A Neoteric Feature Extraction Technique to Predict the Survival of Gastric Cancer Patients.

BACKGROUND: At the time of cancer diagnosis, it is crucial to accurately classify malignant gastric tumors and the possibility that patients will survive. OBJECTIVE: This study aims to investigate the feasibility of identifying and applying a new feature extraction technique to predict the survival of gastric cancer patients. METHODS: A retrospective dataset including the computed tomography (CT) images of 135 patients was assembled. Among them, 68 patients survived longer than three years. Several sets of radiomics features were extracted and were incorporated into a machine learning model, and their classification performance was characterized. To improve the classification performance, we further extracted another 27 texture and roughness parameters with 2484 superficial and spatial features to propose a new feature pool. This new feature set was added into the machine learning model and its performance was analyzed. To determine the best model for our experiment, Random Forest (RF) classifier, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Naïve Bayes (NB) (four of the most popular machine learning models) were utilized. The models were trained and tested using the five-fold cross-validation method. RESULTS: Using the area under ROC curve (AUC) as an evaluation index, the model that was generated using the new feature pool yields AUC = 0.98 ± 0.01, which was significantly higher than the models created using the traditional radiomics feature set (p < 0.04). RF classifier performed better than the other machine learning models. CONCLUSIONS: This study demonstrated that although radiomics features produced good classification performance, creating new feature sets significantly improved the model performance.

The Influence of Health-Related Behavior Profiles on College Students' Perceptions of COVID-19 Safety Measures.

College students often engage in multiple health-related behaviors simultaneously which can lead to negative outcomes and further risky behaviors. During the COVID-19 pandemic, college students reported decreased condom use, increased solitary cannabis use, and increased alcohol consumption. This current study aimed to (1) identify profiles of health-related behaviors (i.e., alcohol consumption, cannabis use, and sexual behaviors), and (2) determine if these profiles would differ in engagement and perceived effectiveness of COVID-19 preventative measures. Participants were 273 college students from a large Northeastern U.S. public university who completed surveys about health-related behaviors during the 2021 academic year. We used a latent profile analysis to identify distinct subgroups of college students based on their engagement in health-related behaviors. Based on fit indices a three-profile solution showed the best fit: low (N = 196), moderate (N = 54), and high (N = 23). Two one-way ANOVAs examined whether profile membership predicted engagement and perceived effectiveness of COVID-19 safety measures. Participants in the low health-related behaviors profile engaged in preventative measures more than students in the other two profiles. However, profile membership did not predict perceived effectiveness of preventative behaviors. Taken together, our results indicate that college students reporting lower levels of health-related behaviors engage in more preventative measures during a pandemic. Understanding distinct health-related behaviors profiles among college students, and their links with COVID-preventative health-related behaviors, can inform prevention strategies.

The Associations Between Drinking Motives and Alcohol-Related Consequences in Graduate Students from Universities in the United States.

Many college students "mature out" of heavy drinking when they graduate. Yet, those who go onto graduate education report engaging in problematic drinking patterns. Drinking motives are one factor that uniquely predicts problematic drinking patterns in college students. Evidence suggests that these unique associations also generalize to individuals' post-college, specifically between drinking motives and specific alcohol-related consequences. However, no research to date has examined the association between drinking motives and alcohol-related consequences in graduate students. The current study aimed to examine the unique associations between drinking motives, and drinks per week and specific alcohol-related consequences. Participants included 330 graduate students from various universities in the United States, recruited through social media. The majority of participants were White (71.3%), 54.9% female identifying, with a mean age of 26. Results revealed that conformity motives were positively associated with drinks per week, self-control consequences, self-care consequences, risky consequences, academic/occupational consequences, and blackout consequences. Social motives were negatively associated with interpersonal consequences and academic/occupational consequences. Enhancement motives were negatively associated with drinks per week, and positively associated with academic/occupational consequences. However, coping motivation was not associated with any of the outcomes. These findings highlight the need to further understand how drinking motives influence specific types of alcohol-related consequences as these associations change post-college. Results can be used to better inform future prevention and interventions for this population.

Development, feasibility, and acceptability of SPoRT: a dating violence and sexual risk prevention intervention for college student-athletes.

