Table 1
Search strings used in different databases.
| DATABASE | SEARCH SYNTAX | N | |
|---|---|---|---|
| PubMed | ((stroke*[Title/Abstract] OR cva*[Title/Abstract] OR poststroke[Title/Abstract] OR post-stroke[Title/Abstract] OR apoplex*[Title/Abstract]) OR ((brain[Title/Abstract] OR cerebell*[Title/Abstract] OR intracran*[Title/Abstract] OR intracerebral[Title/Abstract] OR vertebrobasilar[Title/Abstract]) AND (haemorrhag*[Title/Abstract] OR hemorrhag*[Title/Abstract] OR ischemi*[Title/Abstract] OR ischaemi*[Title/Abstract] OR infarct*[Title/Abstract] OR haematoma*[Title/Abstract] OR hematoma*[Title/Abstract] OR bleed*[Title/Abstract])) AND (rehabilit*[Title/Abstract] OR neurorehabilit*[Title/Abstract] OR treat*[Title/Abstract] OR heal*[Title/Abstract] OR restor*[Title/Abstract] OR cur*[Title/Abstract] OR improve*[Title/Abstract] OR recov*[Title/Abstract]) AND (virtual reality[Title/Abstract] OR Oculus Rift[All Fields] OR HTC Vive[All Fields] OR immersive[Title/Abstract] OR virtual environment[Title/Abstract]) AND (Review[Filter]) | 100 | |
| Web of Science | TS = ((stroke OR cva* OR poststroke OR post-stroke OR apoplex*) OR ((brain OR cerebell* OR intracran* OR intracerebral OR vertebrobasilar) AND (haemorrhag* OR hemorrhag* OR ischemi* OR ischaemi* OR infarct* OR haematoma* OR hematoma* OR bleed* OR damage))) | #1 | 302 |
| TS = (rehabilit* OR neurorehabilit* OR treat* OR heal* OR restor* OR cur* OR improve* OR recov*) | #2 | ||
| TS = (virtual reality OR Oculus Rift OR HTC Vive OR immersive OR virtual environment) | #3 | ||
| TS = (systematic AND review OR review OR meta-analysis OR literature AND search) | #4 | ||
| (#4 AND #3 AND #2 AND #1) OR ((#3 AND #2 AND #1) AND DOCUMENT TYPES: (Review)) | |||
| Scopus | TITLE-ABS-KEY (virtual AND reality) AND TITLE-ABS-KEY (rehabilit*) AND TITLE-ABS-KEY (stroke) AND (LIMIT-TO (DOCTYPE, “re”)) | 121 | |
[i] Note: N = number of search results. We used the search string for “stroke” developed by Veerbeek et al. (2014). The search string for “rehabilitation” was iteratively optimized and the search string for “virtual reality” was restricted as we only wanted to include reviews that use the term “virtual reality”.
Figure 1
Flow chart of in- and exclusion of reviews.
Figure 2
Review characteristics. Panel A represents the number of reviews for different review characteristics. Panel B depicts the number of reviews per publication year. GH = general health conditions, ND = neurological disorders, ABI = acquired brain injury, ADL = activities of daily living.
Figure 3
Panel A represents the strongest and Panel B the weakest connections in the network of review characteristics. Colors represent the number of times that two features co-occurred across the included reviews. The width of the edges are scaled according to the number of reviews in which a pair of features co-occurred. The size of the nodes is scaled according to the number of reviews in which the feature occurred. GH = general health conditions, ND = neurological disorders, ABI = acquired brain injury, ADL = activities of daily living.
Figure 4
Hierarchical clustering of reviews based on aetiology, functional impairment, the main question of the review and whether the review discussed VR or multiple rehabilitation methods including VR. A list of references for each article ID is available online (https://doi.org/10.6084/m9.figshare.11902614.v1). Labels in bold font with a * represent reviews that focused on VR rehabilitation, while labels in plain font represent reviews that focused on multiple rehabilitation methods including VR. ND = neurological disorders, ABI = acquired brain injury, ADL = activities of daily living.
Figure 5
Terms used to define VR (A) and terms used to distinguish VR categories (B).
Figure 6
Network of terms used in VR definitions. Panel A represents the strongest connections and panel B represents the weakest connections in the network. The color represents the number of times that two features co-occurred across the included reviews. Connections smaller than 10 reviews are not visualized. The width of the edges are scaled according to the number of reviews in which a pair of features co-occurred. The size of the nodes is scaled according to the number of reviews in which the feature occurred.
Figure 7
Taxonomy of mixed reality rehabilitation systems, based on the extent to which real and virtual information is mixed, the level of immersion and the main input device. AR = augmented reality, AVR = augmented virtual reality, IVR = immersive virtual reality, SIVR = semi-immersive virtual reality. HMD = head-mounted display. Permission to re-use images was provided by the original copyright owners in written format (Cho et al., 2015; Cordes et al., 2018; Gil-Gómez et al., 2011; Rizzo et al., 2004; Tobler-Ammann et al., 2017; van Kessel et al., 2013; Yin et al., 2016). The category SIVR desktop systems are not illustrated, because articles describing these systems typically do not include pictures of the hardware.
Figure 8
Frequency of different input and output devices (A) and mixed reality systems (B).
Figure 9
Frequencies of different functional impairments for each mixed reality system (A) and frequencies of mixed reality subtype for each publication year (B).