Heat Islands and Chronic Disease: Could African Americans Be More Vulnerable to Heat-Related Health Impacts?

Global warming and environmental heat stress are public health concerns. Urban heat islands, metropolitan areas with higher temperatures compared to their surrounding rural areas, compound the effects of increased environmental heat. In addition to acute heat-related illness, increased environmental heat is linked to exacerbation of chronic diseases. The purpose of this narrative review is to provide an overview of heat islands and how the effects of heat stress intersect with chronic diseases in the African American (AA) community. Across the United States, AAs are more likely to reside in heat islands, resulting in greater exposure to environmental heat. Unfortunately, chronic diseases exacerbated by increased environmental heat disproportionately impact the AA community. Due to the intersection of these disparities, heat-related health risks are likely higher for the AAs. The increased health risks posed by urban heat island exposure on AAs have significant implications for nursing practice, research, and policy.

Sleep and neighborhood socioeconomic status: a micro longitudinal study of chronic low-back pain and pain-free individuals.

Individuals with chronic low back pain (cLBP) frequently report sleep disturbances. Living in a neighborhood characterized by low-socioeconomic status (SES) is associated with a variety of negative health outcomes, including poor sleep. Whether low-neighborhood SES exacerbates sleep disturbances of people with cLBP, relative to pain-free individuals, has not previously been observed. This study compared associations between neighborhood-level SES, pain-status (cLBP vs. pain-free), and daily sleep metrics in 117 adults (cLBP = 82, pain-free = 35). Neighborhood-level SES was gathered from Neighborhood Atlas, which provides a composite measurement of overall neighborhood deprivation (e.g. area deprivation index). Individuals completed home sleep monitoring for 7-consecutive days/nights. Neighborhood SES and pain-status were tested as predictors of actigraphic sleep variables (e.g., sleep efficiency). Analyses revealed neighborhood-level SES and neighborhood-level SES*pain-status interaction significantly impacted objective sleep quality. These findings provide initial support for the negative impact of low neighborhood-level SES and chronic pain on sleep quality.

Identification of DNA methylation associated enrichment pathways in adults with non-specific chronic low back pain.

Chronic low back pain (cLBP) that cannot be attributable to a specific pathoanatomical change is associated with high personal and societal costs. Still, the underlying mechanism that causes and sustains such a phenotype is largely unknown. Emerging evidence suggests that epigenetic changes play a role in chronic pain conditions. Using reduced representation bisulfite sequencing (RRBS), we evaluated DNA methylation profiles of adults with non-specific cLBP (n = 50) and pain-free controls (n = 48). We identified 28,325 hypermethylated and 36,936 hypomethylated CpG sites (p < 0.05). After correcting for multiple testing, we identified 159 DMRs (q < 0.01and methylation difference > 10%), the majority of which were located in CpG island (50%) and promoter regions (48%) on the associated genes. The genes associated with the differentially methylated regions were highly enriched in biological processes that have previously been implicated in immune signaling, endochondral ossification, and G-protein coupled transmissions. Our findings support inflammatory alterations and the role of bone maturation in cLBP. This study suggests that epigenetic regulation has an important role in the pathophysiology of non-specific cLBP and a basis for future studies in biomarker development and targeted interventions.

A Deep Convolutional Neural Network for Annotation of Magnetic Resonance Imaging Sequence Type.

The explosion of medical imaging data along with the advent of big data analytics has launched an exciting era for clinical research. One factor affecting the ability to aggregate large medical image collections for research is the lack of infrastructure for automated data annotation. Among all imaging modalities, annotation of magnetic resonance (MR) images is particularly challenging due to the non-standard labeling of MR image types. In this work, we aimed to train a deep neural network to annotate MR image sequence type for scans of brain tumor patients. We focused on the four most common MR sequence types within neuroimaging: T1-weighted (T1W), T1-weighted post-gadolinium contrast (T1Gd), T2-weighted (T2W), and T2-weighted fluid-attenuated inversion recovery (FLAIR). Our repository contains images acquired using a variety of pulse sequences, sequence parameters, field strengths, and scanner manufacturers. Image selection was agnostic to patient demographics, diagnosis, and the presence of tumor in the imaging field of view. We used a total of 14,400 two-dimensional images, each visualizing a different part of the brain. Data was split into train, validation, and test sets (9600, 2400, and 2400 images, respectively) and sets consisted of equal-sized groups of image types. Overall, the model reached an accuracy of 99% on the test set. Our results showed excellent performance of deep learning techniques in predicting sequence types for brain tumor MR images. We conclude deep learning models can serve as tools to support clinical research and facilitate efficient database management.

