Table 4.
AI research in spinal cord neurological injury and repair nine most representative literature co-citation clusters.
| Cluster-ID | Size | Silhouette | Label LLR |
|---|---|---|---|
| #0 | 73 | 0.923 | Technology |
| #1 | 67 | 0.93 | Exoskeleton |
| #2 | 57 | 0.927 | Treadmill training |
| #3 | 56 | 0.987 | Locomotion |
| #4 | 54 | 0.855 | Muscles |
| #5 | 53 | 0.824 | Rehabilitation robot |
| #6 | 46 | 0.967 | Neuromodulation |
| #7 | 40 | 0.962 | Gait |
| #8 | 40 | 0.957 | Actuation |
| #9 | 35 | 0.971 | Non-invasive brain stimulation |
| #10 | 32 | 0.934 | Brain-machine interface |
| #11 | 32 | 0.966 | Soft robotic glove |
| #12 | 30 | 1 | Deep learning |
| #13 | 26 | 0.966 | Muscle synergies |
| #14 | 25 | 0.936 | Robotic-assisted training |
The silhouettes are the average contour values of the clusters (Tables 3 and this table). Generally, groups with silhouette scores > 0.5 were accepted, and groups with silhouette scores > 0.7 had good clustering performances. The size represents the number of items in each group, and labels represent the clusters using the LLR algorithm.