW power consumption. It has been demonstrated with two typical ECG AI applications, including two-class (i.e., normal/abnormal) classification and five-class classification. The proposed 1-D CNN algorithm performs with a 97.95% accuracy for the two-class classification and 97.9% for the five-class classification, respectively.

IntroductionSolid organ transplant (SOT) recipients can experience bone loss caused by underlying conditions and the use of immunosuppressants. As a result, SOT recipients are at risk for decreased bone mineral density (BMD) and increased fracture incidences. We propose a network meta-analysis (NMA) that incorporates all available randomized control trial (RCT) data to provide the most comprehensive ranking of anti-osteoporotic interventions according to their ability to decrease fracture incidences and increase BMD in SOT recipients.MethodsWe will search MEDLINE, EMBASE, Web of Science, CINAHL, CENTRAL and CNKI for relevant RCTs that enrolled adult SOT recipients, assessed anti-osteoporotic therapies, and reported relevant outcomes. Title and full-text screening as well as data extraction will be performed in-duplicate. We will report changes in BMD as weighted or standardized mean differences, and fracture incidences as risk ratios. SUCRA scores will be used to provide rankings of interventions, and quality of evidence will be examined using RoB2 and CINeMA.DiscussionsTo our knowledge, this systematic review and NMA will be the most comprehensive quantitative analysis regarding the management of bone loss and fractures in SOT recipients. Our analysis should be able to provide physicians and patients with an up-to-date recommendation for pharmacotherapies in reducing incidences of bone loss and fractures associated with SOT. The findings of the NMA will be disseminated in a peer-reviewed journal.

BackgroundThe systematic review of clinical research papers is a labor-intensive and time-consuming process that often involves the screening of thousands of titles and abstracts. The accuracy and efficiency of this process are critical for the quality of the review and subsequent health care decisions. Traditional methods rely heavily on human reviewers, often requiring a significant investment of time and resources. ObjectiveThis study aims to assess the performance of the OpenAI generative pretrained transformer (GPT) and GPT-4 application programming interfaces (APIs) in accurately and efficiently identifying relevant titles and abstracts from real-world clinical review data sets and comparing their performance against ground truth labeling by 2 independent human reviewers. MethodsWe introduce a novel workflow using the Chat GPT and GPT-4 APIs for screening titles and abstracts in clinical reviews. A Python script was created to make calls to the API with the screening criteria in natural language and a corpus of title and abstract data sets filtered by a minimum of 2 human reviewers. We compared the performance of our model against human-reviewed papers across 6 review papers, screening over 24,000 titles and abstracts. ResultsOur results show an accuracy of 0.91, a macro F1-score of 0.60, a sensitivity of excluded papers of 0.91, and a sensitivity of included papers of 0.76. The interrater variability between 2 independent human screeners was κ=0.46, and the prevalence and bias-adjusted κ between our proposed methods and the consensus-based human decisions was κ=0.96. On a randomly selected subset of papers, the GPT models demonstrated the ability to provide reasoning for their decisions and corrected their initial decisions upon being asked to explain their reasoning for incorrect classifications. ConclusionsLarge language models have the potential to streamline the clinical review process, save valuable time and effort for researchers, and contribute to the overall quality of clinical reviews. By prioritizing the workflow and acting as an aid rather than a replacement for researchers and reviewers, models such as GPT-4 can enhance efficiency and lead to more accurate and reliable conclusions in medical research.