Taste the Pain: The Role of TRP Channels in Pain and Taste Perception.

Transient receptor potential (TRP) channels are a superfamily of cation transmembrane proteins that are expressed in many tissues and respond to many sensory stimuli. TRP channels play a role in sensory signaling for taste, thermosensation, mechanosensation, and nociception. Activation of TRP channels (e.g., TRPM5) in taste receptors by food/chemicals (e.g., capsaicin) is essential in the acquisition of nutrients, which fuel metabolism, growth, and development. Pain signals from these nociceptors are essential for harm avoidance. Dysfunctional TRP channels have been associated with neuropathic pain, inflammation, and reduced ability to detect taste stimuli. Humans have long recognized the relationship between taste and pain. However, the mechanisms and relationship among these taste-pain sensorial experiences are not fully understood. This article provides a narrative review of literature examining the role of TRP channels on taste and pain perception. Genomic variability in the TRPV1 gene has been associated with alterations in various pain conditions. Moreover, polymorphisms of the TRPV1 gene have been associated with alterations in salty taste sensitivity and salt preference. Studies of genetic variations in TRP genes or modulation of TRP pathways may increase our understanding of the shared biological mediators of pain and taste, leading to therapeutic interventions to treat many diseases.

Identifying Factors Underlying the Decision for Sickle Cell Carrier Screening Among African Americans Within Middle Reproductive Age.

Guidelines recommend that African Americans know their sickle cell trait status to inform reproductive health decisions. Few studies have applied a behavioral theory to identify factors associated with sickle cell trait screening to inform intervention targets to increase this behavior. We applied a Sickle Cell Trait Screening Framework to identify factors associated with African Americans' intention to ask for sickle cell trait screening. Participants (N = 300), ages 18 to 35, completed a cross-sectional survey. A three-step sequential ordinary least squares regression analysis identified factors influencing intention. Results indicated socio-demographic factors (age, education), knowledge and fear beliefs (screening knowledge, perceived threat), and reasoned action approach (RAA) constructs were associated with intention. RAA constructs influenced intention over knowledge and fear beliefs with an increase in R2 of .468. Perceived behavioral control was more predictive of intention (ß = .576, p < .001). Attitude and perceived norm also had significant weights (ß = .325 and ß = .192, both p < .001, respectively). Findings from this study can inform strategies (e.g., eliminating costs associated with screening, reducing fear of painful tests) to increase sickle cell trait screening among African Americans. Ultimately, more sickle cell carriers will become aware of their trait status and be able to make informed reproductive health decisions.

Biochemical Roles for Conserved Residues in the Bacterial Fatty Acid-binding Protein Family.

Fatty acid kinase (Fak) is a ubiquitous Gram-positive bacterial enzyme consisting of an ATP-binding protein (FakA) that phosphorylates the fatty acid bound to FakB. In Staphylococcus aureus, Fak is a global regulator of virulence factor transcription and is essential for the activation of exogenous fatty acids for incorporation into phospholipids. The 1.2-Å x-ray structure of S. aureus FakB2, activity assays, solution studies, site-directed mutagenesis, and in vivo complementation were used to define the functions of the five conserved residues that define the FakB protein family (Pfam02645). The fatty acid tail is buried within the protein, and the exposed carboxyl group is bound by a Ser-93-fatty acid carboxyl-Thr-61-His-266 hydrogen bond network. The guanidinium of the invariant Arg-170 is positioned to potentially interact with a bound acylphosphate. The reduced thermal denaturation temperatures of the T61A, S93A, and H266A FakB2 mutants illustrate the importance of the hydrogen bond network in protein stability. The FakB2 T61A, S93A, and H266A mutants are 1000-fold less active in the Fak assay, and the R170A mutant is completely inactive. All FakB2 mutants form FakA(FakB2)2 complexes except FakB2(R202A), which is deficient in FakA binding. Allelic replacement shows that strains expressing FakB2 mutants are defective in fatty acid incorporation into phospholipids and virulence gene transcription. These conserved residues are likely to perform the same critical functions in all bacterial fatty acid-binding proteins.

A comparative study of electronic cigarette vapor extracts on airway-related cell lines in vitro.

