Search Results (55)

Background and Objectives: In degenerative spine disease, compensatory mechanisms are activated to maintain upright posture, extending beyond the spine to involve the pelvis, lower limbs, and feet. These adaptations may be accompanied by differences in foot alignment, which could be associated with sagittal balance. The aim of this study is to investigate the relationship between foot alignment and spinal posture in patients with degenerative spine disease and evaluate whether foot deformities are associated with sagittal imbalance in degenerative spine disease. Materials and Methods: We retrospectively reviewed 98 patients with degenerative spine disease who underwent preoperative standing EOS imaging between 2017 and 2025 at a single academic spine centre. Meary’s angle, talocalcaneal angle, and calcaneal pitch were measured on lateral EOS images to classify feet as flat, normal, or cavus. Spinopelvic parameters were extracted from EOS and conventional radiographs. Differences in spinal parameters across foot groups were compared using ANOVA, and linear regression evaluated associations between sagittal vertical axis (SVA) and foot angles. Results: Among spinopelvic parameters, only SVA significantly differed between foot groups, with flatfoot patients showing greater forward imbalance (p = 0.035). Regression analysis demonstrated an inverse relationship between SVA and both talocalcaneal angle (p = 0.003) and calcaneal pitch (p = 0.034), suggesting that greater forward trunk inclination was associated with flatter feet. Degenerative scoliosis patients demonstrated a bimodal pattern with more flat and cavus feet (p = 0.006), while herniated disc patients more often exhibited flatfoot (p = 0.031). Conclusions: Foot posture abnormalities, particularly flatfoot, are associated with sagittal spinal imbalance, suggesting foot posture may be associated with global alignment and could reflect distal postural adaptations. Full article

Background and Objectives: Although the anterior mandible is generally considered a safe region for implant placement, injury to the medial lingual foramen (MLF) may result in significant vascular complications. Accurate identification of this structure is challenging due to its small size, low volumetric representation, and anatomical variability. This study aimed to evaluate the anatomical characteristics of the MLF using cone-beam computed tomography (CBCT) and to develop and validate a deep learning-based approach for its automated detection and segmentation. Materials and Methods: A total of 106 CBCT scans were retrospectively analyzed to assess the morphology and position of the MLF. Manual pixel-wise annotations of the complete canal trajectory were performed on sagittal slices and used to train convolutional neural network models based on a U-Net-derived framework. Multiple configurations, including multi-class, binary, two-dimensional, and three-dimensional approaches, were evaluated. Given the extremely limited volumetric representation of the MLF, severe class imbalance represented a major challenge during model training and evaluation. Model performance was assessed using the Dice similarity coefficient, precision, recall, and Hausdorff distance. External validation was performed on an independent dataset of 10 CBCT scans. Results: The MLF was identified in all patients, with a single canal observed in 63% of cases. The sagittal-plane binary segmentation model achieved the best performance, with a test Dice score of 0.79, precision of 0.88, and recall of 0.73. External validation demonstrated a Dice score of 0.81, precision of 0.89, and recall of 0.71. The 95th percentile Hausdorff distance was 2.6 mm, and the mean center-point localization error was 1.2 mm. The model correctly detected the MLF in 90% of external cases. Conclusions: Deep learning-based segmentation of the MLF is feasible and may support automated localization assistance during preoperative CBCT assessment. Performance was influenced by the alignment between the annotation strategy and model input, highlighting an important consideration for small-structure segmentation. Further validation on larger multicenter datasets is required before clinical implementation can be considered. Full article

