TABLE VI.
Morphological information: details on the clinical aim, pathology/anatomical area, type of data, dataset, AI method, benchmark measure, and results.
| First author (year of publication) | Clinical aim | Pathology/anatomical area | Type of data | Dataset | AI method(s) | Benchmark measure | Results |
|---|---|---|---|---|---|---|---|
| Morpholoical information | |||||||
| Yepes (2018)90 | Determine the quantity of CSF | Neurological disorders | MRI-T1WI | 44 | SVM | Acc | 94% |
| Cherukuri (2018)91 | Determine the quantity of CSF | Neurological disorders | CT | 15 | CNN | Time | 0.003 s |
| Thillaikkarasi (2019)92 | Early detection of brain tumor | Brain tumor | MRI | 40 | CNN SVM | Acc Error Time | 98% 15% 15 ms |
| Sharma (2019)93 | Simulating tissue deformation and locating cancerous nodes | Brain tumor | MRI-T1WI | 6 | HMM | Acc PSNR MSE FRDD | 88% 21 985 mm 72% |
| Pushpa (2019)94 | Detect and classify the tumor type | Brain tumor | MRI | 60 | SVM | Acc | 99% |
| Rundo (2018)99 | Necrosis extraction of brain tumor | Brain tumor | MRI | 32 | Fuzzy C-Means | DSI MAD | 95.93% 0.22 pixel |
| Laukamp (2019)95 | Volumetric assessment of meningiomas | Brain tumor | MRI-T1WI MRI-T2WI | 56 | CNN FCNN | DSI | 81% |
| Chen (2019)96 | Detect mutations in aniopharyngioma patients | Brain tumor | MRI-T1WI | 44 | RF | AUC Acc Sp Se | 89% 86% 85% |
| Soltanine (2018)97 | Segmentation of brain tumor | Brain tumor | MRI MRI-DTI | 30 | RF | DSI Se Error | 89% 96% 2% |
| Sengupta (2018)98 | Segmentation of brain tumor | Brain tumor | MRI-T1WI MRI-T2WI | 9 | SVM | Error | 8.2% |
| Perkuhn (2018)100 | Segmentation of brain tumor | Brain tumor | MRI-T1WI MRI-T2WI MRI-FLAIR | 64 | CNN FCNN | DSI | 86% |
| Liu (2018)101 | Segmentation of brain tumor | Brain tumor | MRI | … | CNN SVM | DSI Acc | 77.03% 94.85% |
| Fabelo (2018)102 | Segmentation of brain tumor | Brain tumor | HSI | 5 | K-means | Acc Se Se | 99% 96% 96% |
| Binaghi (2019)103 | Segmentation of meningiomas | Brain tumor | MRI-T1WI MRI-T2WI | 15 | SVM | JD DSI Error | 81% 88.9% 21.74% |
| Sundaresan (2019)104 | Lesion segmentation | Brain lesions | MRI-T1WI MRI-T2WI MRI-FLAIR | 60 | Supervised learning LOCATE | DSI | 70% |
| Praveen (2018)105 | Segmentation of ischemic stroke lesion | Brain lesion | MRI | 28 | SAE SVM | DSI Sp Acc Se | 94.3% 96.8% 90.4% 92.4% |
| Remedios (2019)106 | Segmentation of brain injury | Brain injury | CT | … | 3 ANN | DSI PCC | 64% 87% |
| Park (2019)107 | Segmentation for DBS | Parkinson's disease | MRI-T2WI | 102 | FCNN | DSI Acc JD | 90.2% 90.4% 81.3% |
| Hadar (2018)108 | Hippocampal segmentation in temporal lobe epilepsy | Epilepsy | MRI-T1WI | 47 | CLNet | DSI | 85% |
| Li (2020)109 | Cerebrovascular segmentation | Cerebral artery | MRI-T1WI | 109 | HMM | DSI | 93% |
| Lee (2019)110 | AVM identification and quantification | Cerebral artery | MRI-T2WI | 39 | Fuzzy C-Means | DSI Se Sp | 79.5% 73.5% 85.5% |