The use of electronic cigarettes (ECs) is rapidly increasing worldwide; however, scientific evidence regarding EC cytotoxicity is limited. The aim of this study was to evaluate the acute cytotoxicity of EC vapor extract (ECE) on airway-related cells in vitro. Cigarette smoke extract (CSE), vapor extract of fifteen brands/flavors of ECs and the extract from the E-vehicle (propylene glycol and glycerin) was collected. Extracts, in concentrations of 100-12.5%, were added to human bronchial epithelial (BEAS-2B, IB3-1 and C38), fibroblast (Wi-38) and macrophage (J774 and THP-1) cell lines. Viability was assessed after 24 h using a standard XTT assay. Viability of <70% of control (no extract) was considered cytotoxic according to UNI EN ISO 10993-5 standards. CSE displayed a concentration-dependent influence on cell viability across all four cell lines with 100% producing the most toxic effect, therefore validating the model and indicating higher cytotoxicity than in ECEs. ECEs did reduce viability although this was not correlated with nicotine content or the E-vehicle. However, several flavors proved cytotoxic, with variation between different brands and cell lines. These data indicate that not all ECs are the same and that use of a particular flavor or brand may have differing effects. The cell line used is also an important factor. More research is crucial to ascertain the health effects of different ECs before they can be accepted as a safe alternative to tobacco cigarettes.

Identification of a two-component fatty acid kinase responsible for host fatty acid incorporation by Staphylococcus aureus.

Extracellular fatty acid incorporation into the phospholipids of Staphylococcus aureus occurs via fatty acid phosphorylation. We show that fatty acid kinase (Fak) is composed of two dissociable protein subunits encoded by separate genes. FakA provides the ATP binding domain and interacts with two distinct FakB proteins to produce acyl-phosphate. The FakBs are fatty acid binding proteins that exchange bound fatty acid/acyl-phosphate with fatty acid/acyl-phosphate presented in detergent micelles or liposomes. The ΔfakA and ΔfakB1 ΔfakB2 strains were unable to incorporate extracellular fatty acids into phospholipid. FakB1 selectively bound saturated fatty acids whereas FakB2 preferred unsaturated fatty acids. Affymetrix array showed a global perturbation in the expression of virulence genes in the ΔfakA strain. The severe deficiency in α-hemolysin protein secretion in ΔfakA and ΔfakB1 ΔfakB2 mutants coupled with quantitative mRNA measurements showed that fatty acid kinase activity was required to support virulence factor transcription. These data reveal the function of two conserved gene families, their essential role in the incorporation of host fatty acids by Gram-positive pathogens, and connects fatty acid kinase to the regulation of virulence factor transcription in S. aureus.

Incorporation of extracellular fatty acids by a fatty acid kinase-dependent pathway in Staphylococcus aureus.

Acyl-CoA and acyl-acyl carrier protein (ACP) synthetases activate exogenous fatty acids for incorporation into phospholipids in Gram-negative bacteria. However, Gram-positive bacteria utilize an acyltransferase pathway for the biogenesis of phosphatidic acid that begins with the acylation of sn-glycerol-3-phosphate by PlsY using an acyl-phosphate (acyl-PO4 ) intermediate. PlsX generates acyl-PO4 from the acyl-ACP end-products of fatty acid synthesis. The plsX gene of Staphylococcus aureus was inactivated and the resulting strain was both a fatty acid auxotroph and required de novo fatty acid synthesis for growth. Exogenous fatty acids were only incorporated into the 1-position and endogenous acyl groups were channeled into the 2-position of the phospholipids in strain PDJ39 (ΔplsX). Extracellular fatty acids were not elongated. Removal of the exogenous fatty acid supplement led to the rapid accumulation of intracellular acyl-ACP and the abrupt cessation of fatty acid synthesis. Extracts from the ΔplsX strain exhibited an ATP-dependent fatty acid kinase activity, and the acyl-PO4 was converted to acyl-ACP when purified PlsX is added. These data reveal the existence of a novel fatty acid kinase pathway for the incorporation of exogenous fatty acids into S. aureus phospholipids.

Unanticipated frequency and consequences of regimen-related diarrhea in patients being treated with radiation or chemoradiation regimens for cancers of the head and neck or lung.

