Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
251
datasets available to search
ShareScore release 0.7.1
Dataset results
251 results for “deep learning models”
High resolution downscaling of CMIP6 Earth System and Global Climate Models using deep learning for Iberia Part 10
<p>Extra data of the GMD paper "High resolution downscaling of CMIP6 Earth System and Global Climate Models using deep learning for Iberia" that did not fit in their respective deposits:</p> <p>Future projections of:</p> <p>Precipitation by all CNN models (BMlinear, BM1, BM10, BMdense) forced by the UKESM1-0-LL GCM.</p> <p>2-meter maximum and minimum temperatures by the BM1 model forced by the NorESM2-MM and UKESM1-0-LL GCMs.</p> <p>2-meter mean temperature by the BMlinear and BM1 models forced by the NorESM2-MM and UKESM1-0-LL GCMs.</p>
High resolution downscaling of CMIP6 Earth System and Global Climate Models using deep learning for Iberia Part 4
<p>Future projections of 2-meter mean temperature by the CNN models (BM1, BM10 and BMdense) forced by the seven GCMs used in the GMD paper "High resolution downscaling of CMIP6 Earth System and Global Climate Models using deep learning for Iberia".</p>
ViTAL deep learning models for lineage assignment under low coverage
<p>ViTAL, is a lineage assignment algorithm which inputs<br> a low-coverage genome, transforms it into embedded genome<br> fragments which are then fed into a classification neural<br> network, that outputs the most likely lineages the input genome<br> might belong to. The ViTAL algorithm is therefore divided into<br> preprocessing phase (MinHash) followed by embedding, and the<br> classification phase (Vision Transformer).</p> <p> </p> <p>The upload contains trained models for lineage assignment.</p>
Pretrained model for tutorial from: "PartSeg v2: Bioimage segmentation using advanced Deep Learning techniques"
Open the record for dataset details and reuse information.
Deep Learning Model to Predict the Recurrence of Stage IA Invasive Lung Adenocarcinoma After Sub-lobar Resection
ClinicalTrials.gov study NCT06659601. IPD Sharing: NO. Countries: 1. Publications: 0.
Multicentric Study for External Validation of a Deep Learning Model for Mammographic Breast Density Categorization
ClinicalTrials.gov study NCT05021055. IPD Sharing: NO. Countries: 0. Publications: 22.
The CT-based Deep Learning Model Predicts Complications in Partial Nephrectomy
ClinicalTrials.gov study NCT06876584. IPD Sharing: YES. Countries: 1. Publications: 0.
Building of Prognosis Model for Patients With Cirrhosis Based on Sarcopenia Assessed by Deep Learning
ClinicalTrials.gov study NCT06531200. IPD Sharing: Not stated. Countries: 0. Publications: 29.
Deep Learning Model Predicts Pathological Complete Response of Lung Cancer Following Neoadjuvant Immunochemotherapy
ClinicalTrials.gov study NCT06285058. IPD Sharing: UNDECIDED. Countries: 0. Publications: 1.
Evaluation and optimization of sequence-based gene regulatory deep learning models
GEO Series GSE254493. Saccharomyces cerevisiae. 36 samples. Type: Other.
HydRA: Deep-learning models for predicting RNA-binding capacity from protein interaction association context and protein sequence
GEO Series GSE221870. Homo sapiens. 76 samples. Type: Other.
Phenotypic Screening with Deep Learning Identifies HDAC6 Inhibitors as Cardioprotective Agents in Mouse Models of Dilated Cardiomyopathy
GEO Series GSE179656. Mus musculus. 25 samples. Type: Expression profiling by high throughput sequencing.
Improved Prediction of Smoking Status via Isoform-Aware RNA-seq Deep Learning Models
GEO Series GSE158699. Homo sapiens. 2653 samples. Type: Expression profiling by high throughput sequencing.
Deep Learning-Based Multimodal Clustering Model for Endotyping and Post-Arthroplasty Response Classification using Knee Osteoarthritis Subject-Matched Multi-Omic Data
GEO Series GSE222979. Homo sapiens. 1242 samples. Type: Non-coding RNA profiling by high throughput sequencing.
