Dataset Information

COVID19-CT-dataset: an open-access chest CT image repository of 1000+ patients with confirmed COVID-19 diagnosis.

ABSTRACT:

Objectives

The ongoing Coronavirus disease 2019 (COVID-19) pandemic has drastically impacted the global health and economy. Computed tomography (CT) is the prime imaging modality for diagnosis of lung infections in COVID-19 patients. Data-driven and Artificial intelligence (AI)-powered solutions for automatic processing of CT images predominantly rely on large-scale, heterogeneous datasets. Owing to privacy and data availability issues, open-access and publicly available COVID-19 CT datasets are difficult to obtain, thus limiting the development of AI-enabled automatic diagnostic solutions. To tackle this problem, large CT image datasets encompassing diverse patterns of lung infections are in high demand.

Data description

In the present study, we provide an open-source repository containing 1000+ CT images of COVID-19 lung infections established by a team of board-certified radiologists. CT images were acquired from two main general university hospitals in Mashhad, Iran from March 2020 until January 2021. COVID-19 infections were ratified with matching tests including Reverse transcription polymerase chain reaction (RT-PCR) and accompanying clinical symptoms. All data are 16-bit grayscale images composed of 512 × 512 pixels and are stored in DICOM standard. Patient privacy is preserved by removing all patient-specific information from image headers. Subsequently, all images corresponding to each patient are compressed and stored in RAR format.

SUBMITTER: Shakouri S

PROVIDER: S-EPMC8114670 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:PurposeTo describe a large, publicly available dataset comprising computed tomography (CT) projection data from patient exams, both at routine clinical doses and simulated lower doses.Acquisition and validation methodsThe library was developed under local ethics committee approval. Projection and image data from 299 clinically performed patient CT exams were archived for three types of clinical exams: noncontrast head CT scans acquired for acute cognitive or motor deficit, low-dose noncontrast chest scans acquired to screen high-risk patients for pulmonary nodules, and contrast-enhanced CT scans of the abdomen acquired to look for metastatic liver lesions. Scans were performed on CT systems from two different CT manufacturers using routine clinical protocols. Projection data were validated by reconstructing the data using several different reconstruction algorithms and through use of the data in the 2016 Low Dose CT Grand Challenge. Reduced dose projection data were simulated for each scan using a validated noise-insertion method. Radiologists marked location and diagnosis for detected pathologies. Reference truth was obtained from the patient medical record, either from histology or subsequent imaging.Data format and usage notesProjection datasets were converted into the previously developed DICOM-CT-PD format, which is an extended DICOM format created to store CT projections and acquisition geometry in a nonproprietary format. Image data are stored in the standard DICOM image format and clinical data in a spreadsheet. Materials are provided to help investigators use the DICOM-CT-PD files, including a dictionary file, data reader, and user manual. The library is publicly available from The Cancer Imaging Archive (https://doi.org/10.7937/9npb-2637).Potential applicationsThis CT data library will facilitate the development and validation of new CT reconstruction and/or denoising algorithms, including those associated with machine learning or artificial intelligence. The provided clinical information allows evaluation of task-based diagnostic performance.

Project description:BackgroundImaging repositories are commonly attached to ongoing clinical trials, but capturing, transmitting, and storing images can be complicated and labor-intensive. Typical methods include outdated technologies such as compact discs. Electronic file transfer is becoming more common, but even this requires hours of staff time on dedicated computers in the radiology department.MethodsWe describe and test an image capture method using smartphone camera video-derived images of brain computed tomography (CT) scans of traumatic intracranial hemorrhage. The deidentified videos are emailed or uploaded from the emergency department for central adjudication. We selected eight scans, mild moderate, and severe subdural and multicompartmental hematomas and mild and moderate intraparenchymal hematomas. Ten users acquired data using seven different smartphones. We measured the time in seconds it took to capture and send the files. The primary outcomes were hematoma volume measured by ABC/2, Marshall scale, midline shift measurement, image quality by a contrast-to-noise ratio (CNR), and time to capture. A radiologist and an imaging scientist applied the ABC/2 method and calculated the Marshall scale and midline shift on the data acquired on different smartphones and the PACS in a randomized order. We calculate the intraclass correlation coefficient (ICC). We measured image quality by calculating the contrast-to-noise ratio (CNR). We report summary statistics on time to capture in the smartphone group without a comparator.ResultsICC for lesion volume, midline shift, and Marshall score were 0.973 (95% CI 0.931, 0.994), 0.998 (95% CI: 0.996, 0.999), and 0.973 (0.931, 0.994), respectively. Lesion conspicuity was not different among the image types via assessment of CNR using the Friedman test, [Formula: see text] of 24.8, P = < .001, with a small Kendall's W effect size (0.591). Mean (standard deviation) time to capture and email the video was 60.1 (24.3) s.ConclusionsTypical smartphones may produce video image quality high enough for use in a clinical trial imaging repository. Video capture and transfer takes only seconds, and hematoma volumes, Marshall scales, and image quality measured on the videos did not differ significantly from those calculated on the PACS.

Dataset Information

COVID19-CT-dataset: an open-access chest CT image repository of 1000+ patients with confirmed COVID-19 diagnosis.

Objectives

Data description

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets