Deep learning to predict rapid progression of Alzheimer's disease from pooled clinical trials: A retrospective study.

The rate of progression of Alzheimer's disease (AD) differs dramatically between patients. Identifying the most is critical because when their numbers differ between treated and control groups, it distorts the outcome, making it impossible to tell whether the treatment was beneficial. Much recent effort, then, has gone into identifying RPs. We pooled de-identified placebo-arm data of three randomized controlled trials (RCTs), EXPEDITION, EXPEDITION 2, and EXPEDITION 3, provided by Eli Lilly and Company. After processing, the data included 1603 mild-to-moderate AD patients with 80 weeks of longitudinal observations on neurocognitive health, brain volumes, and amyloid-beta (Aß) levels. RPs were defined by changes in four neurocognitive/functional health measures. We built deep learning models using recurrent neural networks with attention mechanisms to predict RPs by week 80 based on varying observation periods from baseline (e.g., 12, 28 weeks). Feature importance scores for RP prediction were computed and temporal feature trajectories were compared between RPs and non-RPs. Our evaluation and analysis focused on models trained with 28 weeks of observation. The models achieved robust internal validation area under the receiver operating characteristic (AUROCs) ranging from 0.80 (95% CI 0.79-0.82) to 0.82 (0.81-0.83), and the area under the precision-recall curve (AUPRCs) from 0.34 (0.32-0.36) to 0.46 (0.44-0.49). External validation AUROCs ranged from 0.75 (0.70-0.81) to 0.83 (0.82-0.84) and AUPRCs from 0.27 (0.25-0.29) to 0.45 (0.43-0.48). Aß plasma levels, regional brain volumetry, and neurocognitive health emerged as important factors for the model prediction. In addition, the trajectories were stratified between predicted RPs and non-RPs based on factors such as ventricular volumes and neurocognitive domains. Our findings will greatly aid clinical trialists in designing tests for new medications, representing a key step toward identifying effective new AD therapies.

A combined extract containing Schisandra chinensis (SCE) reduced hepatic triglyceride accumulation in rats fed a high-sucrose diet.

Excessive hepatic lipid accumulation is closely linked to inflammation, insulin resistance, and metabolic syndromes. We hypothesized that a combined extract containing Schisandra chinensis (SCE) could alleviate hepatic lipid accumulation. Male Sprague-Dawley rats fed a high-sucrose diet (HSD) were randomly assigned to three groups (n = 6): normal diet (ND), HSD (60% kcal from sucrose), and HSD + SCE (HSD with 2.44% SCE). Liquid chromatography-tandem mass spectrometry revealed that SCE contains chlorogenic acid (5.514 ± 0.009 mg/g) and schisandrin (0.179 ± 0.002 mg/g) as bioactive components. SCE did not alter the body weight, fat mass, lean mass, or glucose levels. Strikingly, SCE effectively reduced the plasma triglyceride (TG) and hepatic TG levels compared to the HSD group. Adiposity reduction is due to decreased activity of hepatic de novo lipogenic enzymes. These results indicated that SCE has nutraceutical potential for the prevention and treatment of hepatic steatosis. Supplementary Information: The online version contains supplementary material available at 10.1007/s10068-023-01464-1.

Coherent charge hopping suppresses photoexcited small polarons in ErFeO3 by antiadiabatic formation mechanism.

Polarons are prevalent in condensed matter systems with strong electron-phonon coupling. The adiabaticity of the polaron relates to its transport properties and spatial extent. To date, only adiabatic small polaron formation has been measured following photoexcitation. The lattice reorganization energy is large enough that the first electron-optical phonon scattering event creates a small polaron without requiring substantial carrier thermalization. We measure that frustrating the iron-centered octahedra in the rare-earth orthoferrite ErFeO3 leads to antiadiabatic polaron formation. Coherent charge hopping between neighboring Fe3+─Fe2+ sites is measured with transient extreme ultraviolet spectroscopy and lasts several picoseconds before the polaron forms. The resulting small polaron formation time is an order of magnitude longer than previous measurements and indicates a shallow potential well, even in the excited state. The results emphasize the importance of considering dynamic electron-electron correlations, not just electron-phonon-induced lattice changes, for small polarons for transport, catalysis, and photoexcited applications.