Background/Objectives: Scoliosis is a three-dimensional deformity involving coronal curvature, axial rotation, and sagittal imbalance, which may alter thoracic geometry and reduce ventilatory capacity. Traditional two-dimensional radiographic measures incompletely represent the complexity of axial rotation, and the apical vertebra is not always the most rotated vertebra. We aimed to determine whether computed tomography (CT)-derived three-dimensional vertebral rotation indices, particularly global rotational asymmetry between main and compensatory curves, are associated with pulmonary function impairment in a large heterogeneous scoliosis cohort. Methods: We retrospectively reviewed 250 patients with scoliosis who underwent full-spine CT and spirometry within a 1-year interval (2013–2023). Vertebral rotation was measured using the Aaro–Dahlborn method. Rotation indices included apical rotation (R(Apex)), averaged apical rotation across the apical vertebra and adjacent levels (R(Avg)), and rotational asymmetry defined as the absolute difference between rotation in the main and compensatory curves (ΔR(M–C)). Pulmonary function outcomes were FVC (L), FEV1 (L), FVC% and FEV1%. Pearson correlation and multivariate linear regression, adjusted for age, sex, height, and weight, were performed; sensitivity analyses, additionally adjusted for upright Cobb angle and thoracic kyphosis (TK) to evaluate whether ΔR(M–C) provided independent explanatory information, and subgroup analyses by etiology were performed. Results: The cohort had a mean age of 15.6 ± 5.7 years; 49.6% were female. All rotation indices showed significant negative correlations with pulmonary function in the overall cohort. ΔR(M–C) showed the strongest associations with FVC% (r = −0.66) and FEV1% (r = −0.64), as well as with absolute volumes (FVC, r = −0.59; FEV1, r = −0.58). In adjusted multivariate analyses, models incorporating ΔR(M–C) consistently demonstrated the highest explanatory performance compared with models based on R(Apex) or R(Avg). Subgroup analysis revealed the strongest associations in neurofibromatosis-related scoliosis (r = −0.87) and congenital scoliosis (r = −0.71). Associations were attenuated in neuromuscular subtypes and did not reach statistical significance in SMA. In sensitivity analyses adjusting for Cobb angle and thoracic kyphosis, ΔR(M–C) retained a robust independent association with FVC% (unstandardized B = −0.82 percentage points per 1°, 95% CI −0.98 to −0.66; p < 0.001; partial F = 103, p < 0.001), while neither Cobb angle nor TK remained statistically significant after adjustment for ΔR(M–C); comparable patterns were observed across FEV1%, FVC, and FEV1. Conclusions: CT-derived global rotational asymmetry between the main and compensatory curves is strongly associated with pulmonary function impairment in scoliosis and demonstrates superior explanatory performance to single-level rotation indices and retains independent explanatory value after adjustment for conventional 2D radiographic parameters (Cobb angle and thoracic kyphosis). Incorporating a CT-derived metric may complement traditional two-dimensional assessments for functional risk stratification. Full article

Background/Objectives: Spinal alignment is a key determinant of biomechanical function and postural stability, particularly during periods of growth and development. Three-dimensional (3D) surface topography offers a non-invasive method for assessing spinal posture. This study aimed to evaluate spinal alignment parameters in a mixed adolescent and adult population, to investigate sex-related differences, and to analyze biomechanical relationships between spinal components. Methods: A total of 98 participants (aged 11–45 years) underwent 3D spinal surface topography assessment. Descriptive statistics were calculated for sagittal, coronal, and rotational parameters. Group comparisons between sexes were performed using independent samples t-tests. Pearson correlation analysis and linear regression were used to assess the relationships between spinal parameters. Logistic regression analysis was conducted to identify predictors of clinically relevant rotational asymmetry (surface rotation RMS > 6°). Results: Most participants exhibited near-physiological sagittal alignment, with thoracic kyphosis and lumbar lordosis within normal ranges. However, approximately 20% demonstrated clinically relevant rotational asymmetry. Female participants showed significantly higher rotational asymmetry compared to males (p = 0.008), while sagittal parameters did not differ significantly. Strong correlations were observed between thoracic kyphosis and cervical sagittal displacement (r = 0.77). Rotational asymmetry was negatively correlated with sagittal parameters and significantly predicted coronal imbalance (β = 0.38, p < 0.01; R² = 0.21). Conclusions: 3D surface topography provides a non-invasive method for assessing external postural alignment and surface-based asymmetries. Rotational asymmetry appears to represent a relevant component of spinal imbalance and is associated with coronal deviation within a multi-planar framework. These findings support the use of integrated biomechanical assessment in the evaluation of spinal alignment. Full article