PURPOSE: To better understand the indirect effects of standard courses of radiation therapy (RT) on distant tissue toxicity, we evaluated the frequency, course, and health and economic burden of regimen-related diarrhea in a large, multinational group of patients who were being treated for cancers of the head and neck (HNC) or lung (NSCLC). METHODS: In this exploratory, prospective study, 284 patients being treated for HNC and 60 being treated for NSCLC were stratified into four cohorts to evaluate the effect of radiation alone and radiation plus concomitant chemotherapy (CRT) on radiation-induced diarrhea (RID). RID was assessed daily throughout RT using a patient-reported five-point categorical scale. Health and resource use outcomes were evaluated at least weekly during radiation. RESULTS: Moderate to severe RID was reported in all groups and was worse among patient being treated with concomitant chemoradiation (CRT). Whereas 29 % of patients treated with radiation only developed RID, the incidence was 42 % among CRT-treated patients. Tumor site did not impact the rate of RID, but did impact the rate of development and was more acute in patients being treated for NSCLC than for HNC. Patients with significant RID had worse health and resource use outcomes than did patients without RID regardless of the form of treatment. G-tube placement, weight loss, unplanned office visits, and in-patient days were adversely affected by RID. Not surprisingly, patients treated with CRT had poorer health and resource outcomes than RT only patients, even in the absence of RID. CONCLUSION: In addition to local tissue toxicities, our results suggest that focal radiation may also be associated with significant distant tissue-centric injury here represented by RID. While these changes were seen with radiation alone, the addition of chemotherapy increased the incidence and burden of illness. RID adversely impacted resource use. This unanticipated finding supports the hypothesis that focal radiation therapy results in pathobiological changes that extend beyond the radiation field and which can produce distant changes.

Patient-specific mathematical neuro-oncology: using a simple proliferation and invasion tumor model to inform clinical practice.

Glioblastoma multiforme (GBM) is the most common malignant primary brain tumor associated with a poor median survival of 15-18 months, yet there is wide heterogeneity across and within patients. This heterogeneity has been the source of significant clinical challenges facing patients with GBM and has hampered the drive toward more precision or personalized medicine approaches to treating these challenging tumors. Over the last two decades, the field of Mathematical Neuro-oncology has grown out of desire to use (often patient-specific) mathematical modeling to better treat GBMs. Here, we will focus on a series of clinically relevant results using patient-specific mathematical modeling. The core model at the center of these results incorporates two hallmark features of GBM, proliferation [Formula: see text] and invasion (D), as key parameters. Based on routinely obtained magnetic resonance images, each patient's tumor can be characterized using these two parameters. The Proliferation-Invasion (PI) model uses [Formula: see text] and D to create patient-specific growth predictions. The PI model, its predictions, and parameters have been used in a number of ways to derive biological insight. Beyond predicting growth, the PI model has been utilized to identify patients who benefit from different surgery strategies, to prognosticate response to radiation therapy, to develop a treatment response metric, and to connect clinical imaging features and genetic information. Demonstration of the PI model's clinical relevance supports the growing role for it and other mathematical models in routine clinical practice.

Declining newborn intramuscular vitamin K prophylaxis predicts subsequent immunisation refusal: A retrospective cohort study.

AIM: Low rates of childhood immunisation are linked to outbreaks of infectious disease. Identifying and addressing barriers to immunisation may lead to improved immunisation rates. Immunisation and newborn vitamin K prophylaxis have many similarities. We aimed to investigate whether parents who decline newborn vitamin K are also more likely to decline subsequent childhood immunisations. METHODS: We undertook a retrospective cohort study, examining the relationship between vitamin K administration and immunisation uptake by parents of babies born over a 2-year period (January 2010-December 2011) in Dunedin, New Zealand (NZ). Both written and electronic data from a single birthing unit and the NZ National Immunisation Register (NIR) were analysed to ascertain the relationship between declining newborn vitamin K prophylaxis and subsequent immunisation uptake. RESULTS: Records for 3575 babies were examined. Ninety-two per cent of infants received intramuscular, and 5% received oral vitamin K. An increased risk ratio for non-immunisation of 14.1 (95% confidence interval 7.8-25.9) for babies whose parents declined vitamin K was identified. Receiving oral vitamin K was also associated with subsequent non-immunisation, with a risk ratio of 3.5 (95% confidence interval 1.7-7.3). CONCLUSIONS: Parents who decline newborn vitamin K are more likely to decline immunisation for their child. These parents, as well as those that elect for oral vitamin K, are a small but easily identifiable group to whom additional education about the benefits of immunisation could be offered. This is especially pertinent at a time when there is a resurgence of immunisation preventable diseases.

Medical and midwifery attitudes towards vitamin K prophylaxis in New Zealand neonates.

