Browsing by Author "de Jager, Kylie"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Open Access
    Analysis of orthopaedic device development in South Africa: Mapping the landscape and understanding the drivers of knowledge development and knowledge diffusion through networks
    (2021) Salie, Faatiema; Douglas, Tania; de Jager, Kylie
    An orthopaedic medical device refers to a part, implant, prosthetic or orthotic which is used to address damage to the body's musculoskeletal system, primarily by providing stability and mobility. Orthopaedic medical devices play a role in injury-related disorders, which have been highlighted as a key element of the quadruple burden of disease in South Africa. In this thesis, orthopaedic devices are conceptualised as a technological field and a technological innovation system (TIS) framework is applied to understand orthopaedic device development in South Africa. Knowledge development and knowledge diffusion are fundamental components of any innovation system. The thesis hypothesises that the functions “knowledge development” and “knowledge diffusion through networks” of the orthopaedic devices TIS are influenced by contextual factors. The objectives of the study are: to identify the actors who generate knowledge for orthopaedic device development and to characterise the relationships between them; to identify focus areas of orthopaedic device development; to provide insight into the drivers and barriers to knowledge development and diffusion in the TIS; and to identify the contextual factors that influence knowledge dynamics in the TIS. These objectives are investigated using social network analysis based on bibliometric data (scientific publications and patents), keyword networks, a review of institutions, and a set of case studies where the primary data source are interviews with actors. Actors producing knowledge were from the university, healthcare, industry and science council sectors, although science councils played a small role. International actors were shown to bring new ideas into the TIS. The networks were fragmented, illustrating that knowledge diffusion through the networks was limited. This was especially the case in the patent networks as many actors patent in isolation. The keyword networks highlighted unrealised collaboration potential between actors based on their common research interests. The case studies revealed features of cross-sector interaction for orthopaedic device development not evident from network analysis based on bibliometric data. Drivers of knowledge development and knowledge diffusion were: inter-sectoral collaboration; the availability of resources; the affordability of available devices; and the positive externalities of allied TISs. The main barrier to knowledge development and diffusion was in the form of barriers to intersectoral collaboration. These include unmatched expectations from partners in collaboration, different views on intellectual property ownership, and burdensome university administrative processes. The orthopaedic devices TIS was structurally coupled to the embedded TIS and sectoral contexts, and externally linked and structurally coupled to its political context. Knowledge development and diffusion was found to be positively enhanced by innovation in the additive manufacturing TIS, with shared structural elements and resources. Knowledge development and diffusion was influenced by sectoral dynamics of the university, healthcare and industry sectors. This thesis makes the following contributions. First, it applies the TIS framework to a new focus area, namely medical device development, in a developing country context. Second, it makes two unique methodological contributions: it presents an index to capture the extent of sectoral collaboration in a network; and it develops a method for determining the collaboration potential of actors in a network based on cognitive distance.
  • Thumbnail Image
    Item
    Open Access
    Development and Validation of Experimental Protocol and Guidelines for Non-Invasive Superficial and Deep Muscle Electromyography in the Forearm
    (2018) Brijlal, Yasheen; Albertus, Yumna; de Jager, Kylie; Franz, Thomas
    The present study investigated a novel non-invasive superficial and deep surface electromyography (sdEMG) technique to detect and isolate extrinsic muscles of the hand with the aim of developing experimental guidelines to aid future studies. The sdEMG technique comprises of two or more surface electrode arrays encircling the limb under investigation set up in a monopolar EMG recording modality and a blind source separation (BSS) algorithm to decompose the recorded mixed monopolar EMG signals into their constituent components, which is proposed to reflect the underlying EMG activity of each muscle. Three experimental parameters linked to the finger movement protocol (MP) were investigated that varied the effects of timing, randomisation and movement anticipation on the ability of the sdEMG technique to detect and isolate the superficial muscles flexor digitorum superficialis (FDS) and extensor digitorum (ED), and the deep muscles extensor indicis (EI), extensor pollicis longus (EPL), flexor digitorum profundus (FDP) and flexor pollicis longus (FPL). FDS and FDP were split into FDS-Index, FDS-Ring, FDP-Index