Background: Core exercises are widely prescribed, yet their immediate, exercise-specific mechanical effects on spinal alignment remain unclear because most evidence derives from multi-exercise programs. Non-invasive tools capable of detecting acute postural changes could support a better characterization of exercise-specific acute postural responses. This study evaluated whether common core exercises determine distinct acute changes in sagittal and coronal alignment. Methods: Eighty healthy young adults without evident postural abnormalities (23 ± 5.6 years) were block-randomized to crunch (CE), plank (PE), Russian Twist (RE), or control (CG). Rasterstereography recorded spinal postural parameters immediately before and after each condition. Primary outcomes were pre–post change scores (Δ = post − pre) analyzed with linear models and Holm-adjusted planned contrasts versus CG. Results: Versus CG, CE showed greater reductions in sagittal imbalance (Δ −6.75 ± 9.88; adjusted p = 0.034), cervical arrow (Δ −6.15 ± 7.85; adjusted p = 0.011), and kyphosis angle (adjusted p < 0.001). PE differed from CG only for coronal imbalance (Δ −4.25 ± 3.81; adjusted p = 0.022), showing a shift toward more negative values. RE differed from CG only for kyphosis angle (adjusted p = 0.036). Lumbar arrow did not show between-group differences. Conclusions: A single bout of core exercise can transiently modulate selected sagittal and coronal alignment parameters in healthy young adults without evident postural abnormalities, with effects depending on exercise modality. These findings should be considered exploratory, and their clinical significance remains uncertain. Full article

Background: Posterior spinal fusion is the mainstay of treatment for Cobb angle over 50 degrees with satisfactory long-term results. In the surgical management of scoliosis, surgeons usually focus on the amount of coronal curvature correction because it can determine the surgical outcome. Nevertheless, there are many factors that contribute to patients’ satisfaction after surgery, and achieving shoulder balance is one of the most vital factors of a successful surgery. Our objective is to study the differences in managing idiopathic scoliosis with pedicle screws versus hybrid fixation with regard to shoulder imbalance postoperatively. Methods: Continuous variables were described using mean and standard deviation, whereas categorical variables were described using frequencies. The association between predictor independent variables with the analyzed outcomes were expressed as (beta coefficients) with their associated 95% confidence intervals. The Alpha significance level was considered at 0.050 level. Results: The mean angle of the clavicle measured a significant drop post-surgery compared to their pre-surgical measured mean clavicular angle, p-value < 0.001, and so did the coracoid height difference: p-value < 0.001. Furthermore, the participants had measured a significantly lower mean angle of the clavicle compared to their baseline; p-value = 0.029, regardless of their surgery type. The participants mean measured coracoid height difference score had correlated positively with their mean angle of the clavicle: beta coefficient = 1.654, p-value < 0.001; when the coracoid height difference increased, so did the mean angle of the clavicle. Conclusions: Posterior spinal fusion is effective in correcting coronal curvature and improving radiographic shoulder asymmetry in AIS. Significant improvements were observed in Cobb angle, clavicle angle, and coracoid height difference, with pedicle screw constructs providing superior curve correction. These findings reinforce the value of individualized surgical planning that considers coronal, sagittal, and cosmetic alignment goals. Full article

Objectives: In this retrospective comparative cohort study, we aimed to compare surgical efficiency, radiographic facet integrity, and postoperative alignment and mobility between unilateral hemilaminectomy (UL) and laminotomy with laminoplasty (LP) for cervical intradural extramedullary (IDEM) tumors. Methods: Thirty-eight patients (UL: 20; LP: 18) were retrospectively reviewed. Operative variables, tumor characteristics, extent of resection, radiographic facet joint violation (graded 1–4), and sagittal alignment parameters, including global and segmental range of motion (ROM), were evaluated at 1 year postoperatively. Propensity score matching was additionally performed to minimize potential baseline imbalance between groups. Results: The UL group had significantly shorter operative time (178.05 ± 61.89 vs. 276.06 ± 121.76 min, p = 0.003) and lower intraoperative blood loss (p < 0.001) than the LP group. Radiographic facet joint violation (Grade ≥ 2) occurred more frequently in the UL group (25.0% vs. 0%, p = 0.048) but was not associated with postoperative sagittal alignment changes or radiographic instability. Global cervical alignment remained in both groups, but the LP group showed a significantly greater reduction in segmental ROM at 1 year (−6.42 ± 8.29° vs. 0.06 ± 7.72°, p = 0.017). These findings were consistent in the propensity score–matched cohort. Conclusions: UL provides favorable operative efficiency and better preservation of segmental cervical mobility than LP, while maintaining comparable clinical and radiographic outcomes. Although radiographic facet joint violation was more frequent in the UL group, postoperative spinal stability was not compromised in this cohort. UL may serve as a safe and motion-preserving alternative in selected patients with cervical IDEM tumors. Full article