Synthetic Aperture Radar (SAR) images are widely utilized in the field of remote sensing. However, there is a limited body of literature specifically addressing the compression of SAR learning images. To address the escalating volume of SAR image data for storage and transmission, which necessitates more effective compression algorithms, this paper proposes a novel framework for compressing SAR images. Experimental validation is performed using a representative low-resolution Sentinel-1 dataset and the high-resolution QiLu-1 dataset. Initially, we introduce a novel two-stage transformation-based approach aimed at suppressing the low-frequency components of the input data, thereby achieving a high information entropy and minimizing quantization losses. Subsequently, a quality map guidance image compression algorithm is introduced, involving the fusion of the input SAR images with a target-aware map. This fusion involves convolutional transformations to generate a compact latent representation, effectively exploring redundancies between focused and non-focused areas. To assess the algorithm’s performance, experiments are carried out on both the low-resolution Sentinel-1 dataset and the high-resolution QiLu-1 dataset. The results indicate that the low-frequency suppression algorithm significantly outperforms traditional processing algorithms by 3–8 dB when quantifying the input data, effectively preserving image features and improving image performance metrics. Furthermore, the quality map guidance image compression algorithm demonstrates a superior performance compared to the baseline model.

Cardiac arrhythmia (CA) is a severe cardiac disorder that results in a significant number of fatalities worldwide each year. Conventional electrocardiography (ECG) devices are often unable to detect arrhythmia symptoms during patients’ hospital visits due to their intermittent nature. This paper presents a wearable ECG processor for cardiac arrhythmia (CA) detection. The processor utilizes a Hilbert transform-based R-peak detection engine for R-peak detection, a Haar discrete wavelet transform (HDWT) unit for feature extraction, and a Hybrid ECG classifier that combines linear methods and Non-Linear Support Vector Machines (NLSVM) classifiers to distinguish between normal and abnormal heartbeats. The processor is fabricated by the CMOS 110 nm process with an area of 1.34 mm2 and validated with the MIT_BIH Database. The whole design consumes 4.08 μW with an average classification accuracy of 97.34%.

Offensive language detection has received important attention and plays a crucial role in promoting healthy communication on social platforms, as well as promoting the safe deployment of large language models. Training data is the basis for developing detectors; however, the available offense-related dataset in Chinese is severely limited in terms of data scale and coverage when compared to English resources. This significantly affects the accuracy of Chinese offensive language detectors in practical applications, especially when dealing with hard cases or out-of-domain samples. To alleviate the limitations posed by available datasets, we introduce AugCOLD (Augmented Chinese Offensive Language Dataset), a large-scale unsupervised dataset containing 1 million samples gathered by data crawling and model generation. Furthermore, we employ a multiteacher distillation framework to enhance detection performance with unsupervised data. That is, we build multiple teachers with publicly accessible datasets and use them to assign soft labels to AugCOLD. The soft labels serve as a bridge for knowledge to be distilled from both AugCOLD and multiteacher to the student network, i.e., the final offensive detector. We conduct experiments on multiple public test sets and our well-designed hard tests, demonstrating that our proposal can effectively improve the generalization and robustness of the offensive language detector.

Introduction The rates of type 2 diabetes mellitus (T2DM) in children and adolescents have risen globally over the past few years. While a few diabetes pharmacotherapies have been used in this population, their comparative benefits and harms are unclear. Thus, we will conduct a systematic review and network meta-analysis (NMA) of randomised controlled trials (RCTs) to compare the efficacy and safety of pharmacotherapies for managing paediatric T2DM.Methods and analysis We will include RCTs that enrolled T2DM patients ≤18 years of age and who were randomised to monotherapy or combination pharmacotherapies with or without lifestyle interventions. Comparator groups will include placebo or non-pharmacological treatments including lifestyle interventions.Treatment outcomes will include change from baseline in glycated haemoglobin A1c, body mass index z-score, weight, systolic/diastolic blood pressure, fasting plasma glucose, fasting insulin and lipid profiles, T2DM-related complications, as well as the incidence of treatment-related adverse events.Literature searches will be conducted in Medline, Embase, CINAHL, CENTRAL and Web of Science. We will also search the grey literature and the reference list of included trials and relevant reviews. Two reviewers will assess the eligibility of articles identified through our searches and will extract data from eligible studies independently. We will use a modified Cochrane instrument to evaluate the risk of bias. Disagreements will be resolved through consensus or arbitration by a third reviewer.A frequentist random-effects model will be used for conducting NMA. The quality of evidence will be assessed using the Confidence in Network Meta-Analysis platform. We will assess the effect modification through network meta-regression and subgroup analyses for sex, age at study inclusion, duration of T2DM, follow-up duration and risk of bias ratings.Ethics and dissemination This study will not require ethics approval. We will disseminate our findings through publication in a peer-reviewed journal and conference presentations.PROSPERO registration number CRD42022310100.

