Publications by authors named "Foran D"
Cardiovascular disease is a leading cause of morbidity and mortality. New research elucidates increasingly complex relationships between cardiac and metabolic health, giving rise to new possible therapeutic targets. Sphingolipids are a heterogeneous class of bioactive lipids with critical roles in normal human physiology.
View Article and Find Full Text PDFBackground: Academic institutions benefit from researchers adopting leadership positions and, subsequently, leadership development programmes are of increasing importance. Despite this, no evaluation of the evidence basis for leadership development programmes for healthcare researchers has been conducted. In this study, the authors reviewed leadership development programmes for healthcare researchers and aimed to identify their impact and the factors which influenced this impact.
View Article and Find Full Text PDFArticle Synopsis
- Large-scale collaboration in oncology research is essential for advancing cancer biology, precision oncology, and population sciences, which requires innovative data management and analytic tools.
- The informatics community plays a crucial role in automating the organization of diverse clinical data types, including molecular tests and diagnostic imaging, to address the complexities of cancer progression.
- The paper presents a new Clinical & Research Data Warehouse (CRDW) that supports multimodal data, such as genomics and radiology images, integrating machine-learning tools for deeper insights into tumor characteristics beyond traditional methods.
Infertility affects 15% of couples worldwide and in approximately 50% of cases the cause is secondary to an abnormality of the sperm. However, treatment options for male infertility are limited and empirical use of hormone stimulation has been utilised. We review the contemporary data regarding the application of hormone stimulation to treat male infertility.
View Article and Find Full Text PDFArticle Synopsis
- Hormonal therapy's role in improving sperm retrieval success rates for men with non-obstructive azoospermia (NOA) is debated, especially since standard guidelines do not recommend its use but many patients still try it before surgical retrieval.
- A systematic review and meta-analysis aimed to compare sperm retrieval rates in men with NOA receiving hormonal therapy versus those who did not, while also examining differences based on hormone levels.
- Findings from 22 studies indicated that hormonal therapy significantly increased sperm retrieval rates, with an odds ratio of 1.96, suggesting it can be beneficial when used prior to surgical sperm extraction.
Article Synopsis
- Population-based cancer registries in the U.S. gather comprehensive data on cancer cases, including patient demographics, tumor details, treatments, and outcomes, to support cancer statistics and research.* -
- The project aims to enhance the NCI's SEER registry by integrating high-quality biospecimen data through digital pathology and advanced imaging techniques, promoting more consistent and objective analysis of cancer data.* -
- A curated repository of digitized pathology images has been established, alongside the development of automated tools for creating population cohorts and visualizing key features, ultimately improving the retrieval and analysis of cancer specimens.*
Nonrandom selection and multiple blood feeding of human hosts by Anopheles mosquitoes may exacerbate malaria transmission. Both patterns of blood feeding and their relationship to malaria epidemiology were investigated in Anopheles vectors in Papua New Guinea (PNG). Blood samples from humans and mosquito blood meals were collected in villages and human genetic profiles ("fingerprints") were analyzed by genotyping 23 microsatellites and a sex-specific marker.
View Article and Find Full Text PDFBackground: Blackcurrant is rich in anthocyanins that may protect against exercise-induced muscle damage (EIMD) and facilitate a faster recovery of muscle function. We examined the effects of New Zealand blackcurrant (NZBC) extract on indices of muscle damage and recovery following a bout of strenuous isokinetic resistance exercise.
Methods: Using a double-blind, randomised, placebo controlled, parallel design, twenty-seven healthy participants received either a 3 g·day NZBC extract ( = 14) or the placebo (PLA) ( = 13) for 8 days prior to and 4 days following 60 strenuous concentric and eccentric contractions of the biceps brachii muscle on an isokinetic dynamometer.
Background: Opioid misuse is a widespread public health problem, and opioids are often prescribed in the dental environment. These recommendations provide alternatives to opioids to reduce or eliminate dental procedure-related acute pain.
Methods: A multidisciplinary working group developed these clinical recommendations to specifically address procedure-related acute pain.
Liver disease classification from ultrasound using multi-scale CNN.
