Skip to main content

Table 1 Details of the datasets and data processing methods used in the examined studies

From: Enhancing cancer differentiation with synthetic MRI examinations via generative models: a systematic review

Author

Dataset

Patients

Lesion

Modality

Pre-processing

Objective

Beers [38]

BraTS’17

75 LGG

210 HGG

edema

non-enhancing/necrosis

contrast enhancing

T1w, T1ce

T2w, FLAIR

isotropic resample

zero mean

unit variance

Synthesize

Han [39]

BraTS’16

220 HGG

tumor

T1w, T1ce

T2w, FLAIR

skull striping

isotropic resample

resize

Synthesize

Han [41]

BraTS’16

220 HGG

tumor

T1ce

zero padding

Classification

Han [42]

BraTS’16

220 HGG

tumor

T1ce

zero padding

Classification

Han [44]

Private

180 tumor

193 normal

tumor

(brain metastases)

T1ce

skull-striping

crop, resize

Detection

Shin [46]

ADNI/

BraTS’15

54 LGG

220 HGG

edema

non-enhancing/necrosis

contrast enhancing

T1w, T1ce

T2w, FLAIR

Skull-striping/

crop axially, resize

Segmentation

Chang [47]

BraTS’18

210

edema

non-enhancing/necrosis

contrast enhancing

T2w

-

Segmentation

Chang [48]

BraTS’18

210

edema

non-enhancing/necrosis

contrast enhancing

T1w, T1ce

T2w, FLAIR

align

isotropic resample

resize

Segmentation

Slice imputation

Deepak [49]

Figshare

database

233

meningioma

glioma

pituitary tumor

T1ce

Resize

Classification

Qasim [50]

BraTS’19

335

edema

non-enhancing/necrosis

contrast enhancing

T1w, T2w, FLAIR

-

Segmentation

Kwon [52]

BraTS’18

210

edema

non-enhancing/necrosis

contrast enhancing

T2w, FLAIR

Resize

Synthesize

Chen [54]

BraTS’13/

IXI

-

edema

non-enhancing/necrosis

contrast enhancing/

Healthy

-

T1w

-

Classification

Pesteie [55]

BraTS’17

-

edema

non-enhancing/necrosis

contrast enhancing

FLAIR

-

Segmentation

Hamghalam [56]

BraTS’13

30

edema

non-enhancing/necrosis

contrast enhancing

T1ce

-

Segmentation

Qi [57]

BraTS’19

BraTSs

UNS

322

-

-

tumor

T1w, T1ce

T2w, FLAIR

-

Classification

Guo [58]

Private

90

GBM

T1w, Gd-T1w, T2w

FLAIR, APTw

co-registration, skull-stripping

N4-bias field correction

MRI intensity scale standardization

2D slice extraction

Segmentation

Guo [60]

Private

100

GBM

T1w, Gd-T1w, T2w

FLAIR, APTw

co-registration, skull-stripping

N4-bias field correction

MRI standardization

Segmentation

Ge [66]

TCGA-GBM

TCGA-LGG

167

IDH genotype

tumor

T1w, T1ce

T2w, FLAIR

-

Classification

Slice imputation

Ge [67]

TCGA-GBM

TCGA-LGG/

BraTS’17

167/

285

IDH genotype/

edema

non-enhancing/necrosis

contrast enhancing

T1w, T1ce

T2w, FLAIR

-

Classification

Carver [68]

BraTS’18

210

GBM

T1w, T1ce

T2w, FLAIR

resample, skull-stripping

normalization, padding

crop slices

Segmentation

Mok [69]

BraTS’15

220 HGG

54 LGG

edema

non-enhancing/necrosis

contrast enhancing

T1w, T1ce

T2w, FLAIR

skull-stripping

co-registration

Segmentation

Dikici [72]

Private

158

tumor

(brain metastases)

T1ce

isotropic resample

normalization

Detection

Kamli [75]

TCIA

ADNI

20

GBM

T1 pre & T1 post contrast

T2w, FLAIR

skull-striping

normalization/standardisation

registration, denoising

Prediction

Li [77]

Private

23 with PSP

61 with TTP

GBM

DTI

-

Classification

  1. LGG, low-grade gliomas; HGG, high-grade gliomas; T1ce, T1 contrast-enhanced; FLAIR fluid-attenuated inversion recovery; T1adjacent w, T1-weighted; T2w, T2-weighted; GBM, glioblastoma multiforme; Gd-T1w, gadolinium-enhanced; APTw, amide proton transfer weighted; IDH, isocitrate dehydrogenase; PSP, pseudoprogression; TTP, true tumor progression; DTI, diffusion tensor imaging, symbol ’-’ represents that the corresponding information was not provided in the publication