GCH Control Software implements GCH System to measure the loads applied to forearm crutches. Its application is particularly suitable in physiotherapy and traumatology, fields that frequently work on the functional recovery of gait for patients with lower limb musculoskeletal injury [40,41]. The rehabilitation process starts with total or partial unloading of the injured limb, followed by a gradual increase in loads aided by crutches until achieving complete functional recovery [42].

The instrumented crutches must first be described to understand how the software works. GCH System involves the coupling of a miniature force sensor inside the distal area of the crutch, and the measures are wirelessly transmitted. Each crutch contains an ultralow-power microcontroller with an input voltage of 2.4 V direct current and battery/autonomy of 6000 mAh (8 hours). The outgoing signal is detected by a USB receiver connected to a computer (“fixed system”) using a virtual communication port, or by a receptor built into a watch, mobile phone, pendant, or other portable device. In the latter case, patients have autonomy to practice aided gait outdoors with the usual obstacles encountered in daily life, such as steps and slopes, without any professional supervision. The technical specifications of GCH System 2.0 are reported elsewhere [6]. Each of these fixed and portable systems requires an independent implementation program with different features to allow each patient to achieve their specific therapeutic goals. The fixed system is described below in line with the main objective of the study.

The innovative multifunctional desktop program GCH Control Software 1.0 monitors crutch loads, which can be controlled by real-time graphic and numerical displays during assisted gait practice. The software contains a versatile feedback mechanism correcting any wrongly applied loads, ensuring greater clinical accuracy. The integrated database can be used not only to objectively evaluate the progress in loads of the subjects but also to comparatively analyze the system’s efficiency and draw up clinical action protocols. Moreover, we performed a pilot implementation trial to assess the functionality of the feedback mechanism.

The application environment is basically associated with orthopedics and traumatology [6,38,41,45], and particularly to temporary processes of functional gait recovery of patients with lower limb injuries or surgeries [45] such as ligamentoplasty of the anterior cruciate ligament of the knee [36], a knee or hip prosthesis [21], or a knee osteotomy [40]. In all of these cases, treatment of the lesion starts with no load on the injured limb, which is followed as recovery progresses by partial loads that gradually increase up to a total body weight load. The GCH System is also useful in the fields of gerontology [31,32], neurology [46], pediatrics [46,47], or rheumatology [33], as mentioned earlier. If the applications are extended to these fields where the lesions tend to be chronic, orthotic devices must also be maintained over time [48], increasing the opportunities of using the method. New research prospects providing innovative clinical action protocols for users with disabilities or even normal age deterioration are therefore possible.

This study, like many other research projects [4,15,23,43,49-51], advocates technology and information technology effectiveness in functional recovery processes in general, and specifically in gait. However, effectiveness does not only refer to the intervention itself, in this case based on technological feedback, but also extends to the (global or analytical) initial, continuous, and final assessments of each essential gait parameter [52]. GCH Control Software evaluates the maximum force applied to the crutches in each support and therefore indirectly the unloading of the affected lower limb, which is one of the essential parameters [1,39] in assisted gait with two-stage partial unloading. This peak clinical moment matches the midstance phase of the affected limb (ie, where the limb remains upright and with single stance or without the help of the other limb in the air; midswing phase of the contralateral limb) [53,54]. Unloading relies exclusively on the crutches. The software also monitors if the assistive devices touch the ground at the same time and if their loads are balanced and increase progressively without fluctuations. These data are related to other essential parameters of two-stage gait with forearm crutches, as shown by the CHAGS [1] scale of observational aided gait assessment. Three of its ten load-independent items are directly related to this information: “simultaneous support of both crutches and foot,” “step rhythm” (all steps have to be carried out in the same time), and “fluency” (assisted gait is performed automatically, decisively, and without hesitation).

The consequence drawn by the authors is that mutually complementary and objective technological and observational measurement tools need to be applied [55]. Similarly, despite a major goal of technology system-based clinical interventions to offer patients autonomy, they must also rely on the human eye, hands, intuition, and professional decisions [20]. In the specific case of the GCH Control Software feedback mechanism, the physiotherapist’s presence is shown to optimize the technology features.

GCH Control Software can be adapted very easily to each patient’s specific characteristics, regardless of their main disease, such as their level of coordination skills [56], previous experience [56] in handling forearm crutches, age, and visual or acoustic disorders. This is because the software provides a wide range of extrinsic feedback, ensuring individualized interventions for achieving specific or general therapeutic goals when recovering gait functions. This feedback can be provided to the clinical professional, the user, or both at the same time, although the clinical professional, usually a physiotherapist, receives more detailed and extensive information than the user who is performing a dual task [57], and should only be given very specific feedback in small amounts for the intervention to be effective.