BACKGROUND: Student-athletes are one subgroup of college students in the USA at risk for dating violence and sexual risk behaviors. Despite this, research on student-athletes' dating behaviors is limited; existing research pertains primarily to the National Collegiate Athletic Association (NCAA) Division I athletes and focuses on male student-athletes as perpetrators of dating and sexual violence. While some existing programs aim to reduce dating violence and promote healthy relationships, these programs are education based, and not tailored to the specific strengths and challenges of student-athletes. We therefore designed Supporting Prevention in Relationships for Teams (SPoRT), a novel, four-session prevention intervention for Division III student-athletes of all genders to reduce dating violence and sexual risk behavior by targeting knowledge and skills identified in pilot research, incorporating psychoeducation with techniques from cognitive-behavioral therapy, mindfulness, bystander intervention, and normative feedback. METHODS: This study represents stage 1 of the National Institutes of Health (NIH) Stage Model for Behavioral Intervention Development, evaluating the feasibility and acceptability of SPoRT. We describe the development, content, and proposed delivery methods for SPoRT and evaluated the feasibility and acceptability of the program using a mixed-methods approach. Thirty college student-athletes (12 men, 18 women) completed questionnaires and participated in focus groups to provide feedback on the program's length, timing, group size and dynamics, content, and suggestions for making the SPoRT prevention intervention more feasible and acceptable. RESULTS: Our recruitment procedures were successful, and participants rated the program as feasible in terms of delivery methods and logistics. Participants liked that SPoRT was developed based on pilot data collected from student-athletes, brief, and skills based and tailored to athletic team needs. SPoRT was perceived as appropriate and relevant to student-athlete needs in terms of dating violence and sexual risk prevention knowledge and skills. Most participants (63%) rated the program as "excellent" and said they would recommend it to others. CONCLUSIONS: We found SPoRT to be both feasible and acceptable in terms of content and delivery. Suggested modifications will be incorporated into the SPoRT healthy relationships prevention intervention to be tested in an NIH Stage 1 efficacy trial.

Heterogeneous splicing patterns resulting from KIF5A variants associated with amyotrophic lateral sclerosis.

Single-nucleotide variants (SNVs) in the gene encoding Kinesin Family Member 5A (KIF5A), a neuronal motor protein involved in anterograde transport along microtubules, have been associated with amyotrophic lateral sclerosis (ALS). ALS is a rapidly progressive and fatal neurodegenerative disease that primarily affects the motor neurons. Numerous ALS-associated KIF5A SNVs are clustered near the splice-site junctions of the penultimate exon 27 and are predicted to alter the carboxy-terminal (C-term) cargo-binding domain of KIF5A. Mis-splicing of exon 27, resulting in exon exclusion, is proposed to be the mechanism by which these SNVs cause ALS. Whether all SNVs proximal to exon 27 result in exon exclusion is unclear. To address this question, we designed an in vitro minigene splicing assay in human embryonic kidney 293 cells, which revealed heterogeneous site-specific effects on splicing: only 5' splice-site (5'ss) SNVs resulted in exon skipping. We also quantified splicing in select clustered, regularly interspaced, short palindromic repeats-edited human stem cells, differentiated to motor neurons, and in neuronal tissues from a 5'ss SNV knock-in mouse, which showed the same result. Moreover, the survival of representative 3' splice site, 5'ss, and truncated C-term variant KIF5A (v-KIF5A) motor neurons was severely reduced compared with wild-type motor neurons, and overt morphological changes were apparent. While the total KIF5A mRNA levels were comparable across the cell lines, the total KIF5A protein levels were decreased for v-KIF5A lines, suggesting an impairment of protein synthesis or stability. Thus, despite the heterogeneous effect on ribonucleic acid splicing, KIF5A SNVs similarly reduce the availability of the KIF5A protein, leading to axonal transport defects and motor neuron pathology.

Catalytic, Sulfur-Free Chain Transfer Agents That Alter the Mechanical Properties of Cross-Linked Photopolymers.