CancerGPT for few shot drug pair synergy prediction using large pretrained language models.

Large language models (LLMs) have been shown to have significant potential in few-shot learning across various fields, even with minimal training data. However, their ability to generalize to unseen tasks in more complex fields, such as biology and medicine has yet to be fully evaluated. LLMs can offer a promising alternative approach for biological inference, particularly in cases where structured data and sample size are limited, by extracting prior knowledge from text corpora. Here we report our proposed few-shot learning approach, which uses LLMs to predict the synergy of drug pairs in rare tissues that lack structured data and features. Our experiments, which involved seven rare tissues from different cancer types, demonstrate that the LLM-based prediction model achieves significant accuracy with very few or zero samples. Our proposed model, the CancerGPT (with ~ 124M parameters), is comparable to the larger fine-tuned GPT-3 model (with ~ 175B parameters). Our research contributes to tackling drug pair synergy prediction in rare tissues with limited data, and also advancing the use of LLMs for biological and medical inference tasks.

Human papillomavirus infection and cardiovascular mortality: a cohort study.

BACKGROUND AND AIMS: High-risk human papillomavirus (HR-HPV) infection-a well-established risk factor for cervical cancer-has associations with cardiovascular disease (CVD). However, its relationship with CVD mortality remains uncertain. This study examined the associations between HR-HPV infection and CVD mortality. METHODS: As part of a health examination, 163 250 CVD-free Korean women (mean age: 40.2 years) underwent HR-HPV screening and were tracked for up to 17 years (median: 8.6 years). National death records identified the CVD mortality cases. Hazard ratios (HRs) and 95% confidence intervals (CIs) for CVD mortality were estimated using Cox proportional hazard regression analyses. RESULTS: During 1 380 953 person-years of follow-up, 134 CVD deaths occurred, with a mortality rate of 9.1 per 105 person-years for HR-HPV(-) women and 14.9 per 105 person-years for HR-HPV(+) women. After adjustment for traditional CVD risk factors and confounders, the HRs (95% CI) for atherosclerotic CVD (ASCVD), ischaemic heart disease (IHD), and stroke mortality in women with HR-HPV infection compared with those without infection were 3.91 (1.85-8.26), 3.74 (1.53-9.14), and 5.86 (0.86-40.11), respectively. The association between HR-HPV infection and ASCVD mortality was stronger in women with obesity than in those without (P for interaction = .006), with corresponding HRs (95% CI) of 4.81 (1.55-14.93) for obese women and 2.86 (1.04-7.88) for non-obese women. CONCLUSIONS: In this cohort study of young and middle-aged Korean women, at low risks for CVD mortality, those with HR-HPV infection had higher death rates from CVD, specifically ASCVD and IHD, with a more pronounced trend in obese individuals.

Enzymatic Synthesis of α-Glucan Microparticles Using Amylosucrases from Bifidobacterium Species and Its Physicochemical Properties.

α-Glucan microparticles (GMPs) have significant potential as high-value biomaterials in various industries. This study proposes a bottom-up approach for producing GMPs using four amylosucrases from Bifidobacterium sp. (BASs). The physicochemical characteristics of these GMPs were analyzed, and the results showed that the properties of the GMPs varied depending on the type of enzymes used in their synthesis. As common properties, all GMPs exhibited typical B-type crystal patterns and poor colloidal dispersion stability. Interestingly, differences in the physicochemical properties of GMPs were generated depending on the synthesis rate of linear α-glucan by the enzymes and the degree of polymerization (DP) distribution. Consequently, we found differences in the properties of GMPs depending on the DP distribution of linear glucans prepared with four BASs. Furthermore, we suggest that precise control of the type and characteristics of the enzymes provides the possibility of producing GMPs with tailored physicochemical properties for various industrial applications.

Current Status of Latent Tuberculosis Infection Treatment Among Pediatric Patients in Korea: Prescription and Treatment Completion.

