Table 2.

Summary of included studies evaluating the Leap Motion Controller.

Study	Aim	Type of study	Intervention	Sample	Results/Conclusions
[63]	To evaluate the implementation of a low-cost device for touchless PACS control in an interventional radiology suite. To demonstrate that interaction with gestures can decrease the duration of the procedures, the risk of re-intervention, and improve technical performance.	Proof-of-concept and prototype feasibility testing.	Manipulation of images in interventional radiology.	Interventional radiology suite.	The LMCa is a feasible, portable and low-cost alternative to other touchless PACS interaction systems. A decrease in the need for re-intervention was reported, but no explanation was given of how it was measured.
[54]	To present the first experience of using new systems for image control in the operating room: the LMC and OsiriX.	Proof-of-concept.	Manipulation of CTb and MRIc images.	2 general surgeons, 1 urologist, 3 orthopedic surgeons and 2 surgeons	The average training time was 5 min. The system is very cost-effective, efficient and prevents contamination during surgery. First experience of using the LMC to control CT and MRI images during surgery.
[116]	To validate the possibility of performing precise telesurgical tasks by means of the LMC.	Comparative study of the Sigma.7 electro-mechanical device and the LMC.	Peg transferring task and answering a questionnaire. The success rate of peg transfers.	10 researchers.	The results allowed the authors to confirm that fine tracking of the hand could be performed with the LMC. The observed performance of the optical interface proved to be comparable with that of traditional electro-mechanical devices.
[87]	To describe a piece of software for image processing with OsiriX using finger gestures.	Proof-of-concept.	Manipulation of radiological images.	Not described.	It is possible to implement gesture control of medical devices with low-cost, minimal resources. The device is very sensitive to surface dirt and this affects performance. The device favors the occlusion phenomenon.
[113]	To evaluate 2 contactless hand tracking systems, the LMC and MKd, for their potential to control surgical robots.	Experiment.	Manipulation of robots in surgery.	4 trained surgeons.	Neither system has the high level of accuracy and robustness that would be required for controlling medical robots.
[129]	To evaluate the LMC for simple 2-dimensional interaction and the action of entering a value.	Proof-of-concept and prototype testing.	Manipulation of medical information and operating room lights.	A 90-min conference on computer science and untrained users.	The user cases should be carefully classified and the most appropriate gestures for each application should be detected and implemented. Optimal lighting conditions for the LMC have still not been evaluated as unwanted light with deterioration of the IR light emitted may lead to a reduction in the recognition rate.
[81]	To compare the average time required by the conventional method using a mouse and an operating method with a finger-motion sensor.	Observational study.	Manipulation of angiographic images.	11 radiologists who observed a simulated clinical case.	After a practice time of 30 min, the average operation time by the finger method was significantly shorter than that by the mouse method.
[14]	To develop a workstation that allows intraoperative touchless control of diagnostic and surgical images in dentistry.	Prototype user testing.	Manipulation of radiological images.	2 surgeons. A case series of 11 dental surgery procedures.	The system performed very well. Its low cost favors its incorporation into clinical facilities of developing countries, reducing the number of staff required in operating rooms.
[88]	To propose an interface to control hand gestures and gestures with hand-held tools. In this approach, hand-held tools can become gesture devices that the user can use to control the images.	Prototype user testing.	Manipulation of ultrasound images.	12 participants.	Users were able to significantly improve their performance with practice.
[56]	To develop a software application for the manipulation of a 3De pancreatic or liver tumor model by using CT and real-time elastography data.	Proof-of-concept.	Manipulation of CT and real-time elastography images.	15 patients with liver cancer and 10 patients with pancreatic cancer.	A 3D model of liver and pancreatic tumors was successfully implemented with a hands-free interaction device suitable for sterile environments and for aiding diagnostic or therapeutic interventions.
[117]	To present a new gesture recognition system for manipulating 2 surgical robots in a virtual simulator.	Proof-of-concept.	Manipulation of robots in surgery.	2 surgical robots in a virtual simulator.	The device provided satisfactory accuracy and speed. It requires a more complete Application Programming Interface.
[90]	To propose a web-based interface to retrieve medical images using gestures.	User testing. Pilot study.	Manipulation of radiological images.	2 users.	User feedback was positive. Users reported fatigue with prolonged use of gestures. Additional studies are required to validate the interface.
