Context

This dataset contains the tfrecord converted files of the original NIH Chest X-ray dataset. It was created for working with TPU.

Find the original dataset at https://www.kaggle.com/nih-chest-xrays/data

Content

What's inside is more than just rows and columns. Make it easy for others to get started by describing how you acquired the data and what time period it represents, too.

Acknowledgements

We wouldn't be here without the help of others. If you owe any attributions or thanks, include them here along with any citations of past research.

Inspiration

Your data will be in front of the world's largest data science community. What questions do you want to see answered?

Context

Part of NiH dataset where Bounding Boxes are available in format of VinBigData Chest X-ray Abnormalities Detection competition.

NIH Chest X-ray14 Dataset

  Dataset Description

This dataset contains 112120 chest X-ray images with multiple disease labels per image.

  Labels

Atelectasis, Cardiomegaly, Consolidation, Edema, Effusion, Emphysema, Fibrosis, Hernia, Infiltration, Mass, No Finding, Nodule, Pleural_Thickening, Pneumonia, Pneumothorax

  Dataset Structure

Train split: 78484 images Validation split: 16818 images Test split: 16818 images

  Data Format

This dataset is… See the full description on the dataset page: https://huggingface.co/datasets/Manas2703/chest-xray-14.

ChestX-ray8 is a medical imaging dataset which comprises 108,948 frontal-view X-ray images of 32,717 (collected from the year of 1992 to 2015) unique patients with the text-mined eight common disease labels, mined from the text radiological reports via NLP techniques.

Dataset

This dataset was created by Hapon Maksym

Contents

The dataset is genrated by the fusion of three publicly available datasets: COVID-19 cxr image (https://github.com/ieee8023/covid-chestxray-dataset), Radiological Society of North America (RSNA) (https://www.kaggle.com/c/rsna-pneumonia-detection-challenge), and U.S. national library of medicine (USNLM) collected Montgomery country - NLM(MC) (https://lhncbc.nlm.nih.gov/publication/pub9931). These datasets were annotated by expert radiologists. The fused dataset consists of samples of diseases labeled as COVID-19, Tuberculosis, Other pneumonia (SARS, MERS, etc.), and Normal. The dataset can be utilized to train and evaulate deep learning and machine learning models as binary and multi-class classification problem.

Dataset

This dataset was created by Luigi Saetta

Released under CC0: Public Domain

Contents

It contains the following files:

Manas2703/Chest-X-ray-imaging-nih dataset hosted on Hugging Face and contributed by the HF Datasets community

NIH-CXR14-BiomedCLIP-Features Dataset

This dataset is derived from the NIH Chest X-ray Dataset (NIH-CXR14) and processed using the BiomedCLIP-PubMedBERT_256-vit_base_patch16_224 model from Microsoft. It contains image and text features extracted from chest X-ray images and their corresponding textual findings.

  Dataset Description

The original NIH-CXR14 dataset comprises 112,120 chest X-ray images with disease labels from 30,805 unique patients. This processed dataset… See the full description on the dataset page: https://huggingface.co/datasets/Yasintuncer/NIH-CXR14-BiomedCLIP-Features.

Dataset

This dataset was created by Elahi

Contents

Overview

• Introduction
• Object Classes
  • Atelectasis
  • Cardiomegaly
  • Effusion
  • Infiltrate
  • Mass
  • Nodule
  • Pneumonia
  • Pneumothorax

Introduction

This dataset contains chest X-ray images annotated for detecting various thoracic diseases. The dataset addresses the task of identifying specific abnormalities within the lungs.

• Atelectasis: Partial or complete collapse of a lung or lobe.
• Cardiomegaly: Enlargement of the heart.
• Effusion: Accumulation of fluid in the pleural cavity.
• Infiltrate: Presence of denser substances in the lungs, often indicative of inflammation.
• Mass: A larger area indicating potential malignancy.
• Nodule: A small, round or oval-shaped spot in the lung.
• Pneumonia: Infection that inflames the air sacs.
• Pneumothorax: Collapsed lung due to air in the chest.

Object Classes

Atelectasis

Description

Atelectasis appears as a blurring or loss of clarity in the lung area, often near the lung base.

Instructions

Identify areas where clarity in the lung structure diminishes and boundaries become blurred, particularly towards the lung bases. Avoid areas with distinct lung markings or other pathologies.

Cardiomegaly

Description

Cardiomegaly is indicated by a broader cardiac silhouette exceeding the normal lung-heart outline ratios.

Instructions

Draw a bounding box that encompasses the heart if its silhouette extends into the lung areas noticeably more than usual. Do not include areas outside the cardiac outline.

Effusion

Description

Effusion presents as a dimmed, flat shadow at the lung base, indicating fluid accumulation.

Instructions

Look for uniform, smooth shadows at the lung bases. Delineate clearly where fluid outlines are smooth and horizontal, avoiding irregular opacities.

Infiltrate

Description

Infiltrates show as streaky, cloudy regions scattered within the lung field.

Instructions

Mark areas where streakiness or clouding interrupts regular lung patterns, being cautious not to confuse with nodules or masses.

Mass

Description

A mass is a large, localized opacity with clear boundaries, potentially overlying lung structures.

Instructions

Outline large, distinct opacities with well-defined edges. Ensure not to include small, round opacities that fit the nodule description.

Nodule

Description

Nodules are small, round opacities that stand out against the lung background.

Instructions

Identify small, well-circumscribed round spots, ensuring they are distinct from larger masses or infiltrate-like entities.

Pneumonia

Description

Pneumonia areas have patchy or uniform clouding, often distributed across specific lobes.

Instructions

Highlight areas with patchy, uniform clouding, generally across lobes, without distinct solid borders like masses.