BACKGROUND: The treatment of pediatric patients with latent tuberculosis infection (LTBI) is a crucial TB control strategy. LTBI is not a reportable communicable disease, and data regarding LTBI treatment in pediatric patients in Korea are scarce. This study aimed to investigate the prescription patterns and treatment completion rates among pediatric patients with LTBI in Korea by analyzing National Health reimbursement claims data. METHODS: We retrospectively analyzed outpatient prescription records for pediatric patients aged 18 or younger with LTBI-related diagnostic codes from 2016 to 2020. We compared the frequency of prescriptions for the standard treatment regimen (9 months of isoniazid [9H]) and an alternative treatment regimen (3 months of isoniazid plus rifampicin [3HR]). We also assessed the treatment incompletion rates by age group, treatment regimen, treatment duration, the level of medical facility, physician's specialty, and hospital location. We performed multivariable analysis to identify factors influencing treatment incompletion. RESULTS: Among the 11,362 patients who received LTBI treatment, 6,463 (56.9%) were prescribed the 9H regimen, while 4,899 (43.1%) received the 3HR regimen. Patients in the 3HR group were generally older than those in the 9H group. The proportion of 3HR regimen prescriptions significantly greater in the later period (2018-2020), in primary hospitals, under the management of non-pediatric specialists, and in metropolitan regions. The overall treatment incompletion rate was 39.7% (9H group: 46.9%, 3HR group: 30.3%). In the multivariable analysis, 9H regimen prescription was the strongest factor associated with treatment incompletion (adjusted odds ratio, 2.42; 95% confidence interval, 2.20-2.66; P < 0.001). Additionally, management in a primary hospital, a hospital's location in a non-metropolitan region, and management by a non-pediatric specialist were also significant risk factors for treatment incompletion. CONCLUSION: Our study results suggest that promoting the use of 3HR regimen prescriptions could be an effective strategy to enhance treatment completion. Physicians in primary hospitals, hospitals located in non-metropolitan regions, and physicians without a pediatric specialty require increased attention when administering LTBI treatment to pediatric patients to ensure treatment completion.

Out-of-pocket costs associated with chronic respiratory diseases in Korean adults.

OBJECTIVES: Chronic respiratory diseases (CRDs) are a burden on both individuals and society. While previous literature has highlighted the clinical burden and total costs of care, it has not addressed patients' direct payments. This study aimed to estimate the incremental healthcare costs associated with patients with CRDs, specifically out-of-pocket (OOP) costs. METHODS: We used survey data from the 2019 Korea Health Panel Survey to estimate the total OOP costs of CRDs by comparing the annual hospitalizations, outpatient visits, emergency room visits, and medications of patients with and without CRDs. Generalized linear regression models controlled for differences in other characteristics between groups. RESULTS: We identified 222 patients with CRDs, of whom 166 were aged 65 years and older. Compared with the non-CRD group, CRD patients spent more on OOP costs (238.3 USD on average). Incremental costs were driven by outpatient visits and medications, which are subject to a coinsurance of 30% or more and may include items not covered by public insurance. Moreover, CRD patients aged 50-64 years incurred the highest incremental costs. DISCUSSION: The financial burden associated with CRDs is significant, and outpatient visits and medications constitute the largest components of OOP spending. Policymakers should introduce appropriate strategies to reduce CRD-associated burdens.

BHLHE40 drives protective polyfunctional CD4 T cell differentiation in the female reproductive tract against Chlamydia.

The protein basic helix-loop-helix family member e40 (BHLHE40) is a transcription factor recently emerged as a key regulator of host immunity to infections, autoimmune diseases and cancer. In this study, we investigated the role of Bhlhe40 in protective T cell responses to the intracellular bacterium Chlamydia in the female reproductive tract (FRT). Mice deficient in Bhlhe40 exhibited severe defects in their ability to control Chlamydia muridarum shedding from the FRT. The heightened bacterial burdens in Bhlhe40-/- mice correlated with a marked increase in IL-10-producing T regulatory type 1 (Tr1) cells and decreased polyfunctional CD4 T cells co-producing IFN-γ, IL-17A and GM-CSF. Genetic ablation of IL-10 or functional blockade of IL-10R increased CD4 T cell polyfunctionality and partially rescued the defects in bacterial control in Bhlhe40-/- mice. Using single-cell RNA sequencing coupled with TCR profiling, we detected a significant enrichment of stem-like T cell signatures in Bhlhe40-deficient CD4 T cells, whereas WT CD4 T cells were further down on the differentiation trajectory with distinct effector functions beyond IFN-γ production by Th1 cells. Altogether, we identified Bhlhe40 as a key molecular driver of CD4 T cell differentiation and polyfunctional responses in the FRT against Chlamydia.