[64]	To describe the use of the LMC for image manipulation during hepatic transarterial chemoembolization and internal radiotherapy procedures.	Proof-of-concept.	Manipulation of images in interventional radiology.	Not described.	Gesture-based imaging control may lead to increased efficacy and safety with decreased radiation exposure during hepatic transarterial chemoembolization procedures.
[77]	To compare 2 commercial motion sensors (MK and the LMC) to manipulate CT images, in terms of their utility, usability, speed, accuracy and user acceptance.	Two-strand sequential observational study. Qualitative and quantitative descriptive field study using a semi-structured questionnaire.	Manipulation of CT images.	42 participants: radiologists, surgeons and interventional radiologists.	Marginal to average acceptability of the 2 devices. MK was found to be more useful and easier to use, but the LMC was more accurate. Further research is required to establish the design specifications, installation guidelines and user training requirements to ensure successful implementation in clinical areas.
[91]	To evaluate a new method for image manipulation using a motion sensor.	Observational study. User testing and proof-of-concept.	Manipulation of radiological images in dentistry.	14 students. 6 images.	Using the system, several processes can be performed quickly with finger movements. Using gestures was significantly superior to using a mouse in terms of time.
[92]	To develop a new system for manipulating images using a motion sensor.	Observational study.	Manipulation of radiological images in dentistry.	14 students. 25 images.	The operation time with the LMC was significantly shorter than with the conventional method using a mouse.
[108]	To design a virtual 3D online environment for motor skills learning in MISf using exercises from the MISR-VR. The environment is designed in Unity, and the LMC is used as the device for interaction with the MIS forceps.	Letter to the editor.	None.	Not described	If it can be shown that 3D online environments mediated by natural user interfaces enable motor skills learning in MIS, a new field of research and development in the area of surgical simulation will be opened up.
[124]	Patent for accurate 3D instrument positioning.	Patent.	None.	Not described	Representing, on an output display, 3D positions and orientations of an instrument while medical procedures are being performed.
[69]	To describe the configuration for using the LMC in neurosurgery for image manipulation during a surgical procedure.	User testing.	Manipulation of images during a surgical procedure.	Resection of a meningioma and sarcoma surgery.	The learning curve only took 30 min. Although the main disadvantage was the lack of standardization of the gestures, the LMC is a low-cost, reliable and easily personalized device for controlling images in the surgical environment.
[109]	To develop skills in students and professionals using computer simulation technologies based on hand gesture capture systems.	User testing.	Description of the virtual environment.	Not described.	Simulation and new gesture recognition technologies open up new possibilities for the generation of computer-mediated procedures for medical training.
[93]	To present a gesture-controlled projection display that enables a direct and natural physician-machine interaction during CT-based interventions.	User testing (pilot and main).	8 tasks manipulating CT images.	12 participants (biomedical engineers, medical students and radiologists).	Gesture recognition is robust, although there is potential for improvement. The gesture training times are less than 10 min, but vary considerably between study participants.
[94]	To develop an anatomy learning system using the LMC.	User testing.	Manipulation of 220 anatomical images.	30 students and lecturers from an anatomy department.	The anatomy learning system using the LMC was successfully developed and it is suitable and acceptable as a support tool in an anatomy learning system.
[123]	To study the possibility of tracking laparoscopic instruments using the LMC in a box trainer.	Experiment.	3 static experiments and 1 dynamic experiment.	1 user.	The LMC had acceptable precision for tracking laparoscopic instruments in a box trainer.
[126]	To assess the potential of the LMC to track the movement of hands using MIS instruments.	Construct validity, concurrent validity. Comparative study with the InsTrac.	Passing a thread through pegs using the eoSim simulator.	3 experts and 10 novices.	The LMC is able to track the movement of hands using instruments in a MIS box simulator. Construct validity was demonstrated. Concurrent validity was only demonstrated for time and instrument path distance. A number of limitations to the tracking method used by LMC have been identified.
