Wen, J., Thibeau-Sutre, E., Samper-Gonzalez, J., Routier, A., Bottani, S., Durrleman, S., ... & Colliot, O. (2019).
Convolutional Neural Networks for Classification of Alzheimer's Disease: Overview and Reproducible Evaluation.
arXiv preprint arXiv:1904.07773. paper
- 2D slice-level CNN
- 3D patch-level CNN
- ROI-based CNN
- 3D subject-level CNN
- Other approach
- Use clinical data
paper | AD vs CN | sMCI vs pMCI | MCI vs CN | AD vs MCI | Multi-class | Approach | Data Leakage |
---|---|---|---|---|---|---|---|
Aderghal et al., 2017b | 0.84 | - | 0.65 | 0.67 | - | ROI-based | Non detected |
Aderghal et al., 2018 | 0.90 | - | 0.73 | 0.83 | - | ROI-based | Non detected |
Bäckström et al., 2018 | 0.90 | - | - | - | - | 3D subject-level | Non detected |
Cheng et al., 2017 | 0.87 | - | - | - | - | 3D patch-level | Non detected |
Cheng and Liu, 2017 | 0.85 | - | - | - | - | 3D subject-level | Non detected |
Islam and Zhang, 2018 | - | - | - | - | CN vs mild vs moderate vs severe 0.93 |
2D slice-level | Non detected |
Korolev et al., 2017 | 0.80 | - | - | - | - | 3D subject-level | Non detected |
Li et al., 2017 | 0.88 | - | - | - | - | 3D subject-level | Non detected |
Li et al., 2018 | 0.90 | - | 0.74 | - | - | 3D patch-level | Non detected |
Lian et al., 2018 | 0.90 | 0.80 | - | - | - | 3D patch-level | Non detected |
Mingxia Liu et al., 2018a | 0.91 | 0.78 | - | - | - | 3D patch-level | Non detected |
Mingxia Liu et al., 2018c | 0.91 | - | - | - | - | 3D patch-level | Non detected |
Qiu et al., 2018 | - | 0.83 | - | - | - | 2D slice-level | Non detected |
Senanayake et al., 2018 | 0.76 | - | 0.75 | 0.76 | - | 3D subject-level | Non detected |
Shmulev et al., 2018 | - | 0.62 | - | - | - | 3D subject-level | Non detected |
Valliani and Soni, 2017 | 0.81 | - | - | - | AD vs MCI vs CN 0.57 |
2D slice-level | Non detected |
paper | AD vs CN | sMCI vs pMCI | MCI vs CN | AD vs MCI | Multi-class | Approach | Data Leakage |
---|---|---|---|---|---|---|---|
Aderghal et al., 2017a | 0.91 | - | 0.66 | 0.70 | - | ROI-based | Unclear |
Basaia et al., 2019 | 0.99 | 0.75 | - | - | - | 3D subject-level | Unclear |
Hon and Khan, 2017 | 0.96 | - | - | - | - | 2D slice-level | Unclear |
Hosseini Asl et al., 2018 | 0.99 | - | 0.94 | 1.00 | AD vs MCI vs CN 0.95 |
3D subject-level | Unclear |
Islam and Zhang, 2017 | - | - | - | - | CN vs mild vs moderate vs severe 0.74 |
2D slice-level | Unclear |
Lin et al., 2018 | 0.89 | 0.73 | - | - | - | ROI-based | Unclear |
Manhua Liu et al., 2018 | 0.85 | 0.74 | - | - | - | 3D patch-level | Unclear |
Taqi et al., 2018 | 1.00 | - | - | - | - | 2D slice-level | Unclear |
Vu et al., 2017 | 0.85 | - | - | - | - | 3D subject-level | Unclear |
S.-H. Wang et al., 2018 | 0.98 | - | - | - | - | 2D slice-level | Unclear |
Bäckström et al., 2018 | 0.99 | - | - | - | - | 3D subject-level | clear |
Farooq et al., 2017 | - | - | - | - | AD vs LMCI vs EMCI vs CN 0.99 |
2D slice-level | clear |
Gunawardena et al., 2017 | - | - | - | - | AD vs MCI vs CN 0.96 |
3D subject-level | clear |
Vu et al., 2018 | 0.86 | - | 0.86 | 0.77 | AD vs MCI vs CN 0.80 |
3D subject-level | clear |
Wang et al., 2017 | - | - | 0.91 | - | - | 2D slice-level | clear |
Wang et al., 2019 | 0.99 | - | 0.98 | 0.94 | AD vs MCI vs CN 0.97 |
3D subject-level | clear |
Wu et al., 2018 | - | - | - | - | sMCI vs pMCI vs CN 0.95 |
2D slice-level | clear |