Premier Research Group

Brain delivery of lipophilic antioxidants is limited by poor solubility and restricted transport across the blood-brain barrier (BBB). To address this challenge, we developed a plant-derived glucosylceramide (GlcCer)-based w/o/w nanoemulsion designed to exploit glucose transporter (GLUT)-mediated uptake for brain targeting. Unlike synthetic glucose conjugates, GlcCer provides intrinsic amphiphilicity and biocompatibility, enabling self-assembly into stable, surfactant-free nanoemulsions incorporating astaxanthin, curcumin, and sesame oil through a cost-effective and scalable process. The resulting GlcCer-functionalized nanoemulsions demonstrated enhanced physicochemical stability and low cytotoxicity in SH-SY5Y and hCMEC/D3 cells. Quantitatively, the sesame oil-based targeted formulation achieved a 1.6-fold increase in neuronal uptake, stronger ROS suppression, and a 1.4-fold improvement in transendothelial transport compared with nontargeted controls. These findings identify GlcCer-based w/o/w nanoemulsions as a promising and scalable platform for brain-targeted antioxidant delivery.

Abstract We report the development of a novel coreflood testing quantitative evaluation which correlates data science and image processing of slice-by-slice microcore volume consumption with the differential pressure profile achieved during coreflood tests, facilitating the design of optimal treatments for acid stimulation. The outcome of microcore volume consumption studies obtained via high resolution CT scans are broadly applicable, enabling the design of customized acidizing stimulation treatments with minimal propensity for formation damage. The effects of physical testing conditions and fluid composition on microscopic wormhole profile development and core consumption volume using this original method are studied. We report detailed procedures of the coreflood tests, including carbonate core characterization, mineralogy, acid treatment specifications, and resultant wormhole profiles. Particularly, we accurately pinpoint differential pressure variations during the coreflood process in a stepwise manner via in-depth data and imaging analysis of high-resolution CT scans and correlate the changes to the corresponding fractional microcore volumes being consumed by the acid etching reaction at each incremental step. We present the differential pressure changes resulting from the high temperature acid etching reaction during the course of core flooding with corresponding microcore volume consumption along the core-axis in a slice-by-slice mode (with a precision of 0.02"/slice per CT scan). The acidizing fluids used in these tests were diligently formulated for effective modulation of the acid reaction rate with carbonate matrices at elevated temperatures. The resultant wormhole profiles are characterized and optimized under a given set of test parameters (temperature, differential pressure, acid fluid injection rate and treatment fluid composition) using the slice-by-slice technique. The effect of several acid packages on treatment efficiency based on microcore volume consumption is evaluated. In addition to 3D wormhole profiles, the kinetic insights of the acid etching reaction process from head-to-toe of the core are elucidated with microscopic precision, offering previously unavailable details, including the fractional contribution of each slice to the overall microcore volume consumed. By employing in-depth data and imaging analysis, a strong correlation between microcore volume consumption and the incremental differential pressure change is established. This important relationship provides critical light to the strategies leading to the most effective pore volume breakthrough performances for any given core that should be of critical importance when optimizing acid stimulation job designs. For the first time, stepwise changes in differential pressure profiles can be precisely correlated with corresponding microcore volumes consumed as an acid-coreflood etching process proceeds. This vigorous relationship, developed using an unprecedented methodology based on data science and imaging processing, offers potentially game-changing insights that can empower traditionally recommended acid coreflood testing practices, and assist in optimizing the design of reservoir acid stimulation treatments that minimize undesirable consequences of formation damage.

