Share your research data

This double-blind, randomized, placebo-controlled pilot trial evaluated the efficacy of individualized homeopathic medicines (IHMs) in women with heavy menstrual bleeding (HMB). While IHMs did not demonstrate a statistically significant reduction in menstrual blood loss (PBAC) in the primary analysis, consistent trends toward improvement were observed, particularly in per-protocol and sensitivity analyses. Notably, IHMs produced a significant and robust improvement in quality of life (MMAS) compared to placebo. The intervention was safe, with no reported adverse events. These findings suggest that individualized homeopathy may serve as a beneficial, non-invasive adjunct in HMB management, particularly in enhancing patient-centered outcomes, though larger trials with objective measures are needed to confirm efficacy.

This dataset accompanies the study "Optimizing Real-Time Phenotyping in Critical Care Using Machine Learning on Electronic Health Records," (DOI: 10.1016/j.eswa.2026.132084) which hypothesizes that a patient's latent disease state can be continuously and accurately estimated from real-time biomedical signals without requiring full ICU trajectories. It supports replication and evaluation of our predictive framework, which dynamically models phenotype probabilities as data accumulates. All elements are reported in line with the TRIPOD statement to ensure transparency and reproducibility. The training and test data are derived from the MIMIC-IV database and consist of vectorized representations of multivariate, irregularly sampled biomedical time series and associated phenotype labels. These were generated through a structured pipeline that includes cohort selection, event aggregation using fixed-length time bins, and feature engineering to represent both value trends and missingness. Supplementary Tables S.1 to S.6 describe the variables used in this transformation, their sources within the EHR, aggregation methods, and descriptive statistics for both static (e.g., demographics, admission data) and dynamic (e.g., vital signs, lab results, ventilator settings) features across the train and test sets. Table S.7 summarizes the model’s real-time phenotyping performance using multiple evaluation perspectives. The results reveal strong generalization and early predictive value: in the (ls) setting, the model achieved good diagnostic performance (AUROC ≥ 0.8) for 69% of phenotypes and excellent performance (AUROC ≥ 0.9) for 30%. In the real-time (fs) setting—using only the earliest recorded physiological data—the model still achieved good performance for 42% of phenotypes and excellent performance for 7%, demonstrating the feasibility of early, actionable phenotyping. The intermediate (td) evaluation shows that predictive quality improves consistently as more data becomes available, supporting the framework’s ability to track dynamic disease progression in real time. To interpret and use the data: - Each patient stay is represented as a multivariate time series with associated phenotype labels. - Time series are aligned in fixed time intervals (e.g., 2 hours), where each variable is aggregated using statistical functions (e.g., mean, last, sum). - The phenotype labels correspond to ICD-9-CM diagnostic categories assigned at discharge but are used here as latent variables to be estimated continuously. This dataset enables reproducibility of the results and further research in developing machine learning models for early, interpretable, and actionable phenotyping in critical care.

This study was designed as a cross-sectional methodological analysis to evaluate the accuracy, quality, and readability of responses generated by large language models (ChatGPT and Gemini) to frequently asked questions about vertigo. A total of 50 questions were initially generated from three sources: ChatGPT, Gemini, and Google’s “People also ask” section. After removing duplicates and irrelevant items, 20 representative questions were selected. Each question was entered into both models, and only the first response was recorded without any follow-up prompts. The responses were evaluated by five blinded experts (two otolaryngologists, two audiologists, and one physiotherapist). Medical accuracy was assessed using a 4-point Likert scale, and information quality was evaluated using the DISCERN instrument. Mean scores across experts were used for analysis. Readability was assessed using multiple standard indices, including Flesch Reading Ease, Flesch–Kincaid Grade Level, Gunning Fog Index, SMOG, Coleman–Liau Index, and the Automated Readability Index. Additional textual features such as word count, sentence number, average sentence length, and percentage of complex words were also analyzed to determine linguistic complexity.

This dataset contains .inp files for the TOPAS V7 that were used for double-Voigt diffraction data analysis utilizing the Stephens and Aniso macro to analyze contributions to peak-broadening and the size of diffracting domains.

