Recent advancements in deep learning have led to a resurgence of medical imaging and Electronic Medical Record (EMR) models for a variety of applications, including clinical decision support, ...automated workflow triage, clinical prediction and more. However, very few models have been developed to integrate both clinical and imaging data, despite that in routine practice clinicians rely on EMR to provide context in medical imaging interpretation. In this study, we developed and compared different multimodal fusion model architectures that are capable of utilizing both pixel data from volumetric Computed Tomography Pulmonary Angiography scans and clinical patient data from the EMR to automatically classify Pulmonary Embolism (PE) cases. The best performing multimodality model is a late fusion model that achieves an AUROC of 0.947 95% CI: 0.946-0.948 on the entire held-out test set, outperforming imaging-only and EMR-only single modality models.
Abstract
Advancements in deep learning techniques carry the potential to make significant contributions to healthcare, particularly in fields that utilize medical imaging for diagnosis, prognosis, ...and treatment decisions. The current state-of-the-art deep learning models for radiology applications consider only pixel-value information without data informing clinical context. Yet in practice, pertinent and accurate non-imaging data based on the clinical history and laboratory data enable physicians to interpret imaging findings in the appropriate clinical context, leading to a higher diagnostic accuracy, informative clinical decision making, and improved patient outcomes. To achieve a similar goal using deep learning, medical imaging pixel-based models must also achieve the capability to process contextual data from electronic health records (EHR) in addition to pixel data. In this paper, we describe different data fusion techniques that can be applied to combine medical imaging with EHR, and systematically review medical data fusion literature published between 2012 and 2020. We conducted a systematic search on PubMed and Scopus for original research articles leveraging deep learning for fusion of multimodality data. In total, we screened 985 studies and extracted data from 17 papers. By means of this systematic review, we present current knowledge, summarize important results and provide implementation guidelines to serve as a reference for researchers interested in the application of multimodal fusion in medical imaging.
Cross-domain fault diagnosis methods based on transfer learning attempt to leverage knowledge from a domain with sufficient labeled samples to a different but related domain with few or even ...nonlabeled samples. These methods have been widely investigated in the past years. Notwithstanding the efficacy, most existing approaches assume that the label spaces of training and testing data are the same. However, this assumption is not practical in actual applications because new fault category usually happens in the testing stage. A cross-domain open-set transfer diagnosis method is presented in this article to manage the aforementioned problem. Domain adversarial model is employed to discriminate known from unknown target instances. Moreover, multiple auxiliary classifiers introduce a weighting module to evaluate the distinguishing domain knowledge to provide target instances with representative weights. The new adversarial domain adaptation network with diverse supplementary classifiers can effectively identify the unknown and known fault categories in the target domain and bridge the domain shift between the common fault category of the source and target domain. Experiments on two bearing datasets show the effectiveness and advantage of the proposed method.
The selective hydrogenation of acetylene to ethylene in an ethylene‐rich gas stream is an important process in the chemical industry. Pd‐based catalysts are widely used in this reaction due to their ...excellent hydrogenation activity, though their selectivity for acetylene hydrogenation and durability need improvement. Herein, the successful synthesis of atomically dispersed Pd single‐atom catalysts on nitrogen‐doped graphene (Pd1/N‐graphene) by a freeze‐drying‐assisted method is reported. The Pd1/N‐graphene catalyst exhibits outstanding activity and selectivity for the hydrogenation of C2H2 with H2 in the presence of excess C2H4 under photothermal heating (UV and visible‐light irradiation from a Xe lamp), achieving 99% conversion of acetylene and 93.5% selectivity to ethylene at 125 °C. This remarkable catalytic performance is attributed to the high concentration of Pd active sites on the catalyst surface and the weak adsorption energy of ethylene on isolated Pd atoms, which prevents C2H4 hydrogenation. Importantly, the Pd1/N‐graphene catalyst exhibits excellent durability at the optimal reaction temperature of 125 °C, which is explained by the strong local coordination of Pd atoms by nitrogen atoms, which suppresses the Pd aggregation. The results presented here encourage the wider pursuit of solar‐driven photothermal catalyst systems based on single‐atom active sites for selective hydrogenation reactions.