Thermosetting materials generated by photopolymerization frequently suffer from significant shrinkage stress, are often brittle, and have a limited range of mechanical properties. Various classes of chain transfer agents (CTAs) have been investigated and developed to reduce the cross-linking density of photopolymers by terminating chains and initiating new chains in situ. Although CTAs are successful in manipulating the mechanical properties of photopolymers, they are traditionally consumed during the polymerization and are therefore required in high loadings (up to 20 wt % of the total formulation). Moreover, traditional CTAs frequently contain sulfur, which is malodorous and can create unstable formulations. Presented here is a catalytic, sulfur-free CTA that can be added in ppm quantities to existing commercial monomer feedstocks to create photopolymers similar to those prepared using traditional CTAs, but at 10 000-fold lower loadings. These catalysts, which are based on macrocyclic cobaloximes, were found to tunably reduce the molecular weight of the chain proportional to catalyst loading. It was shown, using only commercial monomers, that this catalyst could reduce the glass-transition temperature (Tg), rubbery modulus (E'rubbery), and stiffness of a cross-linked photopolymer while utilizing identical processing conditions and keeping 99.99 wt % of the formulation the same.

The Role of Coping Behaviors and Intoxication in Trauma Symptomology Subsequent to Sexual Victimization.

College women are at an elevated risk for sexual victimization (SV) and secondary physical and psychological consequences. While some women experience negative outcomes such as posttraumatic stress disorder (PTSD), others experience reduced or complete absence of distress following SV. The variation in outcomes may be associated with the victim's level of intoxication, which may in turn affect their processing of and coping with the event. We examined the effects of SV severity on PTSD via coping and intoxication using a moderated mediation analysis among female college students (N = 375). Results demonstrate that coping mediates the association between SV severity and PTSD symptomology; however, intoxication did not moderate these associations. Results suggest that regardless of intoxication, SV severity influences various coping styles and plays an important role in a victim's adjustment post-victimization.

A multi-stage fusion framework to classify breast lesions using deep learning and radiomics features computed from four-view mammograms.

BACKGROUND: Developing computer aided diagnosis (CAD) schemes of mammograms to classify between malignant and benign breast lesions has attracted a lot of research attention over the last several decades. However, unlike radiologists who make diagnostic decisions based on the fusion of image features extracted from multi-view mammograms, most CAD schemes are single-view-based schemes, which limit CAD performance and clinical utility. PURPOSE: This study aims to develop and test a novel CAD framework that optimally fuses information extracted from ipsilateral views of bilateral mammograms using both deep transfer learning (DTL) and radiomics feature extraction methods. METHODS: An image dataset containing 353 benign and 611 malignant cases is assembled. Each case contains four images: the craniocaudal (CC) and mediolateral oblique (MLO) view of the left and right breast. First, we extract four matching regions of interest (ROIs) from images that surround centers of two suspicious lesion regions seen in CC and MLO views, as well as matching ROIs in the contralateral breasts. Next, the handcrafted radiomics (HCRs) features and VGG16 model-generated automated features are extracted from each ROI resulting in eight feature vectors. Then, after reducing feature dimensionality and quantifying the bilateral and ipsilateral asymmetry of four ROIs to yield four new feature vectors, we test four fusion methods to build three support vector machine (SVM) classifiers by an optimal fusion of asymmetrical image features extracted from four view images. RESULTS: Using a 10-fold cross-validation method, results show that a SVM classifier trained using an optimal fusion of four view images yields the highest classification performance (AUC = 0.876 ± 0.031), which significantly outperforms SVM classifiers trained using one projection view alone, AUC = 0.817 ± 0.026 and 0.792 ± 0.026 for the CC and MLO view of bilateral mammograms, respectively (p < 0.001). CONCLUSIONS: The study demonstrates that the shift from single-view CAD to four-view CAD and the inclusion of both DTL and radiomics features significantly increases CAD performance in distinguishing between malignant and benign breast lesions.

Quantifying Intracellular Nanoparticle Distributions with Three-Dimensional Super-Resolution Microscopy.

Super-resolution microscopy can transform our understanding of nanoparticle-cell interactions. Here, we established a super-resolution imaging technology to visualize nanoparticle distributions inside mammalian cells. The cells were exposed to metallic nanoparticles and then embedded within different swellable hydrogels to enable quantitative three-dimensional (3D) imaging approaching electron-microscopy-like resolution using a standard light microscope. By exploiting the nanoparticles' light scattering properties, we demonstrated quantitative label-free imaging of intracellular nanoparticles with ultrastructural context. We confirmed the compatibility of two expansion microscopy protocols, protein retention and pan-expansion microscopy, with nanoparticle uptake studies. We validated relative differences between nanoparticle cellular accumulation for various surface modifications using mass spectrometry and determined the intracellular nanoparticle spatial distribution in 3D for entire single cells. This super-resolution imaging platform technology may be broadly used to understand the nanoparticle intracellular fate in fundamental and applied studies to potentially inform the engineering of safer and more effective nanomedicines.

