Automatic detection of the pectoral muscle in mammograms is widely used in computer-aided diagnostic (CAD) systems for breast cancer. The pectoral muscle region has some prominent features such as the upper corner position, high density, and triangular shape. But, these features may be distorted due to the masses, artifacts, skin folds, and overlapping tissues, and other reasons. Despite recent developments in CAD technology, accurate detection of distorted pectoral muscle images remains a challenging task. In this study, we proposed an automatic method that uses a divided topographic representation to detect distorted pectoral muscle boundaries. After the preprocessing stage, firstly an isocontour map is generated and then divided into horizontal blocks. The contours of the pectoral muscle boundary in the blocks often reveal specific patterns in terms of location, geometric and topological features. We developed a new segmentation algorithm, rule-based contour detection (RBCD), to detect these specific patterned isocontours. The method applied to two datasets consisting of 84 and 201 mammogram images from MIAS and Inbreast databases respectively. Besides, some distorted pectoral muscle samples selected from these datasets were used to further analyze the performance of the proposed method. The mean False-Positive and the mean False-Negative rates of the proposed method for MIAS and Inbreast datasets were 0.92%, 1.26%, and 2.34%, 1.15%, respectively. The quantitative and qualitative results for the distorted pectoral muscle samples show that the proposed method outperformed the compared methods.
Alzheimer's disease (AD), a neurological condition connected with aging, causes cognitive deterioration and has a substantial influence on a patient's daily activities. One of the most widely used clinical methods for examining how AD affects the brain is the electroencephalogram (EEG). Handcraft calculating descriptive features for machine learning algorithms requires time and frequently increases computational complexity. Deep networks provide a practical solution to feature extraction compared to handcraft feature extraction. The proposed work employs a time–frequency (TF) representation and a deep feature extraction-based approach to detect EEG segments in control subjects (CS) and AD patients. To create EEG segments' TF representations, high-resolution synchrosqueezing transform (SST) and traditional short-time Fourier transform (STFT) approaches are utilized. For deep feature extraction, SST and STFT magnitudes are used. The collected features are classified using a variety of classifiers to determine the EEG segments of CS and AD patients. In comparison to the SST method, the STFT-based deep feature extraction strategy produced improved classification accuracy between 79.56% and 92.96%. Cite this article as: O. Karabiber Cura, H. S. Ture, A. Akan, "Detection of Alzheimer's dementia by using deep time-frequency feature extraction," Electrica, 24(1), 109-118, 2024.
Biomimetic optimizers are increasingly used to solve complex engineering problems, yet their performance depends strongly on how effectively they preserve diversity while maintaining selection pressure toward promising regions. In this study, the Starfish Optimization Algorithm (SFOA) is enhanced through fitness–distance-aware selection control, leading to two improved variants: Fitness–Distance Balance Starfish Optimization Algorithm (FDBSFOA) and Dynamic Fitness–Distance Balance Starfish Optimization Algorithm (dFDBSFOA). The proposed framework guides candidate selection using both solution quality and spatial diversity relative to the current best solution, while the dynamic variant further adapts this balance over the course of the search to improve exploration in early iterations and exploitation near convergence. The proposed methods are evaluated on the IEEE CEC2017, CEC2020, and CEC2022 benchmark suites under a unified maximum function evaluation budget, MaxFEs = 10,000 × D, with 21 independent runs, and are further validated on constrained engineering design problems. Performance is assessed using convergence behavior, robustness indicators, computational overhead, and nonparametric statistical tests. The results show that the proposed variants improve the robustness and search efficiency of baseline SFOA, with dFDBSFOA providing the most consistent overall performance while introducing a controlled and interpretable computational overhead. These findings suggest that diversity-aware selection can serve as an effective design principle for strengthening biomimetic optimization frameworks. The current study focuses mainly on continuous, single-objective, and stationary benchmark problems, while the engineering-design validation also includes constrained and discrete/integer-coded cases. Extending the proposed strategy to dynamic, noisy, large-scale mixed-integer, or multi-objective settings remains future work.