Bacterial profile of pork from production to retail based on high-throughput sequencing.

Pork is a common vehicle for foodborne pathogens, including Salmonella spp. and Yersinia enterocolitica. Cross-contamination can occur at any stage of the pork production chain, from farm to market. In the present study, high-throughput sequencing was used to characterize bacterial profiles and track their changes along the whole supply chain. Tracked meat samples (pig on the farm, carcass in the slaughterhouse, unprocessed carcass and processed meat in the processing plant, and fresh pork at the local retail stores) and their associated environmental samples (e.g., water, floor, feed, feces, and workers' gloves) were collected from sequential stages (n = 96) and subjected to 16S rRNA metataxonomic analyses. At the farm, a total of 652 genera and 146 exclusive genera were identified in animal and environmental samples (pig, drain, floor, fan, and feces). Based on beta diversity analysis, it was demonstrated that the microbial composition of animal samples collected at the same processing step is similar to that of environmental samples (e.g., drain, fan, feces, feed, floor, gloves, knives, tables, and water). All animal and environmental samples from the slaughterhouse were dominated by Acinetobacter (55.37 %). At the processing plant, belly meat and neck meat samples were dominated by Psychrobacter (55.49 %). At the retail level, key bacterial players, which are potential problematic bacteria and important members with a high relative abundance in the samples, included Acinetobacter (8.13 %), Pseudomonas (6.27 %), and Staphylococcus (2.13 %). In addition, the number of confirmed genera varied by more than twice that identified in the processing plant. Source tracking was performed to identify bacterial contamination routes in pork processing. Animal samples, including the processing plant's carcass, the pig from the farm, and the unwashed carcass from the slaughterhouse (77.45 %), along with the processing plant's gloves (5.71 %), were the primary bacterial sources in the final product. The present study provides in-depth knowledge about the bacterial players and contamination points within the pork production chain. Effective control measures are needed to control pathogens and major pollutants at each stage of pork production to improve food safety.

Effect of MIND diet on cognitive function in elderly: a narrative review with emphasis on bioactive food ingredients.

As the world becomes a super-aged society, cognitive decline is public health problems that are increasing rapidly. A healthy diet has great potential for maintaining cognitive health. A diet that could delay the onset of neurodegenerative diseases has been developed: the Mediterranean-DASH Intervention for Neurodegenerative Delay (MIND) diet, a hybrid form of the Mediterranean diet and the Dietary Approaches to Stop Hypertension (DASH) diet. In this review, the effects of the MIND diet on improving cognitive function, including memory, are summarized. In most studies, the higher the adherence to the MIND diet, the higher the cognitive function evaluation score, and the lower the incidence of dementia. This is because of the anti-inflammatory and antioxidant effects of the major nutritional components of the MIND diet: folate, carotenoids, polyphenols, and polyunsaturated fatty acids. Adherence to the MIND diet, containing various bioactive food ingredients, is related to cognitive improvement in the elderly population.

The ergodicity question when imaging DNA conformation using liquid cell electron microscopy.

Assessing the ergodicity of graphene liquid cell electron microscope measurements, we report that loop states of circular DNA interconvert reversibly and that loop numbers follow the Boltzmann distribution expected for this molecule in bulk solution, provided that the electron dose is low (80-keV electron energy and electron dose rate 1-20 e- Å-2 s-1). This imaging technique appears to act as a "slow motion" camera that reveals equilibrated distributions by imaging the time average of a few molecules without the need to image a spatial ensemble.

Nanoscale Cathodoluminescence Thermometry with a Lanthanide-Doped Heavy-Metal Oxide in Transmission Electron Microscopy.