Reassessing the Confluence Model of Men's Risk for Sexual Aggression.

Reassessing the confluence model of men's risk for sexual aggression-the confluence model of sexual aggression has been widely used to study men's risk for perpetrating sexual violence. Over time, researchers have attempted to expand this model to improve its predictive utility. Unfortunately, this work has continued to produce similar results with only slight improvements in prediction at best. One explanation for the inability to enhance the model could be due to changes in the dating landscape and shifts in beliefs about gender roles. Therefore, the current study aims to reassess the confluence model using a more contemporary construct, hostile sexism, in an effort to improve the predictive utility of the confluence model of sexual aggression. Participants were 258 college men recruited from a medium-sized public university in the northeastern United States, using an online participant pool of students who volunteered to participate as part of a requirement for a psychology course. Structural equation modeling using mean- and variance-adjusted weighted least squares estimation indicated that the confluence of hostile sexism and impersonal sex appears to be a better predictor of sexual aggression in comparison to the confluence of hostile masculinity and impersonal sex. The results suggest that replacing hostile masculinity with hostile sexism may produce a model that is better able to predict men's risk for perpetrating sexual aggression. These results can provide insight for future iterations of the confluence model, which may include hostile sexism as a core construct. Attitudes that stem from hostile sexism may be a beneficial target for future interventions designed to decrease the frequency of perpetration.

Understanding Attraction, Behavior, and Identity in the Asexual Community.

Models of sexuality have evolved substantially in the past several decades through the inclusion of new aspects which were previously overlooked. Components such as romantic attraction and behavior have also traditionally been included in models of sexuality. However, romantic and sexual orientations do not coincide for all individuals. A population for which this is true and one that has developed a robust language for discussing romantic orientation is the asexual community. The current study aimed to explore romantic and sexual orientation through patterns found within the factors of attraction, behavior, and identity in the asexual community. The current sample composed of individuals who identified as asexual (N = 306, Mage = 27.1) was 61% female, 13% non-binary, and 10% self-described or used multiple labels. Within this sample, aspects of sexual and romantic orientations and experiences were measured, including fluidity, the quantity and type of self-identified labels, desire for romance or sex, and the role of contextual influences on these experiences. These aspects were used as the primary characteristics to construct participant profiles, both complete profiles and factor specific (attraction, behavior, identity). t-distributed stochastic neighbor embedding (tSNE) was used to find patterns of similarity between individual participant profiles. Overall, it appeared that attraction was the factor most closely associated with overall experiences; however, substantial variability existed between participants. These findings provide a mechanism for better understanding of some nuances of romantic and sexual orientation and may be a useful first step toward future inquiry and hypothesis generation.

Access to Substituted Indenol Ethers via a Regioselective Intermolecular Carbopalladation Cascade.

Alkyne carbopalladation reactions can rapidly generate multiple new C-C bonds; however, regioselectivity is challenging for intermolecular variants. Using ynol ethers, we observe complete regiocontrol of migratory insertion followed by a second migratory insertion with a pendant alkene to turn-over the catalytic cycle. The resulting products are oligosubstituted 1-indenol ethers with defined stereochemistry based on the initial alkene geometry. Blocking ß-hydride elimination allowed for C-H and C-C reductive elimination steps for catalyst turnover.

Simultaneous Bilateral Ovarian Torsion in a Transgender Patient.

Ovarian torsions are a diagnostic challenge in the emergency room that result in significant morbidity when missed. Ovarian torsions have a low incidence and can present with atypical and variable symptoms. As such, recognition of ovarian torsion is a key skill that emergency clinicians must master. Typically, ovarian torsion is a unilateral finding but there have been cases of bilateral torsions noted previously in the literature. These rare cases tend to have higher morbidity making it even more important to recognize. This case is unique in that it was a transgendered patient developing synchronous bilateral ovarian torsions, which, to our knowledge, has not been previously described in the literature. This case illustrates the potential for unique presentations of this condition. We advise a high index of suspicion until more data, both objective and subjective gestalt, is known.