Aspect Sentiment Triplet Extraction (ASTE) is a complex and challenging task in Natural Language Processing (NLP). It aims to extract the triplet of aspect term, opinion term, and their associated sentiment polarity, which is a more fine-grained study in Aspect Based Sentiment Analysis. Furthermore, there have been a large number of approaches being proposed to handle this relevant task. However, existing methods for ASTE suffer from powerless interactions between different sources of textual features, and they usually exert an equal impact on each type of feature, which is quite unreasonable while building contextual representation. Therefore, in this paper, we propose a novel Multi-Branch GCN (MBGCN)-based ASTE model to solve this problem. Specifically, our model first generates the enhanced semantic features via the structure-biased BERT, which takes the position of tokens into the transformation of self-attention. Then, a biaffine attention module is utilized to further obtain the specific semantic feature maps. In addition, to enhance the dependency among words in the sentence, four types of linguistic relations are defined, namely part-of-speech combination, syntactic dependency type, tree-based distance, and relative position distance of each word pair, which are further embedded as adjacent matrices. Then, the widely used Graph Convolutional Network (GCN) module is utilized to complete the work of integrating the semantic feature and linguistic feature, which is operated on four types of dependency relations repeatedly. Additionally, an effective refining strategy is employed to detect whether word pairs match or not, which is conducted after the operation of each branch GCN. At last, a shallow interaction layer is designed to achieve the final textual representation by fusing the four branch features with different weights. To validate the effectiveness of MBGCNs, extensive experiments have been conducted on four public and available datasets. Furthermore, the results demonstrate the effectiveness and robustness of MBGCNs, which obviously outperform state-of-the-art approaches.

BackgroundBrain metastases (BM) from non-small-cell lung cancer (NSCLC) are frequent and carry significant morbidity, and current management options include varying local and systemic therapies. Here, we performed a systematic review and network meta-analysis to determine the ideal treatment regimen for NSCLC BMs with targetable EGFR-mutations/ALK-rearrangements.MethodsWe searched MEDLINE, EMBASE, Web of Science, ClinicalTrials.gov, CENTRAL and references of key studies for randomized controlled trials (RCTs) published from inception until June 2020. Comparative RCTs including ≥10 patients were selected. We used a frequentist random-effects model for network meta-analysis (NMA) and assessed the certainty of evidence using the GRADE approach. Our primary outcome of interest was intracranial progression-free survival (iPFS).ResultsWe included 24 studies representing 19 trials with 1623 total patients. Targeted tyrosine kinase inhibitors (TKIs) significantly improved iPFS, with second-and third- generation TKIs showing the greatest benefit (HR=0.25, 95%CI 0.15-0.40). Overall PFS was also improved compared to conventional chemotherapy (HR=0.47, 95%CI 0.36-0.61). In EGFR-mutant patients, osimertinib showed the greatest benefit in iPFS (HR=0.32, 95%CI 0.15-0.69) compared to conventional chemotherapy, while gefitinib + chemotherapy showed the greatest overall PFS benefit (HR=0.26, 95%CI 0.10-0.70). All ALKi improved overall PFS compared to conventional chemotherapy, with alectinib having the greatest benefit (HR=0.13, 95%CI 0.07-0.24).ConclusionsIn patients with NSCLC BMs and EGFR/ALK mutations, targeted TKIs improve intracranial and overall PFS compared to conventional modalities such as chemotherapy, with greater efficacy seen using newer generations of TKIs. This data is important for treatment selection and patient counseling, and highlights areas for future RCT research.Systematic Review Registrationhttps://www.crd.york.ac.uk/prospero/display_record.php?RecordID=179060.