AIM: Neonates are at risk for potentially life-threatening vitamin K deficiency bleeding. This can be readily prevented with prophylactic vitamin K following delivery. In this context, most vitamin K-deficiency bleeding occurs in those whose parents decline newborn vitamin K. One factor influencing parental decision-making is information received from health professionals. This study examined attitudes and perceptions towards newborn vitamin K in relevant health-care professionals. METHODS: A literature review and one-on-one semi-structured interviews were conducted to inform questionnaire design. Midwives and selected medical staff employed in the South Island of New Zealand were then invited to complete an anonymous survey exploring attitudes and perceptions towards vitamin K prophylaxis in newborns. RESULTS: The survey achieved an overall response rate of 57%. Almost all responding medical staff and 76% of midwives agreed with the current New Zealand Ministry of Health vitamin K guideline. All medical staff but only 55% of midwives feel that all babies should receive vitamin K. Differences were also seen between professionals with respect to vitamin K education and risks. CONCLUSION: This is the first study to examine attitudes and perceptions of midwives and doctors to vitamin K prophylaxis in neonates. Considerable discrepancies in attitude are evident, and in some midwives, a lack of confidence in this intervention is apparent. How this affects education to families is unknown. Increased understanding of this phenomenon, along with improved education and communication to professionals and families, is required.

Comparing the effects of infant maternal and sibling separation on adolescent behavior in rats (Rattus norvegicus).

Maternal separation in early life has been observed to have lasting, detrimental effects that impair personal and social development and can persist into adulthood. Maternal separation during infancy can be most detrimental during adolescence, leading to long-term adverse effects on development and social behavior. This research study compared the effects of sibling and maternal separation in infancy on anxiety, sociability, or memory later in adolescence (postnatal day, PND, 50-58) in male and female Long-Evans Rats (Rattus norvegicus). Rat pups were semi-randomly assigned into eight conditions for daily isolation (PND 1-14). The groups were separated by the duration of isolation between 15 minutes (control group) or 180 minutes (experimental group) and the sex of the rat. They were also separated by comfort conditions with the dam present in an adjoining cage versus not present and siblings present or not present during isolation. The result was a 2 (15-min vs. 180-min) x 2 (dam vs. no dam) x 2 (single vs. grouped) x 2 (male vs. female) design. Once pups had reached adolescence (PND 50), researchers tested for differences in anxiety, activity, and social behavior using elevated plus-maze, open field habituation, a three-chamber social interaction, and a social discrimination task. Results indicate that longer isolation was more stressful and caused lower body weight. The female rats showed more anxious behavior in the open field but only if they were in the shorter isolation group. Social interaction showed that the rats isolated with the dam had different effects of isolation. In males, shorter isolation with the dam increased sociability but decreased sociability in females. These complicated findings may be due to the effects of inoculation, which describes how moderate stress combined with comfort may produce adaptation or immunity to stress and affect males and females differently.

Biologically-informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post-treatment glioblastoma.

Intratumoral heterogeneity poses a significant challenge to the diagnosis and treatment of glioblastoma (GBM). This heterogeneity is further exacerbated during GBM recurrence, as treatment-induced reactive changes produce additional intratumoral heterogeneity that is ambiguous to differentiate on clinical imaging. There is an urgent need to develop non-invasive approaches to map the heterogeneous landscape of histopathological alterations throughout the entire lesion for each patient. We propose to predictively fuse Magnetic Resonance Imaging (MRI) with the underlying intratumoral heterogeneity in recurrent GBM using machine learning (ML) by leveraging image-localized biopsies with their associated locoregional MRI features. To this end, we develop BioNet, a biologically-informed neural network model, to predict regional distributions of three tissue-specific gene modules: proliferating tumor, reactive/inflammatory cells, and infiltrated brain tissue. BioNet offers valuable insights into the integration of multiple implicit and qualitative biological domain knowledge, which are challenging to describe in mathematical formulations. BioNet performs significantly better than a range of existing methods on cross-validation and blind test datasets. Voxel-level prediction maps of the gene modules by BioNet help reveal intratumoral heterogeneity, which can improve surgical targeting of confirmatory biopsies and evaluation of neuro-oncological treatment effectiveness. The non-invasive nature of the approach can potentially facilitate regular monitoring of the gene modules over time, and making timely therapeutic adjustment. These results also highlight the emerging role of ML in precision medicine.

Digital twins for health: a scoping review.