As for the feedback modalities considered (ie, acoustic, visual, and acoustic-visual), patients receive two kinds of acoustic feedback: computer-based verbal feedback from the physiotherapist and the beeps generated by the software. The advantage of the verbal feedback [49] is that the physiotherapist can decide what information to give in line with the patient’s needs. This is particularly useful if the subject has poor coordination skills, as noted in the dual tasks [49]. Subjects can be given only information about their results (KR feedback) or also about their performance (KP feedback), and can receive information that only describes a mistake (descriptive feedback) or that suggests the strategy for correcting it (prescriptive feedback). The clinician can also choose between giving the information while the patient is walking (concurrent feedback) or afterward (terminal feedback), and among immediate terminal feedback, delayed terminal feedback, or summary terminal feedback (after several walks). The beeps are immediate terminal feedback, because subjects receive the signal immediately when they begin to take a step using support for the affected limb with crutches. The acoustic signal tells subjects if they have reached the correct maximum load at the end of each support, which is why it is given immediately. However, during free cadence walking [58], steps are taken very quickly, and the period of time between the maximum load and end of the support is too imperceptible so that the signal can be considered a real-time signal, and thus it can be interpreted clinically as concurrent feedback. This is particularly true if the gait is regarded as a cyclical process [59] and each crutch support or step is not assessed independently.

With regard to visual information, the main advantage is its continuity (continuous concurrent feedback) by showing how the loads applied on the crutches steadily increase and decrease in double-bar, single-bar, and balance-bar feedback. Moreover, the double-bar and single-bar feedback show if each crutch support is correct or incorrect (immediate terminal feedback).

The software’s multiple independent biofeedback options can be extended by using various combinations of these options. Numerous studies have assessed the effectiveness of various types of feedback [26,27,49,60,61] and their combinations. A recent systematic review [43] of technology-based feedback and its efficacy in improving gait parameters concluded that immediate visual feedback was the most widely used, followed by terminal and immediate acoustic feedback.

In line with several previous studies [11,23,27,51], the results obtained in this study suggest that feedback mechanism–based interventions are satisfactory. The pilot experiment showed that subjects applied significantly more accurate loads on the forearm crutches after only a few minutes, or even a few seconds in some cases. These preliminary data should be confirmed in subsequent longer studies with larger samples. Nevertheless, these findings suggest that interventions are more effective, and even more efficient, owing to the feedback immediacy. Although the participant inclusion criteria were favorable to achieve positive results (ie, all participants had to have experience in two-stage bilateral assisted walking with partial unloading, and their technical gesture had to be acceptable based on a score of 3-4 points for each CHAGS item [1]), only one participant (Subject 5) had previous experience of load measurements, and with the same software.

The crutch load measurements without feedback (ie, walk 0) showed that all subjects made significant overload and underload mistakes (even with twice the requested load). Most of the subjects (7/10) overloaded the crutches, thereby unloading their body weight more than necessary, with associated clinical disadvantages. For example, in older patients with osteoporosis [32,42], unnecessary unloading is not appropriate because it inhibits osteoblastic action and increases osteoclastic action. Even if there is no underlying disease, the tendency is always to use the largest possible weight-bearing load [42], without damaging the injury, to stimulate circulation and muscle tone, and therefore optimize the patient’s overall functional recovery. This is why frequent two-stage assisted walking without the injured lower limb touching the ground (foot in the air) is not recommended, unless necessary. The negative clinical consequences of underloading the crutches can be even more severe with certain injuries or types of surgery such as noncemented knee or hip prostheses, osteotomy [40,41], or autologous chondrocyte implantation [62]. In such cases, overloading the injured area can lead to further surgery, longer functional recovery times, or other side effects or sequelae, among other drawbacks [42]. Several subjects (3/10), despite being coordinated and having used crutches previously, made these mistakes constantly without feedback. The case of Subject 10 was particularly interesting, whose technical gesture with homogeneous loads were perfect, and remained well below the requested load on the crutches without feedback. This is a clear example of the usefulness of GCH Control Software, even with skilled and experienced patients.

In the three double-bar feedback walks, maximum crutch load accuracy errors decreased in all participants from the first to the third walk. At the start of the tests, and especially the first test, participants made more mistakes and differences between the requested and applied force were larger. Shortly afterward, the visual and acoustic information provided by the software helped them to adjust the loads. The biggest errors were made during the first walk; however, 3 subjects did not make any mistake in the third walk, and another 3 subjects only made one mistake, representing a minimum difference. One subject did not make any mistakes in the second or third walk. However, two of the subjects did not achieve the expected results in the double-bar walk.