and FDP-Ring bands resulting in a total of eight muscles investigated. A standard movement protocol consisting of 12 dynamic movements was designed to target the activation of the investigated muscles during each experimental run. The Timing experiments varied the movement window duration to 3, 5 and 7 seconds using the standard MP sequence. The Randomisation experiment consisted of a randomised MP sequence. The Anticipation experiment presented participants with the current, and next movement instruction in the standard MP sequence. The developed sdEMG system implemented 64 custom-made surface electrodes arranged in three bands positioned around the distal third of the forearm. An OT Bioelettronica® EMG-USB2 256-channel biopotential amplifier was used, set up in a referenced monopolar EMG configuration. Contraction detection apparatus was built consisting of finger exoskeletons and flex sensors to record when finger movements occurred. A forearm testing platform was built to secure the participant’s forearm during experimental testing and a visual participant instruction system was developed to convey the timed movement instructions. Five healthy, right-hand dominant male participants (mean ± SD; age: 24 ± 3 years) without any history of neuromuscular diseases or disabilities were recruited for the study. Each participant completed five experimental runs of the five MP variations while the EMG and flex sensor data was recorded. Independent Component Analysis (ICA) was used as the BSS algorithm and the EMG recordings were decomposed into Independent Components (ICs) which were further processed with a windowed 250ms root mean square (RMS) smoothing filter as well as signal normalisation. The flex sensor data was used to generate synchronised literature-informed predicted EMG (pEMG) waveforms, representing the ideal EMG activation signals for each muscle. The muscle-specific pEMG waveforms were also processed with a 250ms RMS filter and signal normalisation before signal comparisons were made using Pearson’s correlation against all pICs. In each experimental run, the pIC with the highest calculated Pearson’s correlation coefficient (r) value for each pEMG waveform was initially selected as the representative IC (rIC) for that muscle. A rIC selection algorithm was also developed which reassigned pICs that were selected to represent multiple muscles to ensure each muscle was assigned a unique rIC. A case study was conducted to evaluate the effects of the investigated movement protocol parameters upon which experimental guidelines were formed. Fisher-corrected mean population correlation coefficients (ρ) and 95% confidence intervals were calculated to evaluate the effects of timing, randomisation and anticipation of movements. Using an amalgamated population of all the experiments and experimental runs combined, the eight muscles investigated were isolated with ρ values greater than 0.65 indicating moderate isolation (defined as 0.60 ≤ ρ < 0.80), with the exception FDS-Index Band which was poorly isolated (ρ < 0.60) with a ρ value of 0.59. The data did, however, show high variability in all experiments indicating that the sample population was too small and was possibly influenced by poor performing participants. The Timing, Randomisation and Anticipation experiments showed no discernible effects across all participants on the ability of the sdEMG technique to detect and isolate the deep and superficial forearm muscles investigated. The Anticipation experiment also showed that participant reaction delays on average increased steadily during each experimental run suggesting the anticipated visual cues were too complex and potentially confused participants. Concise experimental sdEMG guidelines were developed in which the sdEMG technique was found to be robust to variations of the three movement protocol parameters investigated.
  • Thumbnail Image
    Item
    Open Access
    The medical device development landscape in South Africa: Institutions, sectors and collaboration
    (2017) de Jager, Kylie; Chimhundu, Chipo; Saidi, Trust; Douglas, Tania S
    A characterisation of the medical device development landscape in South Africa would be beneficial for future policy developments that encourage locally developed devices to address local healthcare needs. The landscape was explored through a bibliometric analysis (2000–2013) of relevant scientific papers using co-authorship as an indicator of collaboration. Collaborating institutions thus found were divided into four sectors: academia (A); healthcare (H); industry (I); and science and support (S). A collaboration network was drawn to show the links between the institutions and analysed using network analysis metrics. Centrality measures identified seven dominant local institutions from three sectors. Group densities were used to quantify the extent of collaboration: the A sector collaborated the most extensively both within and between sectors; local collaborations were more prevalent than international collaborations. Translational collaborations (AHI, HIS or AHIS) are considered to be pivotal in fostering medical device innovation that is both relevant and likely to be commercialised. Few such collaborations were found, suggesting room for increased collaboration of these types in South Africa.