The use of digital twins (DTs) has proliferated across various fields and industries, with a recent surge in the healthcare sector. The concept of digital twin for health (DT4H) holds great promise to revolutionize the entire healthcare system, including management and delivery, disease treatment and prevention, and health well-being maintenance, ultimately improving human life. The rapid growth of big data and continuous advancement in data science (DS) and artificial intelligence (AI) have the potential to significantly expedite DT research and development by providing scientific expertise, essential data, and robust cybertechnology infrastructure. Although various DT initiatives have been underway in the industry, government, and military, DT4H is still in its early stages. This paper presents an overview of the current applications of DTs in healthcare, examines consortium research centers and their limitations, and surveys the current landscape of emerging research and development opportunities in healthcare. We envision the emergence of a collaborative global effort among stakeholders to enhance healthcare and improve the quality of life for millions of individuals worldwide through pioneering research and development in the realm of DT technology.

Biologically informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post treatment glioblastoma.

Intratumoral heterogeneity poses a significant challenge to the diagnosis and treatment of recurrent glioblastoma. This study addresses the need for non-invasive approaches to map heterogeneous landscape of histopathological alterations throughout the entire lesion for each patient. We developed BioNet, a biologically-informed neural network, to predict regional distributions of two primary tissue-specific gene modules: proliferating tumor (Pro) and reactive/inflammatory cells (Inf). BioNet significantly outperforms existing methods (p < 2e-26). In cross-validation, BioNet achieved AUCs of 0.80 (Pro) and 0.81 (Inf), with accuracies of 80% and 75%, respectively. In blind tests, BioNet achieved AUCs of 0.80 (Pro) and 0.76 (Inf), with accuracies of 81% and 74%. Competing methods had AUCs lower or around 0.6 and accuracies lower or around 70%. BioNet's voxel-level prediction maps reveal intratumoral heterogeneity, potentially improving biopsy targeting and treatment evaluation. This non-invasive approach facilitates regular monitoring and timely therapeutic adjustments, highlighting the role of ML in precision medicine.

Quantifying intra-tumoral genetic heterogeneity of glioblastoma toward precision medicine using MRI and a data-inclusive machine learning algorithm.

BACKGROUND AND OBJECTIVE: Glioblastoma (GBM) is one of the most aggressive and lethal human cancers. Intra-tumoral genetic heterogeneity poses a significant challenge for treatment. Biopsy is invasive, which motivates the development of non-invasive, MRI-based machine learning (ML) models to quantify intra-tumoral genetic heterogeneity for each patient. This capability holds great promise for enabling better therapeutic selection to improve patient outcome. METHODS: We proposed a novel Weakly Supervised Ordinal Support Vector Machine (WSO-SVM) to predict regional genetic alteration status within each GBM tumor using MRI. WSO-SVM was applied to a unique dataset of 318 image-localized biopsies with spatially matched multiparametric MRI from 74 GBM patients. The model was trained to predict the regional genetic alteration of three GBM driver genes (EGFR, PDGFRA and PTEN) based on features extracted from the corresponding region of five MRI contrast images. For comparison, a variety of existing ML algorithms were also applied. Classification accuracy of each gene were compared between the different algorithms. The SHapley Additive exPlanations (SHAP) method was further applied to compute contribution scores of different contrast images. Finally, the trained WSO-SVM was used to generate prediction maps within the tumoral area of each patient to help visualize the intra-tumoral genetic heterogeneity. RESULTS: WSO-SVM achieved 0.80 accuracy, 0.79 sensitivity, and 0.81 specificity for classifying EGFR; 0.71 accuracy, 0.70 sensitivity, and 0.72 specificity for classifying PDGFRA; 0.80 accuracy, 0.78 sensitivity, and 0.83 specificity for classifying PTEN; these results significantly outperformed the existing ML algorithms. Using SHAP, we found that the relative contributions of the five contrast images differ between genes, which are consistent with findings in the literature. The prediction maps revealed extensive intra-tumoral region-to-region heterogeneity within each individual tumor in terms of the alteration status of the three genes. CONCLUSIONS: This study demonstrated the feasibility of using MRI and WSO-SVM to enable non-invasive prediction of intra-tumoral regional genetic alteration for each GBM patient, which can inform future adaptive therapies for individualized oncology.

Preventing bad behavior in academia.

We gave young scientists this prompt: Describe one change to scientific policy or culture that would substantially decrease incidents of scientific misconduct or other unethical behavior.