Alzheimer's disease (AD) is often underdiagnosed, especially at early stages when symptoms are mild, and patients may benefit from new and recently approved anti‐amyloid therapies. An updated patient care pathway may facilitate timely diagnosis and intervention. We conducted a scan of published information and convened an expert panel of health care professionals to gain insights on early AD care pathways. We developed an Early AD Patient Care Pathway as an implementation guide that includes new diagnostic and treatment modalities and addresses needs and opportunities. The Early AD Patient Care Pathway focuses on patient identification, assessment, diagnosis, treatment, and management and monitoring. Operational readiness considerations aid pathway implementation and include evaluating and addressing gaps in program and business planning, technology, education and resources, access and reimbursement, and care coordination. The pathway is adaptable to health system needs and may be further tested and refined for sustainability.HighlightsWe analyzed published information and insights from subject matter experts on care pathways for early Alzheimer's disease.We identified opportunities to improve screening of at‐risk patients, make appropriate specialist referrals, and ensure timely access and care coordination for diagnosis, treatment, and monitoring as a means of incorporating the latest diagnostics and therapy options.We developed a care pathway that promotes an integrated approach to patient care and is adaptable to various healthcare settings with input from subject matter experts.

ABSTRACT Due to the prevalence of hematoxylin and eosin (H&E) staining in routine histological preparations, understanding the factors that impact stain color characteristics is vital to attain consistently high-quality stains. In the last decade, increased use of digital pathology and image analysis (specifically by optical density [OD] measures) has provided new ways of assessing staining precision. This paper combines data from two studies that tracked H&E staining quality in both nuclear and cytoplasmic components of 12 tissues (11 human and 1 porcine) by OD after overuse of H&E staining reagents from 5 vendors. Both studies showed a decrease in eosin stain intensity by OD and visual inspection (by a histologist) with reagent overuse. This trend varied in degree by tissue type and reagent vendor. Nonetheless, staining quality for H&E staining from all vendors and for all organs remained acceptable (but not always optimal) for microscopic evaluation by the College of American Pathologists (CAP) and National Society for Histotechnology (NSH) staining criteria when stained sections were reviewed by a board-certified veterinary pathologist.

INTRODUCTION Artificial intelligence (AI) is rapidly transforming healthcare and scientific publishing. Reporting guidelines need to be updated to take into account this advance. The STROCSS Guideline 2025 update adds a new AI-focused domain to promote transparency, reproducibility, and ethical integrity in surgical observational studies involving AI. METHODS A Delphi consensus exercise was conducted to update the STROCSS guidelines. A panel of 49 surgical and scientific experts were invited to rate proposed new items. In Round 1, participants scored each item on a nine-point Likert scale and provided feedback. Items not meeting consensus were revised or discarded. RESULTS A 94% response rate occurred among participants (46/49) in the first round. Ratings were analyzed for agreement levels, and consensus was reached on all six proposed AI-related items. A revised STROCSS checklist is presented which incorporates these new AI-related items. Authors are now expected to disclose AI involvement not only in patient care but also in manuscript preparation, as exemplified by this article. CONCLUSION The STROCSS 2025 guideline provides an up-to-date framework for surgical observational studies in the era of AI. Through a robust consensus STROCSS, we have added specific reporting criteria for AI to ensure that any use of artificial intelligence in a surgical observational study is clearly documented, explained, and discussed including with respect to bias and ethics. This update will help maintain the quality, transparency, and clinical relevance of surgical observational studies, ultimately improving their educational value and trustworthiness for the surgical community.

“Can machines think?” – this is the question posed by Alan Turing in 1950 in his classic paper. In the years that followed the term artificial intelligence (AI) was adopted and the field was formally born at the Dartmouth Summer research project on AI. More recently AI has been thrust into the forefront of public consciousness via the media and a market valuation of $638bn in 2024. This is projected to further increase and PwC’s Global Artificial Intelligence Study values the potential contribution to the global economy by 2030 from AI at $15.7tr.