Pd single‐atom catalysts on nitrogen‐doped graphene are successfully fabricated. A Pd1/N‐graphene catalyst (Pd loading of 2.3 wt%) exhibits outstanding activity and selectivity for the selective hydrogenation of acetylene in the presence of excess ethylene under photothermal or direct thermal heating at 125 °C, which is attributed to the suppression of C2H4 hydrogenation to C2H6 by Pd‐N4 surface sites.
The topology of the electronic structure of a crystal is manifested in its surface states. Recently, a distinct topological state has been proposed in metals or semimetals whose spin-orbit band ...structure features three-dimensional Dirac quasiparticles. We used angle-resolved photoemission spectroscopy to experimentally observe a pair of spin-polarized Fermi arc surface states on the surface of the Dirac semimetal Na3Bi at its native chemical potential. Our systematic results collectively identify a topological phase in a gapless material. The observed Fermi arc surface states open research frontiers in fundamental physics and possibly in spintronics.
RNA modulation has become a promising therapeutic approach for the treatment of several types of disease. The emerging field of noncoding RNA-based therapies has now come to the attention of ...cardiovascular research, in which it could provide valuable advancements in comparison to current pharmacotherapy such as small molecule drugs or antibodies. In this review, we focus on noncoding RNA-based studies conducted mainly in large-animal models, including pigs, rabbits, dogs, and nonhuman primates. The obstacles and promises of targeting long noncoding RNAs and circRNAs as therapeutic modalities in humans are specifically discussed. We also describe novel ex vivo methods based on human cells and tissues, such as engineered heart tissues and living myocardial slices that could help bridging the gap between in vivo models and clinical applications in the future. Finally, we summarize antisense oligonucleotide drugs that have already been approved by the Food and Drug Administration for targeting mRNAs and discuss the progress of noncoding RNA-based drugs in clinical trials. Additional factors, such as drug chemistry, drug formulations, different routes of administration, and the advantages of RNA-based drugs, are also included in the present review. Recently, first therapeutic miRNA-based inhibitory strategies have been tested in heart failure patients as well as healthy volunteers to study effects on wound healing (NCT04045405; NCT03603431). In summary, a combination of novel therapeutic RNA targets, large-animal models, ex vivo studies with human cells/tissues, and new delivery techniques will likely lead to significant progress in the development of noncoding RNA-based next-generation therapeutics for cardiovascular disease.
In the modern society when intelligent mobile devices become popular, the Internet breaks through the restrictions on time and space and becomes a ubiquitous learning tool. Designing teaching ...activity for digital learning and flexibly applying technology tools are the key issues for current information technology integrated education. In this study, students are tested and proceeded questionnaire survey to understand the opinions about digital learning. To effectively achieve the research objectives and test the research hypotheses, quasi-experimental research is applied in this study. Total 116 students in 4 classes are selected as the research subjects for the instructional research. The research results conclude that 1. digital learning presents better positive effects on learning motivation than traditional teaching does, 2. digital learning shows better positive effects on learning outcome than traditional teaching does, 3. learning motivation reveals significantly positive effects on learning effect in learning outcome, and 4. learning motivation appears remarkably positive effects on learning gain in learning outcome. It is expected to combine with current teaching trend and utilize the advantages of digital learning to develop practicable teaching strategies for the teaching effectiveness.
To provide a complete toxicity profile, toxicity spectrum, and a safety ranking of immune checkpoint inhibitor (ICI) drugs for treatment of cancer.
Systematic review and network meta-analysis.
...Electronic databases (PubMed, Embase, Cochrane Library, and Web of Science) were systematically searched to include relevant studies published in English between January 2007 and February 2018.
Only head-to-head phase II and III randomised controlled trials comparing any two or three of the following treatments or different doses of the same ICI drug were included: nivolumab, pembrolizumab, ipilimumab, tremelimumab, atezolizumab, conventional therapy (chemotherapy, targeted therapy, and their combinations), two ICI drugs, or one ICI drug with conventional therapy. Eligible studies must have reported site, organ, or system level data on treatment related adverse events. High quality, single arm trials and placebo controlled trials on ICI drugs were selected to establish a validation group.