Int J Comput Assist Radiol Surg
September 2021
Article Synopsis
- Ultrasound is the top choice for diagnosing fatty liver disease due to its noninvasive nature, but traditional methods lack objectivity and accuracy.
- The study introduces an advanced deep learning model that leverages various image processing techniques and multi-feature inputs to enhance classification accuracy for nonalcoholic fatty liver disease from ultrasound data.
- Results show the model achieves over 90% classification accuracy and a 97.8% area under the ROC curve, indicating significant improvements over traditional CNN and machine learning approaches, highlighting its potential in clinical diagnostics.
Chest X-ray image phase features for improved diagnosis of COVID-19 using convolutional neural network.
Int J Comput Assist Radiol Surg
February 2021
Article Synopsis
- The COVID-19 pandemic has increased the need for effective and accessible diagnostic methods, as traditional test kits are limited; chest X-rays (CXR) are identified as a promising alternative due to their speed, cost-effectiveness, and portability.
- This study presents a new multi-feature convolutional neural network (CNN) designed to improve the classification of COVID-19 from enhanced CXR images, combining standard and enhanced imaging techniques.
- The proposed model demonstrated high accuracy in classifying CXR scans—with 95.57% average accuracy and 99% metrics for COVID-19 cases—indicating its potential as a reliable tool for aiding radiologists in diagnosing COVID-19.
Non-obstructive azoospermia is reported to affect 1 in 100 men, and despite advances in surgical practice, the succesful sperm retrieval rate for microdissection testicular sperm extraction surgery (mTESE) is only 46%. This article reviews the potential causes for mTESE failure and provides a management strategy to guide the clinicians on how to treat this challenging cohort of patients.
View Article and Find Full Text PDFArticle Synopsis
- * Results show that Black women had lower expression of certain adipokine receptors, and lower levels of adipokines were linked to higher tumor grades, larger tumor sizes, and aggressive cancer types like ER-, HER2-enriched, and triple-negative.
- * The findings suggest that lower expression of these adipokines and receptors could indicate more aggressive breast cancer phenotypes, necessitating further research to understand their role in breast cancer prognosis.
AI in Medical Imaging Informatics: Current Challenges and Future Directions.
IEEE J Biomed Health Inform
July 2020
This paper reviews state-of-the-art research solutions across the spectrum of medical imaging informatics, discusses clinical translation, and provides future directions for advancing clinical practice. More specifically, it summarizes advances in medical imaging acquisition technologies for different modalities, highlighting the necessity for efficient medical data management strategies in the context of AI in big healthcare data analytics. It then provides a synopsis of contemporary and emerging algorithmic methods for disease classification and organ/ tissue segmentation, focusing on AI and deep learning architectures that have already become the de facto approach.
View Article and Find Full Text PDFTamoxifen is a selective oestrogen receptor modulator (SERM). SERMs act on oestrogen receptors to inhibit oestradiol mediated negative feedback on the hypothalamic-pituitary-gonadal (HPG) axis, thereby upregulating gonadotrophin secretion and release from the pituitary. Hence, Tamoxifen is used to upregulate activation of the HPG axis in the treatment of male-factor infertility.
View Article and Find Full Text PDFArticle Synopsis
- The study investigates how increased body fat might relate to more severe forms of breast cancer, focusing on the roles of specific proteins called adipokines, particularly leptin and adiponectin, and their receptors.
- Researchers analyzed breast tumor samples from 720 women, predominantly Black, using immunohistochemistry to measure the expression of these proteins and linked this data to the aggressiveness of the tumors based on various clinical factors.
- Results indicate that lower levels of leptin receptor expression in tumors were associated with more aggressive cancer types, especially in ER-negative and triple-negative breast cancers, suggesting a potential mechanism for the variations seen in tumor characteristics among individuals.
Pink tooth of Mummery is typically found after trauma. However, this case report describes an unusual occurrence of pink tooth in a 67-year-old Caucasian male after fixed partial denture (FPD) tooth preparation. Pink tooth in this case may be due to one or more factors: tooth reduction and heat generation during tooth preparation; heat generation during polymerization of provisional material; and hyperocclusion of a provisional FPD.