Improving Performance of Breast Lesion Classification Using a ResNet50 Model Optimized with a Novel Attention Mechanism.

Background: The accurate classification between malignant and benign breast lesions detected on mammograms is a crucial but difficult challenge for reducing false-positive recall rates and improving the efficacy of breast cancer screening. Objective: This study aims to optimize a new deep transfer learning model by implementing a novel attention mechanism in order to improve the accuracy of breast lesion classification. Methods: ResNet50 is selected as the base model to develop a new deep transfer learning model. To enhance the accuracy of breast lesion classification, we propose adding a convolutional block attention module (CBAM) to the standard ResNet50 model and optimizing a new model for this task. We assembled a large dataset with 4280 mammograms depicting suspicious soft-tissue mass-type lesions. A region of interest (ROI) is extracted from each image based on lesion center. Among them, 2480 and 1800 ROIs depict verified benign and malignant lesions, respectively. The image dataset is randomly split into two subsets with a ratio of 9:1 five times to train and test two ResNet50 models with and without using CBAM. Results: Using the area under ROC curve (AUC) as an evaluation index, the new CBAM-based ResNet50 model yields AUC = 0.866 ± 0.015, which is significantly higher than that obtained by the standard ResNet50 model (AUC = 0.772 ± 0.008) (p < 0.01). Conclusion: This study demonstrates that although deep transfer learning technology attracted broad research interest in medical-imaging informatic fields, adding a new attention mechanism to optimize deep transfer learning models for specific application tasks can play an important role in further improving model performances.

Applying artificial intelligence technology to assist with breast cancer diagnosis and prognosis prediction.

Breast cancer remains the most diagnosed cancer in women. Advances in medical imaging modalities and technologies have greatly aided in the early detection of breast cancer and the decline of patient mortality rates. However, reading and interpreting breast images remains difficult due to the high heterogeneity of breast tumors and fibro-glandular tissue, which results in lower cancer detection sensitivity and specificity and large inter-reader variability. In order to help overcome these clinical challenges, researchers have made great efforts to develop computer-aided detection and/or diagnosis (CAD) schemes of breast images to provide radiologists with decision-making support tools. Recent rapid advances in high throughput data analysis methods and artificial intelligence (AI) technologies, particularly radiomics and deep learning techniques, have led to an exponential increase in the development of new AI-based models of breast images that cover a broad range of application topics. In this review paper, we focus on reviewing recent advances in better understanding the association between radiomics features and tumor microenvironment and the progress in developing new AI-based quantitative image feature analysis models in three realms of breast cancer: predicting breast cancer risk, the likelihood of tumor malignancy, and tumor response to treatment. The outlook and three major challenges of applying new AI-based models of breast images to clinical practice are also discussed. Through this review we conclude that although developing new AI-based models of breast images has achieved significant progress and promising results, several obstacles to applying these new AI-based models to clinical practice remain. Therefore, more research effort is needed in future studies.

A Comparison of Computer-Aided Diagnosis Schemes Optimized Using Radiomics and Deep Transfer Learning Methods.

Objective: Radiomics and deep transfer learning are two popular technologies used to develop computer-aided detection and diagnosis (CAD) schemes of medical images. This study aims to investigate and to compare the advantages and the potential limitations of applying these two technologies in developing CAD schemes. Methods: A relatively large and diverse retrospective dataset including 3000 digital mammograms was assembled in which 1496 images depicted malignant lesions and 1504 images depicted benign lesions. Two CAD schemes were developed to classify breast lesions. The first scheme was developed using four steps namely, applying an adaptive multi-layer topographic region growing algorithm to segment lesions, computing initial radiomics features, applying a principal component algorithm to generate an optimal feature vector, and building a support vector machine classifier. The second CAD scheme was built based on a pre-trained residual net architecture (ResNet50) as a transfer learning model to classify breast lesions. Both CAD schemes were trained and tested using a 10-fold cross-validation method. Several score fusion methods were also investigated to classify breast lesions. CAD performances were evaluated and compared by the areas under the ROC curve (AUC). Results: The ResNet50 model-based CAD scheme yielded AUC = 0.85 ± 0.02, which was significantly higher than the radiomics feature-based CAD scheme with AUC = 0.77 ± 0.02 (p < 0.01). Additionally, the fusion of classification scores generated by the two CAD schemes did not further improve classification performance. Conclusion: This study demonstrates that using deep transfer learning is more efficient to develop CAD schemes and it enables a higher lesion classification performance than CAD schemes developed using radiomics-based technology.

