Fnirs-based analysis of motor and cognitivemovements in lower limb amputee and replantation

In this thesis study, cortical activation patterns during motor execution and motor imagery tasks were evaluated using functional near-infrared spectroscopy (fNIRS) in individuals who had undergone lower limb amputation and replantation, in comparison with healthy individuals. The study included a total of 15 healthy participants (10 males, 5 females) with a mean age of 32.33 ± 6.81 years and right-side dominance, along with 9 patients who had undergone lower limb amputation and replantation at various levels. Participants performed a total of 40 different motor and motor imagery tasks involving the toe, ankle, knee, and hip joints. First, a channel configuration suitable for analyzing lower limb-specific motor and motor imagery tasks was established. In the system consisting of 8 sources and 8 detectors, a total of 21 optical channels were defined, and fNIRS data were recorded. Channel placements were optimized to target the medial surfaces of the motor cortex representing the lower limb and the primary somatosensory areas. During data processing, instrumental, physiological, and motion-related artifacts were eliminated through appropriate filtering and correction methods in the preprocessing stage. The channel planning prioritized regions corresponding to the M1-leg area, premotor cortex, primary somatosensory cortex, and somatosensory association cortex. In this way, a neuroanatomically targeted configuration enabling detailed examination of motor activity in both the left and right hemispheres was achieved. This study particularly focused on the roles of the premotor and supplementary motor areas, primary motor cortex, primary somatosensory cortex, and association cortices in the reorganization process. The 40 motor and motor imagery tasks (10 right motor, 10 left motor, 10 right imagery, 10 left imagery) covered the toe, ankle, knee, and hip joints. Initially, the motor tasks were performed, followed by motor imagery tasks conducted with eyes closed under auditory instruction. All tasks were designed in a block paradigm consisting of 5 seconds of rest, 8 seconds of activity, and 8 seconds of rest, repeated four times. Before analyzing the fNIRS data, a comprehensive preprocessing pipeline was applied to eliminate noise and artifacts from various sources that could affect the signals. A low-pass filter was applied to remove device-specific high-frequency instrumental noise, and a band-pass filter in the range of 0.01–0.20 Hz was used to suppress physiological artifacts (such as heartbeat, respiration, and Mayer waves). To correct for motion artifacts such as head movements, a hybrid approach was adopted that combined wavelet-based filtering, spline interpolation, and Savitzky-Golay smoothing. While the wavelet-based method effectively removed abrupt and sharp signal distortions, spline interpolation corrected low-frequency drifts, and the Savitzky-Golay filter was effective in smoothing high-frequency noise. Analyses were conducted based on statistical tests and contrast maps, allowing for the identification of significant group-level differences and their spatial visualization at the cortical level. The findings of the study revealed that, following lower limb amputation, cortical activity shifted from the primary motor cortex to the somatosensory cortex, indicating a process of neuroplastic reorganization. Moreover, in individuals who had undergone replantation, cortical activation patterns differed from those of healthy individuals, with evidence of reshaped cortical representations. Contrast maps provided a clear spatial depiction of these differences beyond statistical results, enabling a more detailed tracking of reorganization. During motor imagery tasks, both amputee and replanted individuals exhibited a broader and more widespread cortical activation profile compared to healthy participants. Increased activity particularly in the somatosensory areas suggested that sensory-motor integration processes were also restructured within the scope of neuroplasticity. Channel-based time series data analyses also suggested reorganization in terms of temporal epochs in amputee individuals. In conclusion, this thesis demonstrates that brain organization undergoes both structural and functional reconfiguration following lower limb amputation and replantation. These processes can be reliably monitored using the fNIRS method, and the obtained data have the potential to guide the development of individualized rehabilitation approaches. Furthermore, the study emphasizes the sensitivity of fNIRS in detecting motor tasks specific to the lower limb, offering significant contributions to future brain-computer interface applications and neurotechnological developments.