Search

PurposeReal world evidence is crucial to understanding the diffusion of new oncologic therapies, monitoring cancer outcomes, and detecting unexpected toxicities. In practice, real world evidence is challenging to collect rapidly and comprehensively, often requiring expensive and time-consuming manual case-finding and annotation of clinical text. In this Review, we summarise recent developments in the use of artificial intelligence to collect and analyze real world evidence in oncology.MethodsWe performed a narrative review of the major current trends and recent literature in artificial intelligence applications in oncology.ResultsArtificial intelligence (AI) approaches are increasingly used to efficiently phenotype patients and tumors at large scale. These tools also may provide novel biological insights and improve risk prediction through multimodal integration of radiographic, pathological, and genomic datasets. Custom language processing pipelines and large language models hold great promise for clinical prediction and phenotyping.ConclusionsDespite rapid advances, continued progress in computation, generalizability, interpretability, and reliability as well as prospective validation are needed to integrate AI approaches into routine clinical care and real-time monitoring of novel therapies.

What professional development do assessment professionals need to succeed in higher education? To meet the rising demands in the field, assessment professionals seek ways to improve their skill sets. The current study utilized survey responses from over 200 higher education assessment professionals across the United States to assess the amount of training assessment professionals had when starting their position, the professional development they have sought since being in the position, and what training is still needed in the field. A thematic analysis of three open-ended survey questions revealed the eclectic backgrounds of assessment professionals, their preferred trainings while in the role, and the content and modalities requested for future professional development.

Abstract Since previous studies have suggested that the RNAs of human endogenous retrovirus (HERV) might be involved in regulating innate immunity, it is important to investigate the HERV transcriptome patterns in innate immune cell types such as CD14 + monocytes. Using single cell RNA-seq datasets from resting or stimulated PBMCs mapped to 3,220 known discrete autonomous proviral HERV loci, we found individual-specific variation in HERV transcriptomes between HERV loci in CD14 + monocytes. Analysis of paired datasets from the same individual that were cultured in vitro with LPS or without (i.e. control) revealed 36 HERV loci in CD14 + monocytes that were detected only after activation. To extend our analysis to in vivo activated CD14 + monocytes, we used two scRNA-seq datasets from studies that had demonstrated activation of circulating CD14 + monocytes in patients with physical trauma or patients hospitalized with COVID-19 infections. For direct comparison between the trauma and COVID-19 datasets, we first analyzed 1.625 billion sequence reads from a composite pangenome control of 21 normal individuals. Comparison of the sequence read depth of HERV loci in the trauma or COVID-19 samples to the pangenome control revealed that 39 loci in the COVID-19 and 11 HERV loci in the trauma samples were significantly different (Mann-Whitney U test), with 9 HERV loci shared between the COVID-19 and trauma datasets. The capacity to compare HERV loci transcriptome patterns in innate immune cells, like CD14 + monocytes, across different pathological conditions will lead to greater understanding of the physiological role of HERV expression in health and disease.

Abstract Objective Biallelic titin truncating variants (TTNtv) have been associated with a wide phenotypic spectrum, ranging from complex prenatal muscle diseases with dysmorphic features to adult‐onset limb‐girdle muscular dystrophy, with or without cardiac involvement. Given the size and complexity of TTN, reaching an unequivocal molecular diagnosis and precise disease prognosis remains challenging. Methods In this case series, 12 unpublished cases and one already published case with biallelic TTNtv were collected from multiple international medical centers between November 2022 and September 2023. TTN mutations were detected through exome or genome sequencing. Information about familial and personal clinical history was collected in a standardized form. RNA‐sequencing and analysis of TTN exon usage were performed on an internal sample cohort including postnatal skeletal muscles, fetal skeletal muscles, postnatal heart muscles, and fetal heart muscles. In addition, publicly available RNA‐sequencing data was retrieved from ENCODE. Results We generated new RNA‐seq data on TTN exons and identified genotype–phenotype correlations with prognostic implications for each titinopathy patient (whether worsening or improving in prenatal and postnatal life) using percentage spliced in (PSI) data for the involved exons. Interestingly, thanks to exon usage, we were also able to rule out a titinopathy diagnosis in one prenatal case. Interpretation This study demonstrates that exon usage provides valuable insights for a more exhaustive clinical interpretation of TTNtv; additionally, it may serve as a model for implementing personalized medicine in many other genetic diseases, since most genes undergo alternative splicing.

Abstract Long-term agricultural experiments are essential to measure the impacts of farming practices on crop yields, soil fertility and biogeochemical processes. However, these impacts often only manifest at decadal timescales, requiring committed and consistent data collection that exceeds the timelines for most experiments. The second oldest agricultural experiment in the world, the Morrow Plots at University of Illinois Urbana-Champaign (USA) has examined the impact of crop rotation and fertility treatments on maize (Zea mays L.) yields since 1876. While results have been widely reported since 1888, the publicly available longitudinal dataset described here now allows for validation of those past results, as well as new analyses and investigations. A multi-disciplinary team identified, collected, and aggregated multiple historical data sources into one comprehensive and FAIR (Findable, Accessible, Interoperable and Reusable) dataset that synthesizes yield data and management practices from 1888–2021. Updated versions of the dataset will continue to be published as additional data from this ongoing experiment are made available and as new historical data sources are uncovered.