Background/Objectives: Transforaminal lumbar interbody fusion (TLIF) and lateral lumbar interbody fusion (LLIF) are widely utilized techniques for degenerative lumbar diseases. This study compared radiological and clinical outcomes of LLIF and TLIF in patients undergoing lumbar fusion. Methods: This non-randomized prospective observational study enrolled 117 patients (LLIF: n = 17; TLIF: n = 100), with an inherent imbalance in group sizes, who underwent one- or two-level lumbar interbody fusion. Primary outcome was segmental lordotic angle (SLA) at the operated level. Secondary outcomes included disc height, lumbar lordotic angle, sagittal vertical axis, and patient-reported outcomes. Assessments were conducted at baseline, 3, 6, 12, and 24 months. Linear mixed models analyzed longitudinal data. Results: Mean SLA improvement was not significantly different between the groups (LLIF: 3.04° vs. TLIF: 3.18°, p = 0.782). No significant differences were observed for disc height (p = 0.518), lumbar lordotic angle (p = 0.718), or sagittal vertical axis (p = 0.866). Patient-reported outcomes improved significantly in both groups. Linear mixed model analysis revealed no significant between-group effects for Oswestry Disability Index (p = 0.335) or low back pain (p = 0.069). TLIF showed higher rates of dural tears and wound complications, while LLIF had more sympathetic chain injuries and transient psoas weakness. Overall complication rates were comparable (p > 0.05). Conclusions: TLIF and LLIF demonstrate comparable radiographic and clinical outcomes at 24-month follow-up. Surgical technique selection should be individualized based on patient-specific anatomical and clinical factors, considering distinct approach-specific complication patterns. Full article

Background: Total knee arthroplasty (TKA) effectively relieves pain in end-stage osteoarthritis, yet a proportion of patients remain dissatisfied despite advances in surgical technique. Medial-congruent (MC) bearings are designed to recreate native medial-pivot kinematics, which depend on appropriate medial compartment soft tissue tension. Robotic-assisted TKA (RA-TKA) has been shown to improve the accuracy and soft tissue balance. However, evidence of its additional benefits in MC TKA remains limited. Methods: We retrospectively identified consecutive primary TKAs with the same MC bearing performed between April 2022 and June 2024 at a tertiary center. After performing 1:1 propensity score matching to reduce baseline imbalance, 36 patients who received RA-TKA and 36 who underwent manual TKA (M-TKA) were included. Primary outcomes were evaluated with the 12-month Oxford Knee Score (OKS) and KOOS-JR. Secondary outcomes included radiographic alignment parameters, outlier rates, operative time, liner thickness, and hospital stay. Results: Baseline characteristics and liner thickness were comparable, and operative time was longer in the RA-TKA group than in the M-TKA group. Both RA-TKA and M-TKA produced significant 12-month improvements in OKS and KOOS-JR with no difference in mean scores. RA-TKA had fewer posterior tibial slope outliers (mean slope 4.3° ± 1.8 vs. 5.9° ± 3.1; outlier rate 16.7% vs. 41.7%; p = 0.02), whereas coronal alignment parameters did not differ between groups. Conclusions: RA-TKA with MC bearing provides functional outcomes comparable to M-TKA and may decrease sagittal alignment variability; long-term follow-up studies are needed to determine whether this potential benefit translates into sustained functional gains or improved implant survivorship. Full article

Background: Adult scoliosis following spinal fusion and subsequent hardware removal presents complex challenges, especially when deformity progresses in both the sagittal and coronal planes. Revision surgery is often recommended but it entails considerable risks. Conservative approaches, including Schroth-based physiotherapy and bracing, offer promising alternatives for select patients, particularly older adults with comorbidities or those who decline further surgery. Case Presentation: A woman with adolescent idiopathic scoliosis (diagnosed at age 13) underwent spinal fusion with Harrington rod instrumentation and costoplasty at age 24. She was de-instrumented two years later due to infection and developed progressive thoracic hyperkyphosis, coronal imbalance, and 12 cm loss of height over two decades. At age 47, she chose conservative management instead of revision surgery. Methods: She completed a 28-month program combining Schroth-based physiotherapy supervised by a certified therapist and part-time bracing. Outcomes included radiographs, inclinometry, spirometry, spinal-muscle ultrasound, height/posture measures, and SRS-22 and ODI scores. Results: Substantial clinical improvements were observed. Thoracic kyphosis decreased from 80° to 45° (44%) and the coronal thoracic curve was reduced from 48° to 32° (33%). Lumbar lordosis increased by 6°, standing height rose by 4 cm, and forced vital capacity improved by 900 mL (33%). The Oswestry Disability Index (ODI) score improved from 4% to 0%, and SRS-22 domains (pain, function, self-image, mental health, satisfaction) showed meaningful gains. The patient returned to full physical activity and avoided further surgery. Conclusion: This case highlights the potential of Schroth-based physiotherapy with bracing to reverse postural and functional decline in complex post-fusion scoliosis. It offers a viable non-surgical option when revision surgery poses a high risk or is declined and aligns with modern rehabilitative goals focused on long-term quality of life. Full article