ABSTRACT The Eagle Ford Group is a mudrock system deposited during the Late Cretaceous across a flooded carbonate platform (referred to as the Comanche Platform in south and central Texas) that rimmed the Gulf coast region of the southern United States. During the Cenomanian–Turonian the Comanche Platform was situated at the confluence of the Western Interior Seaway and the Gulf of Mexico. Bathymetric features on the Comanche Platform include: 1) the San Marcos Arch, which separates the East Texas Basin to the northeast from the Maverick Basin further to the southwest, 2) isolated fault-bound troughs including the Karnes, Atascosa, and Gonzales Troughs, and 3) structural remnants of the down-dip Aptian through middle Cenomanian Stuart City and Sligo Reef margins. Collectively, bathymetric features and oceanographic circulation patterns resulted in a complex depositional network that affected sediment source input, bottom water redox potentials, and carbonate productivity. The onset of Oceanic Anoxic Event-2 (OAE-2) at the Cenomanian–Turonian boundary (ca. 93.5 Ma) coincides with the transition from lower Eagle Ford (LEF) to upper Eagle Ford (UEF) deposition and is marked by a positive ~2–3‰ carbon isotope excursion measured from bulk organic matter. OAE-2 also reflects a transition from anoxic organic-matter rich mudrocks typical of the LEF to oxic, carbonate-rich, and organic-matter poor mudrocks typical of the UEF. As part of a regional study of the Eagle Ford Group, 42 core have been characterized using high resolution X-ray fluorescence core scanning coupled to lithofacies, biostratigraphy, volcanic ash dating, and stratigraphic wireline log analysis. In south Texas, onset of LEF deposition at ca. 96.25 Ma is dominated by compositionally banded TOC-rich argillaceous–calcareous mudstones to wackestones deposited under anoxic bottom water conditions (elevated concentrations of Ni, Cu, Zn, Mo, and V). These TOC-rich anoxic mudstones are interbedded with diagenetic calcite cemented limestones and concretions. Importantly, anoxic conditions and maximum TOC preservation occurred in the LEF approximately 2–4 million years prior to the onset of OAE-2. At the top of the LEF, initiation of OAE-2 coincides with an increase in clay content, decrease in TOC content, decrease in redox-sensitive trace elements Mo and V, and increase in bioturbation, all of which are indicative of an increase in oxidation of bottom water. Not all cores preserve the 2–3‰ carbon isotope excursion associated with OAE-2, indicating that either erosion or non-deposition occurred during the UEF, likely the result of ocean current reworking, clinoform deposition during the high stand system of the UEF, or the result of bathymetric geometries. ‘Recovery’ from OAE-2 is marked by a decrease in carbon isotope values measured from bulk organic matter back to values typical of LEF, but the lithofacies that dominate the UEF are more oxygenated, contain less TOC, and are more carbonate-rich than observed in the LEF. Collectively, integrated core data from the Comanche Platform in south Texas demonstrate that oxic bottom waters persisted during OAE-2, which is different from the Tethys Ocean that experienced increased TOC-preservation and anoxic bottom water conditions contemporaneous with the OAE-2 carbon isotope excursion.

Breast cancer today continues to be a worthy global health concern because its early diagnosis has effective treatment means that make it come with efficient diagnostic techniques. In this paper, a new procedure is proposed to deal with breast ultrasound image segmentation with an intention of improving the identification of breast cancer lesions. The proposed methodology is built upon aU-NET architcture, a CNN structure in an encoderdecoder fashion along with skip connections; it concentrates on accurate segmentation of the specific pathologies in the ultrasound images. The dataset used incorporate breast ultrasound images which are a verydiverse collection and are thoroughly labeled for segmentation ground truth. Cross Entropy loss and Dice loss is used as the loss functions for the training process, fine-tuned with the Adam optimizer. The robustness of this method is evident through various segmentation results based on the experiment performed on different portions of the dataset; the high value of 0.72 of the dice coefficients and a test accuracy of 93% also signal viability of this work. It also establishes the possibility of deep learning approaches for breast cancer detection by using the automatic image processing. Looking at this research from this perspective will also expand the understanding of its applicability in clinical practice which, in turn, may benefit patient prognosis relying on radiologists' support in cancer detection at early stages. This research lays the foundation for the possible future studies of more complex deep learning networks in breast ultrasound imagery and offers hope for the improved diagnostic methods in this field and, in consequence, better treatment options for patients.