This dataset contains a sample of the raw data (excel files) obtained from a mechanical 360-degree scanning sonar (Ping360 Scanning Image Sonars, Blue Robotics Inc., California, USA) deployed in a commercial sea-cage (140 m circumference, 27 m depth, the upper 15 m cylindrical + 12 m conical; a lice skirt was mounted to 5 m depth) rearing Atlantic salmon (Salmo salar) in Norway. Fish were just transferred from the land-based facility to the sea cage, and we wanted to follow the evolution in their behavior (vertical and horizontal distribution) their first month at sea. Fish were on average 510g (post smolts) and density in the cage at the start was 4.3 kg/m3. For more details on the experimental design we refer readers to the article link provided below. Due to storage limits, 3 full days (24 h each) are included (one folder per date). Raw data from 38 days (not always 24h continuous) is available on request but all scans from the raw data are presented as videos (10 images per sec) in the folder "All videos from complete dataset". Description xlsx files: Timestamps are in CET and embedded in filenames (yyyymmdd_hhmmss-x). First column: angle (radians, 0–399) First row: distance (0–1199, equal to 30 m) Remaining cells: echo strength (Analog-to-Digital Converter values, 0–597) Folder "Data processing": Includes a Python script to plot sonar data, input/output examples, and a labeled scan ("Scanwithexplanations.png") in polar coordinates. The sonar is in the middle of the figure. The circular lines show the distance from the sonar and each line corresponds to a 5 m distance. Apart from the visible cage structures (cage walls, cone leading to the cage bottom and water surface), observations inside the cage are salmon. Any visible echo strength above the water surface corresponds to echo reverberation on the surface. Cage deformation due to current is also visible on the scans and examples of deformed and undeformed cage are provided in the folder "Cage deformation". Folder "Water current": Excel file with two sheets: Horizonspeed = overall sideways movement of the water in cm/sec Eastspeed = current in the eastern direction in cm/sec First column: timestamp (dd.mm.yyyy HH:MM in UTC) First range: depth (in meters) Data collected on one of the most exposed buoys of the fish farm's mooring frame.

This dataset contains the research data supporting the manuscript entitled “High-voltage electrostatic field thawing stabilizes water immobilization and preserves ventral-skin golden appearance in frozen large yellow croaker (Larimichthys crocea)”. The deposited files include: (1) Supplementary Table S1, containing the list of 38 identified volatile compounds and their normalized semi-quantitative responses across thawing treatments; (2) Supplementary Table S2, containing the complete VIP ranking from the OPLS-DA model based on the normalized responses of the 38 identified volatile compounds; and (3) an additional raw sensory evaluation dataset containing anonymized panelist-level scores and fish-level mean values used for statistical analysis. Treatment abbreviations are consistent with those used in the manuscript (RT, WB, UAT, MW, and HVEF).

This dataset contains the input and output files from the three-dimensional morphodynamic modeling of reef islands using the non-hydrostatic XBeach model (XBeach-NH), as presented in the associated manuscript "Three-dimensional morphodynamic modeling of reef islands with varying geomorphic and sedimentary attributes". The data is organized into the following directories, each representing a set of sensitivity analyses or the baseline scenario: ReferenceCase/: Files for the baseline or standard scenario simulation. Island height/: Simulations varying the initial island crest elevation. Beach slope/: Simulations varying the beach slope. Island shape/: Simulations comparing different island shape. Sediment grain size/: Simulations varying the sediment grain size. Wave process data/: Hydrodynamic outputs for both fixed and mobile bed conditions. Initial island/: Topographic and initial condition files. Each directory typically contains the model parameter files (params.txt), initial topography grids, and key output files.

The dataset contains the Cross-Correlation Functions (CCF) and velocity change (dv/v) from different methods of Haiyuan Fault Zone . Due to file size limitations, just the Z-component cross-correlations are shown here. We also share local temperature data from Zhongning meteorological station during study period (ID 53705, China).

This dataset contains raw behavioral data from 90 experimental trials testing host protection and predator-avoidance in five palaemonid shrimp species across three treatments (bare tank, synthetic anemone, live anemone Entacmaea quadricolor). Data includes: - Shrimp survival time (Table S1) - Frequency of hunting attempts by fish Epinephelus lanceolatus (Table S2) - Frequency of tail-flipping in shrimps (Table S6) - Proportion of time spent on continuous behaviors of shrimps and fish (Table S3 and S5) - Proportion of time spent on different spatial regions of shrimps (Table S4) Data supports the manuscript: [Tse, T. W., Wong, T. H. H. & Tsang, L. M. (in press). Protective benefit of sea anemone host to symbiotic palaemonid shrimps and predator-avoidance behaviours among palaemonid shrimps. Zoology].

This dataset consists of heart rate estimates over time from 33 individuals of marine invertebrates (belonging to nine species from three taxonomic groups - gastropods, bivalves and crustaceans) derived either from manual counts or from the algorithm of the associated manuscript. Heart rate estimates from these two methods were compared to generate various error estimates that support the algorithm accuracy and high data utilization. Detailed methods for obtaining heart rate measurements for each individual are provided in the supplementary information (Supplementary S3) of the associated manuscript.