When navigated by the available energy of a system, often provided in the form of heat, physical processes or chemical reactions fleet on a free-energy landscape, thus changing the structure. In in situ transmission electron microscopy (TEM), where material structures are measured and manipulated inside the microscope while being subjected to external stimuli such as electrical fields, laser irradiation, or mechanical stress, it is necessary to precisely determine the local temperature of the specimen to provide a comprehensive understanding of material behavior and to establish the relationship among energy, structure, and properties at the nanoscale. Here, we propose using cathodoluminescence (CL) spectroscopy in TEM for in situ measurement of the local temperature. Gadolinium oxide particles doped with emissive europium ions present an opportunity to utilize them as a temperature probe in CL measurements via a ratiometric approach. We show the thermometric performance of the probe and demonstrate a precision of ±5 K in the temperature range from 113 to 323 K with the spatial resolution limited by the size of the particles, which surpasses other methods for temperature determination. With the CL-based thermometry, we further demonstrate measuring local temperature under laser irradiation.

Changes in the BOLD signal of S1 and BA3 per finger/phalanx as a response to high-frequency vibratory stimulation.

PURPOSE AND METHOD: The purpose of this study was to determine the changes in the Blood Oxygen Level Dependent signal of Primary somatosensory area (S1) and Brodmann area 3 (BA3) per finger and phalanx in comparison to the activation voxel when 250 Hz vibratory stimulation with high sensitivity for the Pacinian corpuscle was given to the four fingers and three phalanges. RESULTS: The result of analyzing the activation voxel showed a significant difference for S1 per finger and phalanx, but for BA3, no significant difference was observed despite a similar trend to S1. In contrast, the activation intensity (BOLD) displayed a significant difference for S1 per finger and phalanx and for BA3, where the activation voxel had no significant variation. In addition, while the result of S1 did not indicate whether the index or the little fingers had the highest sensitivity based on the BOLD signal per finger, the result of BA3 marked the strongest BOLD signal for the little finger as a response to 250 Hz vibratory stimulation. The activation intensity per phalanx was the highest for the intermediate phalanx for S1 and BA3, which was in line with a previous study comparing the activation voxel. CONCLUSIONS: The method based on the intensity of the nerve activation is presumed to have high sensitivity as the signal intensity is monitored within a specific, defined area. Thus, for the extraction of brain activation patterns of micro-domains, such as BA3, monitoring the BOLD signal that reflects the nerve activation intensity more sensitively is likely to be advantageous.

Thickness Tolerance in Large-Area Organic Photovoltaics with Fluorine-Substituted Regioregular Conjugated Polymer.

Large areas and simple processing methods are necessary for the commercialization of organic photovoltaics (OPVs). However, the efficiency drop due to the variation in thickness of OPVs limits their large-scale applications. Regioregular polymers with good crystallinity and packing properties that exhibit high charge mobility and extraction ability can help overcome these limitations. In this study, a regioregular polymer named PDBD-2FBT was synthesized. The crystallinity and packing properties of PDBD-2FBT were enhanced by a simple thermal treatment. Using PDBD-2FBT material as a donor and Y6-HU as an acceptor, we fabricated binary blend OPV devices. The devices with optimized active layer thickness achieved a power conversion efficiency (PCE) of 14.14%. A PCE of 13.18% was maintained even in thick-film conditions (400 nm), and thickness tolerance was observed. Based on the thickness tolerance, a 5-line module measuring 36 cm2 was fabricated via the bar-coating method, and a PCE of approximately 10% was achieved.

Association between body shape misperception and unhealthy eating behaviors among Korean adolescents.

BACKGROUND/OBJECTIVES: Body shape misperception (BSM) is the disagreement between the subjectively perceived body size and body mass index. This study investigated the association between BSM and unhealthy eating behaviors (UEB) among Korean adolescents. SUBJECTS/METHODS: This cross-sectional study used data from the 2019 Korea Youth Risk Behavior Web-based Survey with 55,748 participants. UEB were measured according to the weekly consumption frequency of caffeinated energy drinks, fast food, carbonated beverages, and sugar-sweetened beverages. The covariates included demographic and socioeconomic characteristics and health-related factors. Multiple logistic regression analysis determined the association between BSM and UEB. RESULTS: Among the participants, 37,607 (67.5%) reported UEB. The gap between UEB among those with BSM was the largest between the underestimated and accurately estimated groups. Participants who underestimated their body shape were likelier to engage in UEB (adjusted odds ratio [AOR], 1.18; 95% confidence interval [CI], 1.11-1.25). Both sexes with underestimation of body size showed an association with UEB compared to those with accurate estimations (girls: AOR, 1.19; 95% CI, 1.09-1.30; boys: AOR, 1.16; 95% CI, 1.08-1.26). CONCLUSIONS: Underestimating body shape can provoke UEB among Korean adolescents. The need for appropriate school health interventional programs to prevent underestimating body shape is emphasized to avoid UEB.