View Article and Find Full Text PDFArticle Synopsis
- - The study focuses on improving the grading of prostatic adenocarcinoma by analyzing histopathological image features in relation to survival outcomes using various advanced methods, like CNNs.
- - Researchers utilized multiple techniques, including texture analysis and convolutional neural networks, to assess how these image features correlate with established prognostic factors like Gleason patterns and PSA levels.
- - Findings indicate that a specific CNN method yielded the most accurate assessment of prostate cancer recurrence, suggesting potential for broader applications in predicting outcomes across different cancer types in future research.
Soil, being diverse and ubiquitous, can potentially link a suspect or victim to a crime scene. Recently scientists have examined the microbial makeup of soil for determining its origin, and differentiating soil samples is well-established. However, when soil is transferred to evidence its microbial makeup may change over time, leading to false exclusions.
View Article and Find Full Text PDFTwo well-differentiated pancreatic neuroendocrine tumor mouse models.
Cell Death Differ
January 2020
Article Synopsis
- MEN1 is a genetic syndrome that causes patients to develop neuroendocrine tumors, specifically pancreatic neuroendocrine tumors (PanNETs), and other mutations may accelerate this process.
- Researchers created two mouse models (MPR and MPM) that showed accelerated tumor development when both Men1 and Pten genes were inactivated, resulting in tumor characteristics similar to human MEN1 cases.
- The study found that using the mTOR inhibitor rapamycin slowed tumor growth in these mouse models, providing a new platform for testing treatments targeting specific cancer pathways related to neuroendocrine tumors.
Unsupervised Domain Adaptation for Classification of Histopathology Whole-Slide Images.
Front Bioeng Biotechnol
May 2019
Article Synopsis
- Computational image analysis improves the accuracy and consistency of cancer diagnoses by applying algorithms to digitized histopathology specimens, but variations in sample preparation across labs can affect performance.
- To address these challenges, the study proposes unsupervised domain adaptation to enable effective transfer of diagnostic knowledge without needing new labels or annotations.
- The researchers evaluate two strategies—color normalization and adversarial training—finding that adversarial training using convolutional neural networks enhances classification results significantly across various testing conditions.
Article Synopsis
- - Prostate cancer is the most prevalent nonskin-related cancer in the U.S., particularly among men, and the Gleason score is a key indicator for predicting patient outcomes.
- - Recent advancements in genomic sequencing have helped in classifying prostate cancer, showing that patients with a Gleason score of 7 experience varied disease recurrence and survival rates.
- - The study developed a unified system that leverages deep neural networks to extract computational biomarkers, achieving superior predictive results compared to traditional clinical methods, underscoring the potential for neural networks in improving prostate cancer treatment and precision medicine.
Article Synopsis
- Prostate cancer is the most prevalent non-skin cancer in men, affecting 1 in 7, and treatment choices are complex, factoring in patient health and potential side effects.
- The Gleason score, which assesses cancer based on prostate gland patterns, is a key predictor of patient outcomes, but there's variability in recurrence for those with a score of 7.
- The study investigates correlations between histopathology images and genomic data to find better prognostic markers, showing that integrating image features with advanced models can improve predictions of disease recurrence compared to traditional methods.
We propose a sparse Convolutional Autoencoder (CAE) for simultaneous nucleus detection and feature extraction in histopathology tissue images. Our CAE detects and encodes nuclei in image patches in tissue images into sparse feature maps that encode both the location and appearance of nuclei. A primary contribution of our work is the development of an unsupervised detection network by using the characteristics of histopathology image patches.
View Article and Find Full Text PDFAdversarial Domain Adaptation for Classification of Prostate Histopathology Whole-Slide Images.
Med Image Comput Comput Assist Interv
September 2018
Article Synopsis
- Automatic Gleason grading for prostate cancer histopathology is essential for accurate diagnosis and treatment, but variations in tissue preparation can hinder accuracy across different institutions.
- The authors propose using unsupervised domain adaptation, allowing the transfer of knowledge from a trained model without needing labeled target images, which enhances the model's performance across different histopathology slides.
- Their method, validated on two prostate cancer datasets, demonstrates significant improvement in predicting Gleason scores compared to standard models, thanks to adversarial training and a Siamese architecture designed to maintain consistency in feature space.