Emotion recognition has been used widely in various applications such as mental health monitoring and emotional management. Usually, emotion recognition is regarded as a text classification task. Emotion recognition is a more complex problem, and the relations of emotions expressed in a text are nonnegligible. In this paper, a hierarchical model with label embedding is proposed for contextual emotion recognition. Especially, a hierarchical model is utilized to learn the emotional representation of a given sentence based on its contextual information. To give emotion correlation-based recognition, a label embedding matrix is trained by joint learning, which contributes to the final prediction. Comparison experiments are conducted on Chinese emotional corpus RenCECps, and the experimental results indicate that our approach has a satisfying performance in textual emotion recognition task.

Rich in phenolic compounds, Moringa oleifera leaf extract (ME) exhibits significant antioxidant activity both in vitro and in vivo. ME has already been widely used in fields of medicine, functional food, and cosmetics. Ultrasonic extraction (UE) method has been improved to be one of the most effective ways to extract phenols from M. oleifera leaves. The purpose of this study was to optimize ultrasonic extraction of phenols by response surface methodology (RSM). Four parameters were discussed, such as ethanol concentration, solvent-sample ratio, extraction temperature, and extraction time. Also, purification methods of the crude ME by organic solvent extraction and column chromatography were examined. Antioxidant activities of ME and each fraction were evaluated by DPPH, ABTS, and hydroxy radical-scavenging activities and reducing power. The phenol content of the purified ME reached up to 962.6 mg RE/g, extremely higher than the crude extract 107.22 ± 1.93 mg RE/g. The antioxidant activity of the purified ME was also significantly improved. Furthermore, phenols were identified by using the HPLC-MS method, and the results showed that there were 6 phenolic acids and derivatives and 7 flavonoids in ME. Quercetin-3-O-β-D-glucoside isolated from ME showed excellent DPPH and ABTS radical-scavenging abilities, which were comparable to VC.

As a foundation and typical task in natural language processing, text classification has been widely applied in many fields. However, as the basis of text classification, most existing corpus are imbalanced and often result in the classifier tending its performance to those categories with more texts. In this paper, we propose a background knowledge based multi-stream neural network to make up for the imbalance or insufficient information caused by the limitations of training corpus. The multi-stream network mainly consists of the basal stream, which retained original sequence information, and background knowledge based streams. Background knowledge is composed of keywords and co-occurred words which are extracted from external corpus. Background knowledge based streams are devoted to realizing supplemental information and reinforce basal stream. To better fuse the features extracted from different streams, early-fusion and two after-fusion strategies are employed. According to the results obtained from both Chinese corpus and English corpus, it is demonstrated that the proposed background knowledge based multi-stream neural network performs well in classification tasks.

The well-known Jarzynski equality, often written in the form e − β Δ F = 〈 e − β W 〉 , provides a non-equilibrium means to measure the free energy difference Δ F of a system at the same inverse temperature β based on an ensemble average of non-equilibrium work W. The accuracy of Jarzynski’s measurement scheme was known to be determined by the variance of exponential work, denoted as var e − β W . However, it was recently found that var e − β W can systematically diverge in both classical and quantum cases. Such divergence will necessarily pose a challenge in the applications of Jarzynski equality because it may dramatically reduce the efficiency in determining Δ F . In this work, we present a deformed Jarzynski equality for both classical and quantum non-equilibrium statistics, in efforts to reuse experimental data that already suffers from a diverging var e − β W . The main feature of our deformed Jarzynski equality is that it connects free energies at different temperatures and it may still work efficiently subject to a diverging var e − β W . The conditions for applying our deformed Jarzynski equality may be met in experimental and computational situations. If so, then there is no need to redesign experimental or simulation methods. Furthermore, using the deformed Jarzynski equality, we exemplify the distinct behaviors of classical and quantum work fluctuations for the case of a time-dependent driven harmonic oscillator dynamics and provide insights into the essential performance differences between classical and quantum Jarzynski equalities.