A baseline of copper associated with antifouling paint in marinas within a large fjord estuary.

Marinas have been shown to contribute elevated concentrations of copper (Cu) to marine waters. The Cu can come primarily from antifouling paints which are designed to discourage biofouling of boat hulls. Legislation in Washington State, USA is being developed to limit or regulate the amount and rate of diffusion of Cu from antifouling paints. This study provides baseline data for Cu in five marinas of different configuration and size within Puget Sound, a large fjord estuary. Samples were collected over a year from multiple environmental media. We find strong evidence that Cu accumulates inside marinas to higher concentrations than outside marinas. Marinas that are more enclosed accumulated higher concentrations of Cu than more open marinas. Using a power analysis, we assessed the adequacy of the baseline dataset to measure progress as a result of future legislation towards the reduction of Cu to Puget Sound from marinas.

Improving mammography lesion classification by optimal fusion of handcrafted and deep transfer learning features.

Objective.Handcrafted radiomics features or deep learning model-generated automated features are commonly used to develop computer-aided diagnosis schemes of medical images. The objective of this study is to test the hypothesis that handcrafted and automated features contain complementary classification information and fusion of these two types of features can improve CAD performance.Approach.We retrospectively assembled a dataset involving 1535 lesions (740 malignant and 795 benign). Regions of interest (ROI) surrounding suspicious lesions are extracted and two types of features are computed from each ROI. The first one includes 40 radiomic features and the second one includes automated features computed from a VGG16 network using a transfer learning method. A single channel ROI image is converted to three channel pseudo-ROI images by stacking the original image, a bilateral filtered image, and a histogram equalized image. Two VGG16 models using pseudo-ROIs and 3 stacked original ROIs without pre-processing are used to extract automated features. Five linear support vector machines (SVM) are built using the optimally selected feature vectors from the handcrafted features, two sets of VGG16 model-generated automated features, and the fusion of handcrafted and each set of automated features, respectively.Main Results.Using a 10-fold cross-validation, the fusion SVM using pseudo-ROIs yields the highest lesion classification performance with area under ROC curve (AUC = 0.756 ± 0.042), which is significantly higher than those yielded by other SVMs trained using handcrafted or automated features only (p < 0.05).Significance.This study demonstrates that both handcrafted and automated futures contain useful information to classify breast lesions. Fusion of these two types of features can further increase CAD performance.

Applying a radiomics-based CAD scheme to classify between malignant and benign pancreatic tumors using CT images.

BACKGROUND: Pancreatic cancer is one of the most aggressive cancers with approximate 10% five-year survival rate. To reduce mortality rate, accurate detection and diagnose of suspicious pancreatic tumors at an early stage plays an important role. OBJECTIVE: To develop and test a new radiomics-based computer-aided diagnosis (CAD) scheme of computed tomography (CT) images to detect and classify suspicious pancreatic tumors. METHODS: A retrospective dataset consisting of 77 patients who had suspicious pancreatic tumors detected on CT images was assembled in which 33 tumors are malignant. A CAD scheme was developed using the following 5 steps namely, (1) apply an image pre-processing algorithm to filter and reduce image noise, (2) use a deep learning model to detect and segment pancreas region, (3) apply a modified region growing algorithm to segment tumor region, (4) compute and select optimal radiomics features, and (5) train and test a support vector machine (SVM) model to classify the detected pancreatic tumor using a leave-one-case-out cross-validation method. RESULTS: By using the area under receiver operating characteristic (ROC) curve (AUC) as an evaluation index, SVM model yields AUC = 0.750 with 95% confidence interval [0.624, 0.885] to classify pancreatic tumors. CONCLUSIONS: Study results indicate that radiomics features computed from CT images contain useful information associated with risk of tumor malignancy. This study also built a foundation to support further effort to develop and optimize CAD schemes with more advanced image processing and machine learning methods to more accurately and robustly detect and classify pancreatic tumors in future.