36 head-to-head phase II and III randomised trials (n=15 370) were included. The general safety of ICI drugs ranked from high to low for all adverse events was as follows: atezolizumab (probability 76%, pooled incidence 66.4%), nivolumab (56%, 71.8%), pembrolizumab (55%, 75.1%), ipilimumab (55%, 86.8%), and tremelimumab (54%, not applicable). The general safety of ICI drugs ranked from high to low for severe or life threatening adverse events was as follows: atezolizumab (49%, 15.1%), nivolumab (46%, 14.1%), pembrolizumab (72%, 19.8%), ipilimumab (51%, 28.6%), and tremelimumab (28%, not applicable). Compared with conventional therapy, treatment-related adverse events for ICI drugs occurred mainly in the skin, endocrine, hepatic, and pulmonary systems. Taking one ICI drug was generally safer than taking two ICI drugs or one ICI drug with conventional therapy. Among the five ICI drugs, atezolizumab had the highest risk of hypothyroidism, nausea, and vomiting. The predominant treatment-related adverse events for pembrolizumab were arthralgia, pneumonitis, and hepatic toxicities. The main treatment-related adverse events for ipilimumab were skin, gastrointestinal, and renal toxicities. Nivolumab had a narrow and mild toxicity spectrum, mainly causing endocrine toxicities. Integrated evidence from the pooled incidences, subgroup, and sensitivity analyses implied that nivolumab is the best option in terms of safety, especially for the treatment of lung cancer.
Compared with other ICI drugs used to treat cancer, atezolizumab had the best safety profile in general, and nivolumab had the best safety profile in lung cancer when taking an integrated approach. The safety ranking of treatments based on ICI drugs is modulated by specific treatment-related adverse events.
PROSPERO CRD42017082553.
The activity of electrocatalysts strongly depends on the number of active sites, which can be increased by downsizing electrocatalysts. Single‐atom catalysts have attracted special attention due to ...atomic‐scale active sites. However, it is a huge challenge to obtain atomic‐scale CoOx catalysts. The Co‐based metal–organic frameworks (MOFs) own atomically dispersed Co ions, which motivates to design a possible pathway to partially on‐site transform these Co ions to active atomic‐scale CoOx species, while reserving the highly porous features of MOFs. In this work, for the first time, the targeted on‐site formation of atomic‐scale CoOx species is realized in ZIF‐67 by O2 plasma. The abundant pores in ZIF‐67 provide channels for O2 plasma to activate the Co ions in MOFs to on‐site produce atomic‐scale CoOx species, which act as the active sites to catalyze the oxygen evolution reaction with an even better activity than RuO2.
Through the efficient and mild O2‐plasma process, atomic‐scale CoOx species in a ZIF‐67 matrix are on‐site formed, which are in favor of combining with OH* to act as active sites to directly catalyze the oxygen evolution reduction with an even better activity than commercial RuO2.
There is an urgent need to identify antivirals to curtail the COVID-19 pandemic. Herein, we report the sensitivity of SARS-CoV-2 to recombinant human interferons α and β (IFNα/β). Treatment with ...IFN-α or IFN-β at a concentration of 50 international units (IU) per milliliter reduces viral titers by 3.4 log or over 4 log, respectively, in Vero cells. The EC50 of IFN-α and IFN-β treatment is 1.35 IU/ml and 0.76 IU/ml, respectively, in Vero cells. These results suggest that SARS-CoV-2 is more sensitive than many other human pathogenic viruses, including SARS-CoV. Overall, our results demonstrate the potential efficacy of human Type I IFN in suppressing SARS-CoV-2 infection, a finding which could inform future treatment options for COVID-19.
•Type I Interferons inhibit SARS-CoV-2, the causative agent for COVID-19, in cultured cells.•Treatment with IFN-α or IFN-β at 50 IU/ml reduces viral titers by 3.4-log or over 4-log, respectively, in Vero cells.•The EC50 of IFN-α and IFN-β treatment is 1.35 IU/ml and 0.76 IU/ml, respectively, in Vero cells.•Type I IFNs have been used in clinical therapies and thus could be repurposed in COVID-19 treatment.