Table 3.
The characteristics of the patients included in this study.
| ID | Study (author, yr) | Country | Sample | Patient | Normal people | Model | Inputs | Outputs |
|---|---|---|---|---|---|---|---|---|
| 1 | Abbasian, 2020 | Iran | 612 | 306 | 306 | K-nearest neighbor, DTL, ensemble model, | Chest HRCT | Classification (COVID-19; non-COVID-19) |
| 2 | Ahuja, 2020 | India | 406 | 95 | 72 | ResNet, SqueezeNet | CT scans | Classification (COVID-19; non-COVID) |
| 3 | Amyar, 2020 | France | 150 | 50 | 100 | CNN,DenseNet, ensemble model, ResNet, AlexNET, VGG, EffcientNet, Inception V3 | CT scans | Classification (Covid-19+; Normal; Others) + Two images (Image reconstruction; Infection and segmentation) |
| 4 | Attallah, 2020 | China | 744 | 347 | 397 | GoogleNet, ShuffleNet, ensemble model, AlexNET, ResNet | CT scans | Classification (COVID-19; non-COVID-19) |
| 5 | Gozes, 2020 | China, US | 206 | 56 | 100 | ResNet | Full thoracic CT | A lung abnormality localization map; Quantitative opacity measurements |
| 6 | Harmon, 2020 | China, Japan, Italy, US | 2617 | 326 | 1011 | DenseNet, ensemble model | Whole lung regions of CT scans | Classification (yes COVID-19; no COVID-19) |
| 7 | Jaiswal, 2020 | India | 374 | 190 | 184 | VGG, DenseNet, ensemble model | CT scans | Classification (COVID-19 (+); COVID-19 (−)) |
| 8 | Gifani, 2020 | China | 387 | 216 | 171 | Xception, DenseNet, Inception V3, ensemble model, ResNet, EffcientNet | CT scans | Classification |
| 9 | Horry, 2020 | Australia,US, China | 150 | 81 | 69 | VGG | X-Ray, Ultrasound, CT scan | Classification (COVID-19; Normal; Pneumonia) |
| 10 | Jin, 2020 | China | 2688 | 751 | 1937* | ResNet | Multichannel image, lung-masked slices | Classification (non-pneumonia; CAP; Influenza; COVID-19) |
| 11 | Kadry, 2020 | Lebanon, India | 500 | 250 | 250 | ensemble model, DTL, Random Forst, K-nearest neighbor | CT scans | Classification (Normal; COVID-19) |
| 12 | Krzysztof, 2020 | Poland | 203 | 98 | 105 | ensemble model, ResNet, DenseNet | Full CT lung scans, radiograph images (Front views & lateral views) | Classification (fungal pneumonia; COVID-19; healthy chest; viral pneumonia; bacterial pneumonia) |
| 13 | Liu, 2020 | China | 88 | 61 | 27 | DTL, ensemble model, Logistic regression, K-nearest neighbor, | CT scans | Classification (COVID-19; GP) |
| ID | Study (author, yr) | Country | Sample | Patient | Normal people | Model | Inputs | Outputs |
| 14 | Maghdid, 2020 | Iraq, UK | 23 | 17 | 6 | AlexNET, CNN | X-ray, CT scans | Classification (Negative; Positive) |
| 15 | Mobiny, 2020 | China | 105 | 47 | 58 | Inception V3, DenseNet, ResNet | X-ray, CT scans | Classification (Negative; Positive) |
| 16 | Pathak, 2020 | India | 530 | 270 | 260 | CNN, DTL, ResNet | Chest CT images | Classification |
| 17 | Pham, 2020 | US | 746 | 349 | 397 | Inception V3, ensemble model, AlexNET,VGG,ResNet,MoblieNet,ShuffleNet,DenseNet,GoogleNet, SqueezeNet,Xception | Chest CT images | Classification (COVID+COVID-) |
| 18 | Ragab, 2020 | Brazil | 120 | 60 | 60 | ensemble model, AlexNET, ResNet, GoogleNet, ShuffleNet | Whole CT image slices | Classification (COVID-19 pneumonia; Healthy) |
| 19 | Sharma, 2020 | Italy,India,China, Moscow | 2200 | 1400† | 800 | Inception V3 ensemble model, DenseNet, MoblieNet | CT scans | Classification (COVID-19; non-COVID-19) |
| 20 | Saeedi, 2020 | China | 746 | 349 | 397 | Inception V3, ensemble model, DenseNet, MoblieNet | CT scans (COVID-19 CT scans showing typical patches on the outer edges of the lung) | Classification (COVID-19; Normal health; Other viral pneumonia) |
| 21 | Yang, 2020 | China | 295 | 70 | 70 | DenseNet | CT scans | Classification (COVID; Non-COVID) |
| 22 | Zheng, 2021 | China | 659 | 262 | 397‡ | DenseNet, ResNet, VGG | CT scans | Classification (Patients; Healthy person) |
| 23 | Chen, 2021 | China | 610 | 39 | 53§ | ResNet | CT images (whole lung, include the chest wall and armpits on both sides) | Classification (Healthy; COVID-19; Bacterial Pneumonia; Typical Viral Pneumonia) (Image-level and human-level) |
| 24 | Gao, 2021 | China | 1202 | 656 | 423 | ResNet, CNN, VGG | CT scans | Classification (COVID-19; Normal control; Other pneumonias) |
| 25 | Javaheri, 2021 | US,Iran, Canada | 335 | 226∥ | 109 | CNN | Thick-section CT scans | Classification (Covid-19; normal) (image level and individual level) segmentation of lesions; FCN |
| 26 | Liu, 2021 | China | 2800 | 233 | 289 | DenseNet | 3D CT images | Classification (Covid-19; CAP; Control) |
| ID | Study (author, yr) | Country | Sample | Patient | Normal people | Model | Inputs | Outputs |
| 27 | Ozyurt, 2021 | China | 746 | 349 | 397 | CNN, DNN | A stack of 64 axial images of size 384 of whole chest CTs | Classification (COVID-19 pneumonia; non-COVID-19 pneumonia) |
| 28 | Shah, 2021 | US | 73 | 34 | 39 | Inception V3, VGG, ensemble model, DenseNet, ResNet | Chest CT images | Classification (COVID-19; Healthy) |
| 29 | Song, 2021 | China | 274 | 188¶ | 86 | VGG, ensemble model, DenseNet, ResNet | CT scans | Classification (COVID-19 positive; COVID-19 negative) |
| 30 | Tan, 2021 | China | 470 | 275 | 195 | VGG | Chest CT images | Classification (COVID-19; Bacteria pneumonia) (image-level prediction and individual-level prediction) |
| 31 | Zhu, 2021 | China | 1592 | 275 | 235 | VGG, ResNet, GoogleNet | CT scans | Classification (COVID-19; Normal) |
The table is shown to the summaries of the characteristics of the patients included in this study including demographics, clinical features and the inputs and outputs of the models.
COVID-19 = corona virus disease 2019.
*including 1229 non-pneumonia, 668 CAP, 42 Influenza.
†including 800 COVID-19, 600 Other viral pneumonia.
‡including 100 bacterial pneumonia, 219 typical viral pneumonia, 78 healthy.
§including 38 other pneumonias, 15 normal controls.
∥including 111 infections with CAP and 115 other viral sources, whose CT images may be misdiagnosed as COVID-19.
¶including 88 COVID-19, 100 patients infected with bacteria pneumonia.