[118]	To explore the use of the LMC in endonasal pituitary surgery and to compare it with the Phantom Omni.	Comparative study between the LMC and the Phantom Omni.	16 resections of simulated pituitary gland tumors using a robot manipulated by the Phantom Omni and by the LMC.	3 neurosurgeons.	Users were able to achieve a very similar percentage of resection and procedure duration using the LMC.
[95]	To try to interact with medical images via a web browser using the LMC.	Prototype user testing.	Rotation, panning, scaling and selection of slices of a reconstructed 3D model based on CT or MRI.	1 user.	It is feasible to build this system and interaction can be carried out in real time.
[58]	To analyze the value of 2 gesture input modalities (the Myo armband and the LMC) versus 2 clinically established methods (task delegation and joystick control).	User study. Comparative study.	Simulating a diagnostic neuroradiological vascular treatment with 2 frequently used interaction tasks in an experimental operating room.	10 neuroradiologists	Novel input modalities have the potential to carry out single tasks more efficiently than clinically established methods.
[120]	To investigate the potential of a virtual reality simulator for the assessment of basic laparoscopic skills, based on the LMC	Face and construct validity.	3 basic tasks: camera navigation, instrument navigation, and two-handed operation.	2 groups of surgeons (28 experts and 21 novices).	This study provides evidence of the potential use of the LMC for assessing basic laparoscopic skills. The proposed system allows the dexterity of hand movements to be evaluated.
[52]	To evaluate the feasibility of using 3 different gesture control sensors (MK, the LMC and the Myo armband) to interact in a sterile manner with preoperative data as well as in settings of an integrated operating room during MIS.	Pilot user study.	2 hepatectomies and 2 partial nephrectomies on an experimental porcine model.	3 surgeons	Natural user interfaces are feasible for directly interacting, in a more intuitive and sterile manner, with preoperative images and integrated operating room functionalities during MIS. The combination of the Myo armband and voice commands provided the most intuitive and accurate natural user interface.
[127]	To evaluate the LMC as a tool for the objective measurement and assessment of surgical dexterity among users at different experience levels.	Construct validity study.	Surgical knot tying and manual transfer of objects.	11 participants.	The study showed 100% accuracy in discriminating between expert and novice performances.
[66]	To design an affordable and easily accessible endoscopic third ventriculostomy simulator based on the LMC, and to compare it with the NeuroTouch for its usability and training effectiveness.	Concurrent and construct validity study.	4 ellipsoid practice targeting tasks and 36 ventricle targeting tasks.	16 novice users and 2 expert neurosurgeons	An easy-access simulator was created, which has the potential to become a training tool and a surgical training assessment tool. This system can be used for planning procedures using patient datasets.
[119]	To present the LMC as a novel control device to manipulate the RAVEN-II robot.	Comparative study between the LMC and the electro-mechanical Sigma.7.	Comparison of peg manipulations during a training task with a contact-based device (Sigma.7).	3 operators.	With contactless control, manipulability is not as good as it is with contact-based control. Complete control of the surgical instruments is feasible. This work is promising for the development of future human-machine interfaces dedicated to robotic surgical training systems.
[98]	To evaluate the effect of using virtual reality surgery on the self-confidence and knowledge of surgical residents (the LMC and Oculus Rift).	Multisite, single-blind, parallel, randomized controlled trial.	The study group used the virtual reality surgery application. The control group used similar content in a standard presentation.	95 residents from 7 dental schools.	Immersive virtual reality experiences improve the knowledge and self-confidence of the surgical residents.
[97]	To develop and validate a novel training tool for Le Fort I osteotomy based on immersive virtual reality (the LMC and Oculus Rift).	Face and content validity.	A pre-intervention questionnaire to understand training needs and a postintervention feedback questionnaire.	7 consultant oral and maxillofacial surgeons.	The results confirmed the clinical applicability of virtual reality for delivering training in orthognathic surgery.
[70]	To investigate the feasibility and practicability of a low-cost multimodal head-mounted display system in neuroendoscopic surgery (the LMC and Oculus Rift).	Proof-of-concept in the operating room.	Ventriculocysto- cisternostomy. Ventriculostomy. Tumoral biopsy.	21 patients with ventricular diseases. 1 neurosurgeon.	The head-mounted display system is feasible, practical, helpful, and relatively cost efficient in neuroendoscopic surgery.

^aLMC: Leap Motion Controller.

^bCT: Computed Tomography.

^cMRI: magnetic resonance imaging.

^d3D: 3-dimensional.

^eMK: Microsoft Kinect.

^fMIS: minimally invasive surgery.