Time-Frequency Multi-Domain 1D Convolutional Neural Network with Channel-Spatial Attention for Noise-Robust Bearing Fault Diagnosis.

This paper proposes a noise-robust and accurate bearing fault diagnosis model based on time-frequency multi-domain 1D convolutional neural networks (CNNs) with attention modules. The proposed model, referred to as the TF-MDA model, is designed for an accurate bearing fault classification model based on vibration sensor signals that can be implemented at industry sites under a high-noise environment. Previous 1D CNN-based bearing diagnosis models are mostly based on either time domain vibration signals or frequency domain spectral signals. In contrast, our model has parallel 1D CNN modules that simultaneously extract features from both the time and frequency domains. These multi-domain features are then fused to capture comprehensive information on bearing fault signals. Additionally, physics-informed preprocessings are incorporated into the frequency-spectral signals to further improve the classification accuracy. Furthermore, a channel and spatial attention module is added to effectively enhance the noise-robustness by focusing more on the fault characteristic features. Experiments were conducted using public bearing datasets, and the results indicated that the proposed model outperformed similar diagnosis models on a range of noise levels ranging from -6 to 6 dB signal-to-noise ratio (SNR).

Visualization with Prediction Scheme for Early DDoS Detection in Ethereum.

Blockchain technologies have gained widespread use in security-sensitive applications due to their robust data protection. However, as blockchains are increasingly integrated into critical data management systems, they have become attractive targets for attackers. Among the various attacks on blockchain systems, distributed denial of service (DDoS) attacks are one of the most significant and potentially devastating. These attacks render the systems incapable of processing transactions, causing the blockchain to come to a halt. To address the challenge of detecting DDoS attacks on blockchains, existing visualization schemes have been developed. However, these schemes often fail to provide early DDoS detection since they typically display only past and current system status. In this paper, we present a novel visualization scheme that not only portrays past and current values but also forecasts future expected system statuses. We achieve these future predictions by utilizing polynomial regression with blockchain data. Additionally, we offer an alternative DDoS detection method employing statistical analysis, specifically the coefficient of determination, to enhance accuracy. Through our experiments, we demonstrate that our proposed scheme excels at predicting future blockchain statuses and anticipating DDoS attacks with minimal error. Our work empowers system managers of blockchain-based applications to identify and mitigate DDoS attacks at an earlier stage.

Saponins Derived from the Korean Endemic Plant Heloniopsis koreana Inhibit Diffuse-type Gastric Cancer Cells.

Diffuse-type gastric cancer (GC) is a type of stomach cancer that occurs in small clusters of cells that are widely spread. It does not typically manifest with symptoms until the advanced stages and often goes undetected in routine imaging tests. In addition, there is no specific targeted therapy for diffuse-type GC and it has a high mortality risk. Hence, it is worthwhile to discover molecules against this cancer. In this study, the extract of Heloniopsis koreana, which is endemic to Korea, exhibited cytotoxicity against two diffuse-type GC cell lines, MKN1 and SNU668. This led to the isolation of 10 compounds, including a new cinnamic acid glycoside. Of the compounds, saponin Th (4) and SNF 11 (5) showed potent activities with IC50 values of 3.66 and 3.85 µM, respectively, in MKN1 cells, and 1.8 and 1.98 µM, respectively, in SNU668 cells. These compounds prevented cancer cell division, invasion, and colony formation in both cell lines. In addition, these compounds induced cancer cell death through conventional cell death pathways, showing an increase in ADP-ribose polymerase, caspase 3, and BAX and a decrease in BCL2.