Subject 2 began with very high loads and made 8/10 mistakes in the first walk, which decreased, but to a lesser extent than observed for the majority of participants (ie, 3 errors made in the second walk and 4 errors made in the third walk). During the 3 walks, it was observed that the crutches did not always load evenly, and the right and left bars fluctuated, even though the final weight of each crutch met the requirements (6/30 crutch balance errors). The physiotherapist decided to continue the intervention and apply the third type of feedback (balance bar) to allow the subject to focus their attention on leaning on both crutches at the same time and ensure even crutch loads, regardless of the total amount. As expected, the subject’s performance improved remarkably with only one walk, after scoring 3/4 points in the CHAGS scale item on simultaneously using both crutches and the affected limb [1]. Therefore, occasionally, the subject did not meet this goal. The GCH Control Software feedback seems to not only facilitate load monitoring and accuracy but can also detect and correct other erroneous parameters that are equally essential in crutch-assisted gait (eg, leaning on both crutches simultaneously as in the case of Subject 2). This subject is a clear example that technological assessment and feedback (GCH System linked with GCH Control Software), combined with observational assessment and direct feedback from the physiotherapist (CHAGS score), can be used together and in a complementary manner, thereby individualizing each intervention [1].

The number of errors made by Subject 9 dropped in the double-bar feedback walks (from 7 to 5); however, there were still numerous disparate inaccuracies, in line with the prefeedback walk. Nevertheless, the subject maintained the crutch load balance. After analyzing these data, the physiotherapist asked the subject to continue the intervention with the single-bar display, and the subject made only 3 mistakes in the first walk and 1 mistake in the second walk. This improvement was explained by having focused the subject’s attention [56] on the total load without displaying independent data from each crutch, which were not necessary in this case. The subject also walked more times and therefore spent more time practicing, which must also be taken into account for interpreting this improvement [56].

At the end of the pilot experiment, the subjects commented that they had been surprised by the difference between how they actually performed and how they thought they had performed. They found the test to be easy, comfortable, and effective. Even those who needed a few extra screens spoke positively about GCH Control Software and GCH System in general. They also stated that despite hearing the acoustic signals, they were generally more aware of the visual signals.

This study was limited to an extent by the sample size and short duration of the intervention. Longer prospective clinical trials should be performed, with large samples and control groups to ascertain the effectiveness and efficiency of the different types of graphical biofeedback (double bar, single bar, and balance), acoustic feedback (verbal and/or beep), or combined feedback available with GCH Control Software. For example, acoustic feedback using beeps and visual feedback could be applied separately, followed by evaluation of whether (and in which cases) acoustic feedback alone suffices in the long term. As explained earlier, the more options there are, the easier it is to customize or adapt the system to each patient during the treatment, both in line with individual characteristics (eg, coordination) and specific temporary musculoskeletal injuries or chronic situations due to illness or age. Clinical trials are therefore required to address different pathologies and contexts in the various branches of medicine to develop effective clinical action protocols.

Given the rapid progress of information technology, future adaptations of GCH Control Software to new communication systems will be necessary. However, these adaptations would be implemented at the lower levels of the software stack. The way feedback information is presented to patients and physiotherapists would not be affected by these adaptations.

This study describes GCH Control Software, the multifunctional desktop software associated with GCH System 2.0, which measures loads exerted on forearm crutches by patients who need to partially unload a pathological lower limb.

The program graphically and numerically monitors loads in real time during assisted gait interventions, and uses a versatile, efficient feedback mechanism to correct the accuracy of wrongly applied loads. It also shows if subjects use both crutches at the same time in bilateral walking and how evenly they distribute the load between them. The program features a patient and test database that can be used to objectify applied load progress and comparatively analyze the system’s effectiveness in different clinical contexts, enabling drawing up clinical action protocols based on scientific evidence.

Satisfactory findings were obtained in the complementary pilot implementation tests of the feedback mechanism functionality. Although an ideal sample was selected, all subjects made errors exerting loads on the forearm crutches during bilateral gait in the baseline test, indicating the necessity or appropriateness of the method presented. They achieved greater accuracy with only one brief session. Although one subject found it difficult to balance the loads between the two crutches and fluidly apply the loads, these issues were ultimately resolved with feedback. However, any clinical data provided herein should be interpreted with caution. Future research should focus on how participants’ accuracy changes over several weeks to confirm the findings.