Background: This study investigated how static postural parameters influence dynamic spinopelvic balance across varying walking speeds. One hundred healthy young adults (aged 18–25) underwent rasterstereographic assessment (DIERS 4Dmotion^®) to quantify static global alignment metrics including craniovertebral angle (CVA), Q-angle, sagittal and coronal imbalance, pelvic rotation, torsion, obliquity, vertebral rotation, thoracic kyphosis, lumbar lordosis, and pelvic tilt, followed by dynamic spinopelvic analysis during treadmill walking at 1, 2, 4, and 5 km/h. Methods: Multiple linear regression models were used to determine the predictive value of static postural measures for dynamic outcomes at each speed. At slower walking speeds (1–2 km/h), static alignment variables significantly predicted dynamic spinopelvic parameters (adjusted R² = 0.53–0.73; RMSE = 0.59–0.81), with CVA, sagittal imbalance, and pelvic torsion emerging as the most consistent predictors. Results: At higher speeds (4–5 km/h), predictive strength declined substantially (adjusted R² = 0.04–0.34), indicating a shift from posture-driven to neuromuscular-governed gait control. The Q-angle showed limited and inconsistent predictive value across all conditions. Conclusions: Overall, static postural alignment, particularly CVA, sagittal imbalance, and pelvic torsion, serves as a moderate predictor of spinopelvic dynamics at slow to moderate gait speeds but loses explanatory power as velocity increases, emphasizing the growing role of neuromuscular control in maintaining dynamic balance. These findings highlight the clinical relevance of integrating both static and dynamic assessments to comprehensively evaluate postural and locomotor function. Full article

Background/Objectives: Three-dimensional (3D) spinal and pelvic alignment plays a critical role in maintaining anticipatory postural control. However, the extent to which specific multiplanar alignment parameters influence feedforward activation of trunk stabilizing muscles remains unclear. This study aimed to determine whether sagittal, coronal, and transverse postural deviations predict anticipatory muscle activation patterns during externally induced perturbations. Methods: Surface electromyography (EMG) was recorded from bilateral external oblique (EO), lumbar multifidus (LM), and transversus abdominis/internal oblique (TrA/IO) muscles in 100 asymptomatic young adults (18–25 years) performing dynamic right-leg raises. Spinal and pelvic alignment was quantified using rasterstereography, including sagittal and coronal imbalance, pelvic tilt, torsion, rotation, vertebral rotation, and spinal curvatures (kyphotic and lordotic angles). Regression models examined how these parameters predicted EMG onset latency and activation amplitude. Results: Distinct alignment patterns were associated with altered anticipatory control. Increased vertebral rotation and greater sagittal imbalance were linked to delayed activation of EO and LM, while asymmetries in pelvic torsion and tilt were related to less efficient TrA/IO recruitment. Conversely, more balanced spinal curvatures corresponded with earlier, more coordinated muscle activation across the trunk. Conclusions: Multiplanar spinal and pelvic alignment significantly influences anticipatory neuromuscular strategies. Identifying how specific postural deviations disrupt feedforward activation provides a functional basis for targeted rehabilitation programs aiming to restore alignment, enhance trunk stability, and prevent recurrent postural dysfunction. Full article