Abstract Personal and family history of suicidal thoughts and behaviors (PSH and FSH, respectively) are significant risk factors associated with suicides. Research is limited in automatic identification of such data from clinical notes in Electronic Health Records. This study developed deep learning (DL) tools utilizing transformer models (Bio_ClinicalBERT and GatorTron) to detect PSH and FSH in clinical notes derived from three academic medical centers, and compared their performance with a rule-based natural language processing tool. For detecting PSH, the rule-based approach obtained an F1-score of 0.75 ± 0.07, while the Bio_ClinicalBERT and GatorTron DL tools scored 0.83 ± 0.09 and 0.84 ± 0.07, respectively. For detecting FSH, the rule-based approach achieved an F1-score of 0.69 ± 0.11, compared to 0.89 ± 0.10 for Bio_ClinicalBERT and 0.92 ± 0.07 for GatorTron. Across sites, the DL tools identified more than 80% of patients at elevated risk for suicide who remain undiagnosed and untreated.

Precise gamma-ray spectral analysis is crucial in high-stakes applications, such as nuclear security. Research efforts toward implementing machine learning (ML) approaches for accurate analysis are limited by the resemblance of the training data to the testing scenarios. The underlying spectral shape of synthetic data may not perfectly reflect measured configurations, and measurement campaigns may be limited by resource constraints. Consequently, ML algorithms for isotope identification must maintain accurate classification performance under domain shifts between the training and testing data. To this end, four different classifiers (Ridge, Random Forest, Extreme Gradient Boosting, and Multilayer Perceptron) were trained on the same dataset and evaluated on twelve other datasets with varying standoff distances, shielding, and background configurations. A tailored statistical approach was introduced to quantify the similarity between the training and testing configurations, which was then related to the predictive performance. Wilcoxon signed-rank tests revealed that the OVR-wrapped XGB significantly outperformed the other algorithms, with confidence levels of 99.0% or above for the 133Ba, 60Co, 137Cs, and 152Eu sources. The findings from this work are significant as they outline techniques to promote the development of robust ML-based approaches for isotope identification.

Abstract By 2050, 1 in 4 people worldwide will be living with hearing impairment. We propose a digital Speech Hearing Screener (dSHS) using short nonsense word recognition to measure speech-hearing ability. The importance of hearing screening is increasing due to the anticipated increase in individuals with hearing impairment globally. We compare dSHS outcomes with standardized pure-tone averages (PTA) and speech-recognition thresholds (SRT). Fifty participants (aged 55 or older underwent pure-tone and speech-recognition thresholding. One-way ANOVA was used to compare differences between hearing impaired and hearing not-impaired groups, by the dSHS, with a clinical threshold of moderately impaired hearing at 35 dB and severe hearing impairment at 50 dB. dSHS results significantly correlated with PTAs/SRTs. ANOVA results revealed the dSHS was significantly different (F(1,47) = 38.1, p

Abstract Bronchopulmonary dysplasia (BPD) is a common complication of preterm birth. Despite this, genetic drivers of BPD are poorly understood. The objective of this study is to better understand the impact of single nucleotide polymorphisms (SNPs) previously associated with BPD by examining associations with other phenotypes. We drew pediatric subjects from the biorepository at the Center for Applied Genomics to identify associations between these SNPs and 2,146 imputed phenotypes. Methylation data, external cohorts, and in silico validation methods were used to corroborate significant associations. We identified 60 SNPs that were previously associated with BPD. We found a significant association between rs3771150 and rs3771171 and mean eosinophil percentage in a European cohort of 6,999 patients and replicated this in external cohorts. Both SNPs were also associated with asthma, COPD and FEV1/FVC ratio. These SNPs displayed associations with methylation probes and were functionally linked to ST2 (IL1RL1) levels in blood and lung tissue. Our findings support a genetic justification for the epidemiological link between BPD and asthma. Given the well-established link between ST2 and type 2 inflammation in asthma, these findings provide a rationale for future studies exploring the role of type 2 inflammation in the pathogenesis of BPD.

This study involved the creation and validation of a self-rated social presence measure. Study 1 utilized focus groups to create items. The focus group participants were presented with a set of items based upon past literature; through discussion of these items, a preliminary measure was created. Study 2 involved an exploratory factor analysis on the preliminary measure to eliminate items that did not work well with each other. This reduced the measure from 54 to 23 items. Study 3 validated a 21-item self-rated measure of creating a social presence (PSP), which can be used to determine if people have difficulty projecting themselves as real individuals willing to interact with other online communicators.

Positive academic climates are critical to helping students thrive, and kindness innovations might enhance these climates. This conceptual paper's purpose is to share insights from a consensus building event focused on fostering relationships and knowledge-sharing among an international group of multidisciplinary students, faculty, and staff who explored ways to bring a kindness framework into post-secondary education. Participants underscored kindness as critical for students' experiences and university culture, and identified several levels of influence requiring intervention focus. Ideas and strategies emerging from the event might serve to encourage student-led kindness initiatives and prompt university personnel to integrate kindness into post-secondary institutions.

Using intergroup contact and stigma management communication theories, this work examines the effect of cultural teaching and learning in Spanish language classes on cultural stigma, social distance, and ethnocentrism. Additionally, this study examines how individuals respond to positive and negative messages about foreigners. Survey results indicate cultural integration in the foreign language classroom was associated with increased cultural knowledge and, indirectly, decreased cultural stigma and social distance toward those from foreign cultures. The findings of this study extend prior intergroup contact work by demonstrating the potential for exposure to a culture, rather than its members, to decrease stigma.