Optimizing Bank Loan Approval with Cutting-Edge Deep Learning model
<p>Abstract</p><p>For any bank or financial institution, managing loans and controlling leverage is one of the most<br>important tasks they have to undertake. A bank cannot function efficiently without a well-<br>designed loan-to-deposit business model. As technology continues to evolve, the mechanism of<br>handling and granting loans underwent a significant change with the introduction of use cases<br>concerning machine learning and data science.<br>Hence, this data-driven research utilized advanced machine learning techniques to analyze and<br>manipulate the data, aiming to predict the best possible way to recommend a loan to a client.<br>These predictions are based on modified yet unique features created from the data obtained from<br>the client. The dataset was tested using two different methodologies: a logistic regression model<br>and a Neural Network algorithm. Both of these methodologies produced high-level accuracy<br>rates. However, the latter outperformed the currently used methodologies by over 20%, resulting<br>in an accuracy of 90%.<br>The successful research results were obtained due to the use of a perfectly balanced, unbiased,<br>and cleaned dataset, as well as the well-executed combination of activation functions for the<br>Neural Network model. A performance assessment was conducted based on a confusion matrix<br>evaluation to demonstrate its feasibility and performance </p>
Input data for deep learning model-analog
<p>This repository contains input data required to run the Deep Learning Model-Analog (<a href="https://github.com/kinyatoride/DLMA">GitHub</a>), as presented in the paper titled "Using Deep Learning to Identify Initial Error Sensitivity for Interpretable ENSO Forecasts" by Toride et al. A preprint is available at <a href="https://arxiv.org/abs/2404.15419">https://arxiv.org/abs/2404.15419</a>.</p> <p>The <code>cesm2</code> directory contains the Community Earth System Model Version 2 Large Ensemble (<a href="https://doi.org/10.26024/kgmp-c556" rel="nofollow">CESM2-LE</a>), while the <code>real</code> directory contains the Ocean Reanalysis System 5 (<a href="https://doi.org/10.24381/cds.67e8eeb7" rel="nofollow">ORAS5</a>) datasets. These datasets have been processed to provide detrended monthly anomalies and have been interpolated to two different resolutions: 2° × 2° and 5° × 5°. The 5°×5° files are used as input, while the 2°×2° files are used for analog forecasting.</p>
High Spatiotemporal Resolution Estimation of Global Surface CO Concentrations Using a Deep Learning Model
<p><span>A high-performance Convolutional Neural Network (CNN)-based Residual Network (ResNet) was developed for estimating daily worldwide CO concentrations at a high spatial resolution of 0.07</span><span>°</span><span> from June 2018 to May 2021, using the global TROPOMI Total Column of atmospheric CO (TCCO) product and reanalysis datasets. The proposed framework achieved a desirable estimation accuracy, with <em>R</em>-values (correlation coefficients) of 0.90 and 0.96 for daily and monthly predictions, respectively. The daily surface CO concentration dataset from our study is potentially useful for further relevant sustainable studies.</span></p>
Deep learning sequence-based ab initio prediction of variant effects on expression and disease risk (sequence model release)
<p>(This is the updated version that has been converted a standard pytorch model format)</p> <p>This is the deep learning sequence model used in </p> <p>Jian Zhou, Chandra L. Theesfeld, Kevin Yao, Kathleen M. Chen, Aaron K. Wong, and Olga G. Troyanskaya, Deep learning sequence-based ab initio prediction of variant effects on expression and disease risk, Nature Genetics, 2018.</p> <p>Note the full software is available from https://github.com/FunctionLab/ExPecto and this release is created for the convenience of use and under the same non-commercial license. The model weights can be loaded with pytorch load_state_dict function (for an example please find <a href="https://github.com/FunctionLab/ExPecto/blob/master/chromatin.py">https://github.com/FunctionLab/ExPecto/blob/master/chromatin.py</a>). We also provide a web server for browsing mutations with strong predicted effects at https://hb.flatironinstitute.org/expecto/, which are currently limited to mutations within 1kb to TSS or are 1000 Genomes variants.</p> <p>Trivia: we code-named our models with whale names. This model has an unofficial codename DeepSEA "Beluga".</p>
Large-Population based Deep Learning Models in Classifying Primary Bone Tumors and Bone Infections based on Radiographs: a Retrospective and Multi-reader Multi-center Study
<p>This retrospective multicenter study collected patients via consecutive sampling between 2013 and 2022 from two cohorts: training cohort (from the Second Xiangya Hospital of Central South University) and testing cohort (from Xiangya Hospital of Central South University and Hunan Children's Hospital of Central South University). These lesions were identified to have bone involvement through pre-operative radiographs and were histologically diagnosed following biopsy or surgery. <br>(i) For the inclusion criteria, lesions were confirmed and diagnosed as PBTs according to the 2020 World Health Organization (WHO) system for the classification for tumors of bone 1 while bone infections were confirmed and proven by histology and (or) bacterial culture. The other vital inclusion criteria are evident as well as available clinical information and pre-operative radiographs. <br>(ii) The screening criteria: (a) radiographs were from patients diagnosed between 2013 and 2022 (b) in selected three hospitals; (c) radiographs with robust quality for reliable assessments of the bone lesions and (d) all of these radiographs were pre-operative. <br>Clinical characteristics like age, gender, and the location of the lesion of interest and so on were obtained from the patients' electronic medical records after data desensitization and standardization.<br>Radiographs were kept and downloaded as Digital Imaging and Communications in Medicine (DICOM) files from the picture archiving and communication system (PACS) at their original sizes and resolutions. All of these radiograph images have undergone desensitization processing of disengaging patient protected health information from DICOM data to meet the relevant legal criteria and requirements of US (HIPAA) as well as European (GDPR). Delineating the region of interest (ROI) was performed by two proficient radiologists. ROIs were meticulously outlined via Click 2 Crop (version 5.2.2) (https://click-2-crop.en.softonic.com/) to closely segment pertinent entities present in each PBT or bone infection. The smallest rectangular box that can completely cover the ROI was manually annotated as the boundary box by senior seniority radiologist to ensure accuracy. Afterwards, the annotated ROIs were used as ground truth for the model development process.</p>
Dataset of paper: Deep Learning Models for Automated Identification of Scheduling Policies
<p>The training set and testing set for the paper: Deep Learning Models for Automated Identification of Scheduling Policies</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.