Background: Alzheimer’s disease (AD) is a leading worldwide cause of cognitive impairment, necessitating accurate, inexpensive diagnostic tools to enable early recognition. Methods: In this study, we present a robust deep learning approach for AD classification based on structural MRI scans, ConvNeXt, an emergent convolutional architecture inspired by vision transformers. We introduce AlzaSet, a clinically curated T1-weighted MRI dataset of 79 subjects (63 with Alzheimer’s disease [AD], 16 cognitively normal controls [NC]) acquired on a 1.5 T Siemens Aera in axial, coronal, and sagittal planes, respectively (12,947 slices in total). Images are neuroradiologist-labeled. Results are reported per plane, with awareness of the class imbalance at the subject level. We further present AlzaSet, a novel, expertly labeled clinical dataset with axial, coronal, and sagittal perspectives from AD and cognitively normal control subjects. Three ConvNeXt sizes (Tiny, Small, Base) were compared and benchmarked against existing state-of-the-art CNN models (VGG16, VGG19, InceptionV3, DenseNet121). Results: ConvNeXt-Base consistently outperformed the other models on coronal slices with an accuracy of 98.37% and an AUC of 0.992. Coronal views were determined to be most diagnostically informative, with emphasis on visualization of the medial temporal lobe. Moreover, comparison with recent ensemble-based techniques showed superior performance with comparable computational efficiency. Conclusions: These results indicate that ConvNeXt-capable models applied to clinically curated datasets have strong potential to provide scalable, real-time AD screening in diverse settings, including both high-resource and resource-constrained settings. Full article

Background: Postoperative sagittal imbalance, including hyperextension and flexion contracture after total knee arthroplasty (TKA), adversely affects long-term outcomes. Conventional techniques depend on surgeons’ visual estimation, which may vary with patient anatomy. This study compared the accuracy of sagittal alignment assessment between a robot-assisted system and surgeons’ visual evaluation and analyzed the influence of body mass index (BMI) and anatomic factors on concordance. Methods: Sixty unilateral TKAs performed between October 2023 and May 2024 using the MAKO system were retrospectively reviewed. Sagittal mechanical axis angles were obtained from the robotic system (M group) and visually estimated by two blinded orthopedic surgeons from intraoperative lateral photographs (S group) for 9, 10, and 11 mm trial inserts. Final alignment was verified by C-arm radiographs. Inter-observer reliability was analyzed using Cohen’s κ and ICC, and correlations with BMI, thigh circumference, and limb proportions were assessed. Results: In patients with overweight/obesity, the S group significantly underestimated sagittal alignment (p < 0.001), whereas the M group maintained consistent accuracy regardless of BMI. With thinner inserts (9 mm), hyperextension was often overestimated (mismatch 55%, p < 0.0001), and with thicker inserts (11 mm), flexion contracture was underestimated (mismatch 46.7%, p = 0.001). Inter-observer reliability was good (κ = 0.717, ICC = 0.816). Conclusions: Visual assessment may underestimate sagittal alignment, especially in patients with obesity or those with abundant soft tissue. The MAKO robotic system provided consistent, objective alignment evaluation regardless of body habitus. Robotic-assisted quantitative assessment enables more accurate and reproducible sagittal alignment, supporting safer and more reliable TKA outcomes. Full article

We propose a novel hybrid deep learning framework that synergistically integrates Convolutional Neural Networks (CNNs), Spiking Neural Networks (SNNs), and Generative Adversarial Networks (GANs) for robust and accurate classification of high-resolution frontal and sagittal human gait video sequences—capturing both lower-limb kinematics and upper-body posture—from subjects with Knee Osteoarthritis (KOA), Parkinson’s Disease (PD), and healthy Normal (NM) controls, classified into three disease-type categories. Our approach first employs a tailored CNN backbone to extract rich spatial features from fixed-length clips (e.g., 16 frames resized to 128 × 128 px), which are then temporally encoded and processed by an SNN layer to capture dynamic gait patterns. To address class imbalance and enhance generalization, a conditional GAN augments rare severity classes with realistic synthetic gait sequences. Evaluated on the controlled, marker-based KOA-PD-NM laboratory public dataset, our model achieves an overall accuracy of 99.47%, a sensitivity of 98.4%, a specificity of 99.0%, and an F1-score of 98.6%, outperforming baseline CNN, SNN, and CNN–SNN configurations by over 2.5% in accuracy and 3.1% in F1-score. Ablation studies confirm that GAN-based augmentation yields a 1.9% accuracy gain, while the SNN layer provides critical temporal robustness. Our findings demonstrate that this CNN–SNN–GAN paradigm offers a powerful, computationally efficient solution for high-precision, gait-based disease classification, achieving a 48.4% reduction in FLOPs (1.82 GFLOPs to 0.94 GFLOPs) and 9.2% lower average power consumption (68.4 W to 62.1 W) on Kaggle P100 GPU compared to CNN-only baselines. The hybrid model demonstrates significant potential for energy savings on neuromorphic hardware, with an estimated 13.2% reduction in energy per inference based on FLOP-based analysis, positioning it favorably for deployment in resource-constrained clinical environments and edge computing scenarios. Full article