Health Data Science / 2022 / Article / Tab 2

Review Article

Knowledge Graph Applications in Medical Imaging Analysis: A Scoping Review

Table 2

Overview of datasets used in the disease classification articles.

RefYearTaskDatasetDataset info

[40]2019Binary classificationLIDC-IDRI [53]1,018 chest CT scans with lung nodules present. CT scans were obtained from seven institutions
[45]2020Multilabel classificationIU X-ray [37]3,955 radiology reports, 7470 chest X-ray images
[41]2020Binary classificationCOVID-19 CT report [54]349 COVID-19 images and 379 non-COVID images and their corresponding Chinese reports
Chest X-ray images (pneumonia) [55]5,863 chest X-ray images with two classes: pneumonia and Normal
[50]2020Multilabel classificationCheXpert [39]224,316 chest radiographs of 65,240 patients, with 14 common disease labels
ChestX-Ray14 [38]112,120 frontal-view X-ray images, with the text-mined 14 common disease labels
[46]2021Multilabel classificationIU X-ray [37]3,955 radiology reports, 7470 chest X-ray images
MIMIC-CXR [56]377,110 chest X-ray images and the related 227,835 reports
[47]2021Multilabel classificationCheXpert [39]224,316 chest radiographs of 65,240 patients, with 14 common disease labels
NIH chest X-ray [38]112,120 frontal-view X-ray images with the text-mined 14 common disease labels
[48]2021Multilabel classificationChest ImaGenome [57]242,072 images and the corresponding scene graphs
[42]2021Binary classificationAutism brain imaging data exchange (ABIDE) [58]fMRI and the corresponding phenotypic data of 1,112 subjects
[49]2021Multilabel classificationCheXpert [39]224,316 chest radiographs of 65,240 patients, with 14 common disease labels
[43]2021Binary classificationDDSM [59]2,620 scanned film mammography studies.
[52]2021Binary classification and multilabel classificationCOVID-19 [60], COVID-19 Radiography [61, 62]150 CXR of COVID-19, 150 other pneumonia and another 150 instances for normal CXR images
[51]2021Binary classificationCOVID-19 multimodal dataset1,393 doctor–patient dialogues and 3706 images (347 X-ray +2,598 CT +761 ultrasound) about COVID-19 patients and 607 non-COVID-19 patient dialogues and 10,754 images (9658 X-ray + 494 CT +761 ultrasound)
[44]2021Multi-label classification7PC [63]1,011 lesion cases, and report comprehensive results