Purpose: Upper extremity motor impairments following stroke substantially limit independence in daily living. Hand-tracking virtual reality (VR) rehabilitation systems may support intensive task-oriented practice while enhancing motivation; however, evidence integrating objective kinematic indicators with usability and patient-reported outcomes for controller-free consumer-grade VR remains limited. The primary aim of this single-arm pilot study was to evaluate the feasibility, safety (tolerability), and usability of a hand-tracking VR rehabilitation system delivered as an adjunct to conventional physiotherapy in individuals with stroke; describing its preliminary in-game kinematic profile and exploring participants’ experiences through open-ended feedback were secondary aims. The study was explicitly not designed or powered to test clinical efficacy. Materials and Methods: Ten individuals with stroke completed a 20-session (8-week) single-arm pilot intervention; all participants concurrently received standard hospital-based physiotherapy (median 3 sessions/week, ∼45 min/session). Early-phase (sessions 1–5) and late-phase (sessions 16–20) within-subject performance were compared using the Wilcoxon signed-rank test with Holm–Bonferroni correction across four pre-specified primary outcomes. Movement smoothness was assessed using the Spectral Arc Length (SPARC), usability was evaluated using the System Usability Scale (SUS), and cybersickness was monitored with the Simulator Sickness Questionnaire (SSQ). Results: Sixteen patients were screened, of whom fourteen started the intervention and ten completed the 8-week per-protocol program (intervention completion rate, 10/14 = 71.4%; per-protocol session adherence among the ten completers, 200/200 = 100%; no SSQ-defined adverse events). Within-subject comparisons showed a -point increase in success rate (adjusted ), a m/s increase in mean movement speed (adjusted ), a ms reduction in pause duration (adjusted ), and a -point Hodges–Lehmann median-difference increase in SS-QOL (95% CI –; participant-level median ; adjusted ). The mean SUS score was , indicating “good” usability. Thematic analysis of post-intervention open-ended feedback identified three themes—motivation and engagement, the value of feedback, and design and comfort suggestions—that converged with the high adherence and good usability ratings. Correlations between VR-derived kinematic change and clinical change were non-significant trends (). Conclusions: A controller-free hand-tracking VR system was found to be feasible, well tolerated, and rated as having good usability when delivered as an adjunct to conventional therapy. Because the study used a single-arm design, included only ten participants, and did not control for the confounding effect of concurrent conventional physiotherapy or natural recovery, the observed within-subject changes cannot be causally attributed to the VR intervention and should be interpreted as exploratory feasibility signals. Adequately powered randomized controlled trials with stroke-specific clinical scales (e.g., FMA-UE, ARAT) are required before clinical efficacy can be claimed.
Virtual reality (VR)-based rehabilitation is an established modality for upper extremity motor recovery; however, existing systems frequently rely on fixed, random, or therapist-tuned target placement that disregards patient-specific motor capacity and population-level priors. This study proposes a cross-patient collaborative swarm intelligence framework that derives zone-based patient profiles from real VR trajectories and augments them with a similarity-weighted cohort prior distilled from clinically similar patients’ successful trajectory clouds and zone-transition graphs. A hybrid Ant Colony Optimization (ACO)–Particle Swarm Optimization (PSO) algorithm optimizes 12 targets per session across a 27-zone () workspace using a five-component fitness function encompassing reachability, zone balance, movement efficiency, heatmap-guided challenge coverage, and swarm-flow consistency. The framework was evaluated retrospectively on a single-center cohort of 36 post-stroke patients and 6373 sessions under a leakage-safe simulation protocol with 70/30 chronological splits; outcomes are model-based proxy success rates derived from each patient’s profile rather than directly observed task success. The hybrid strategy achieved a mean simulated success rate of 85.5% ± 5.5%, a 36.4% relative improvement over random placement (Wilcoxon , Cohen’s ); the leakage-safe split yielded 80.1% on the held-out segment versus 61.1% for random, with no statistically significant train–test gap (). Ablation confirmed both PSO and ACO are individually necessary ( pp, ). Total session-start computation is 78 ms on standard CPU hardware. These findings constitute a proof-of-concept that collaborative personalized swarm optimization can substantially outperform heuristic target placement under in silico evaluation; clinical efficacy in terms of standardized motor outcome measures remains to be established in a prospective randomized controlled trial, and the findings should be replicated across centers, task modes, and a larger cohort before generalization.