Pneumothorax

Description

Pneumothorax is shown by visibility of the visceral pleural line with no vessel markings beyond.

Instructions

Mark regions where a clear pleural line can be seen without any vascular markings beyond it towards the chest wall. Avoid mislabeling lung collapse regions not conforming to air presence.

The analysis of chest radiography imaging is of paramount importance for healthcare institutions since it is one of the most used imaging modalities for patient diagnosis, management and follow-up. In recent years, several studies have demonstrated that automatic algorithms, namely using deep learning, can successfully detect and segment the anatomical structures seen in chest radiography images, namely the lungs, heart and clavicles. However, only a limited number of public data is available for training, particularly for the heart and clavicles. This dataset provides segmentation masks for the heart and clavicles in the Montgomery dataset, a public repository of 138 chest radiographies for which lung segmentation masks are publicly available (Jaeger et al. Two public chest x-ray datasets for computer-aided screening of pulmonary diseases. Quant. Imaging Med. Surg., 4(6):475–477, December 2014.). Segmentation masks were obtained through manual annotation by a PhD candidate in biomedical Image analysis with experience in chest radiography imaging using the Heartex Label Studio software.

For additional information please refer to our manuscript: R. Brioso et al "Semi-supervised Multi-structure Segmentation in Chest X-Ray Imaging" CBMS 2023

Dataset

This dataset was created by kambarakun

Contents

NIH-CXR14-ELIXR Dataset

  Overview

The NIH-CXR14-ELIXR dataset contains embeddings generated by the ELIXR model using the NIH CXR14 (Chest X-Ray) dataset. These embeddings can be used for various computer vision and medical imaging tasks, including image analysis, retrieval, and multimodal learning. Project GitHub Repository

  Repository Structure

├── dataset_infor.ipynb # Notebook with dataset information ├── datasets.json # Metadata for the… See the full description on the dataset page: https://huggingface.co/datasets/Yasintuncer/nih-cxr14-elixr.

Background: The aggressive and heterogeneous nature of lung cancer has thwarted efforts to reduce mortality from this cancer through the use of screening. The advent of low-dose helical computed tomography (CT) altered the landscape of lung-cancer screening, with studies indicating that low-dose CT detects many tumors at early stages. The National Lung Screening Trial (NLST) was conducted to determine whether screening with low-dose CT could reduce mortality from lung cancer.

Methods: From August 2002 through April 2004, we enrolled 53,454 persons at high risk for lung cancer at 33 U.S. medical centers. Participants were randomly assigned to undergo three annual screenings with either low-dose CT (26,722 participants) or single-view posteroanterior chest radiography (26,732). Data were collected on cases of lung cancer and deaths from lung cancer that occurred through December 31, 2009. This dataset includes the low-dose CT scans from 26,254 of these subjects, as well as digitized histopathology images from 451 subjects.

Results: The rate of adherence to screening was more than 90%. The rate of positive screening tests was 24.2% with low-dose CT and 6.9% with radiography over all three rounds. A total of 96.4% of the positive screening results in the low-dose CT group and 94.5% in the radiography group were false positive results. The incidence of lung cancer was 645 cases per 100,000 person-years (1060 cancers) in the low-dose CT group, as compared with 572 cases per 100,000 person-years (941 cancers) in the radiography group (rate ratio, 1.13; 95% confidence interval [CI], 1.03 to 1.23). There were 247 deaths from lung cancer per 100,000 person-years in the low-dose CT group and 309 deaths per 100,000 person-years in the radiography group, representing a relative reduction in mortality from lung cancer with low-dose CT screening of 20.0% (95% CI, 6.8 to 26.7; P=0.004). The rate of death from any cause was reduced in the low-dose CT group, as compared with the radiography group, by 6.7% (95% CI, 1.2 to 13.6; P=0.02).

Conclusions: Screening with the use of low-dose CT reduces mortality from lung cancer. (Funded by the National Cancer Institute; National Lung Screening Trial ClinicalTrials.gov number, NCT00047385).

Data Availability: A summary of the National Lung Screening Trial and its available datasets are provided on the Cancer Data Access System (CDAS). CDAS is maintained by Information Management System (IMS), contracted by the National Cancer Institute (NCI) as keepers and statistical analyzers of the NLST trial data. The full clinical data set from NLST is available through CDAS. Users of TCIA can download without restriction a publicly distributable subset of that clinical data, along with the CT and Histopathology images collected during the trial. (These previously were restricted.)

Hyperparameters explored during model development and training.

Top 16 models classifying large and moderate pneumothorax, excluding small pneumothoraces in training.

Dataset

This dataset was created by Mikhail

Contents

Pneumothorax is a common medical emergency defined as the abnormal collection of air in the pleural space between the lung and chest wall. Its typical symptoms include chest pain and dyspnea, leading to oxygen deficiency or even life-threatening in severe cases. Therefore, an efficient and automatic pneumothorax diagnosis algorithm would be useful in many clinical scenarios. Recently, deep learning methods have achieved impressive progress in medical image segmentation tasks. However, a large-scale dataset is one of the critical components for the success of deep learning. On the other hand, there are few public chest X-ray images with pneumothorax.

To stimulate the researchers' interest in the pneumothorax diagnosis algorithm, we released a new data set PTX-498 here. It contains 498 chest X-ray images of pneumothorax collected from three hospitals, and each image contains pixel-level annotations. All images were resized to 1024×1024. The raw image intensity was clipped within the range from 2.5th to 97.5th percentile and then normalized to 0 to 255. The contours of the pneumothorax area were labelled by two senior radiologists using ITK-SNAP. The dataset was anonymized and every record related to patients' privacy was removed. Only the image data and the corresponding labels were included in PTX-498.

Dataset

This dataset was created by 废物和菜鸡

Contents