Searcharxiv⌕ Search

arXiv subjects

Mohammad Mashayekhi

Publications and source records attributed to Mohammad Mashayekhi.

4 recordsLinked to original sources

Deep Learning-Assisted Optimization of Large Domino-Tiled Phased Arrays for Dual-Band LEO Satellite Communications

Irregular domino tiling reduces the number of transmit/receive modules (TRMs) of large phased arrays, but its design requires repeated pattern evaluations whose cost grows rapidly with aperture size. This paper presents a hybrid deep learning and optimization framework for large domino-tiled arrays. A local-flip Markov chain generator, about $5.2\times$ faster than a randomized greedy baseline, produces balanced tilings with nearly equal numbers of horizontal and vertical dominoes, and a Tabu Search engine reaches a sidelobe level (SLL) about $0.8$~dB lower than a genetic algorithm, simulated annealing, and iterated local search at comparable directivity. These tools build the dataset on which a convolutional neural network (CNN) surrogate is trained to predict, from the tiling structure alone, whether a candidate yields a favorable multi-angle SLL. Gating the Tabu Search with the surrogate removes $60.6$--$70.6\%$ of the full evaluations and speeds up the search by $2.56$--$3.80$ times for apertures from $20\times20$ to $100\times100$ elements, without degrading the solution quality. A convex amplitude--phase optimization of the fixed tilings then gives equal or higher directivity than two state-of-the-art methods at a comparable SLL, with more than an order of magnitude shorter computation time. Finally, the framework is applied to a $36\times36$ dual-band circularly polarized array for low Earth orbit (LEO) terminals at $11.2$ and $14.0$~GHz, whose bands are a factor of $1.25$ apart, using $648$ dominoes and a common element spacing selected from a sweep. The array scans to $50^\circ$ with no grating lobe and, compared with a reference irregular array, gives higher directivity over most of the scan range, a lower SLL at large scan angles, and a spacing that is more practical for the lower band.

eess.SP↗

AI-Powered Inverse Design of Ku-Band SIW Resonant Structures by Iterative Residual Correction Network

Designing high-performance substrate-integrated waveguide (SIW) filters with both closely spaced and widely separated resonances is challenging. Consequently, there is a growing need for robust methods that reduce reliance on time-consuming electromagnetic (EM) simulations. In this study, a deep learning-based framework was developed and validated for the inverse design of multi-mode SIW filters with both closely spaced and widely separated resonances. A series of SIW filters were designed, fabricated, and experimentally evaluated. A three-stage deep learning framework was implemented, consisting of a Feedforward Inverse Model (FIM), a Hybrid Inverse-Forward Residual Refinement Network (HiFR\textsuperscript{2}-Net), and an Iterative Residual Correction Network (IRC-Net). The design methodology and performance of each model were systematically analyzed. Notably, IRC-Net outperformed both FIM and HiFR\textsuperscript{2}-Net, achieving systematic error reduction over five correction iterations. Experimental results showed a reduction in mean squared error (MSE) from 0.00191 to 0.00146 and mean absolute error (MAE) from 0.0262 to 0.0209, indicating improved accuracy and convergence. The proposed framework demonstrates the capability to enable robust, accurate, and generalizable inverse design of complex microwave filters with minimal simulation cost. This approach is expected to facilitate rapid prototyping of advanced filter designs and could extend to other high-frequency components in microwave and millimeter-wave technologies.

cs.LG↗

Clinical Semantic Intelligence (CSI): Emulating the Cognitive Framework of the Expert Clinician for Comprehensive Oral Disease Diagnosis

The diagnosis of oral diseases presents a problematic clinical challenge, characterized by a wide spectrum of pathologies with overlapping symptomatology. To address this, we developed Clinical Semantic Intelligence (CSI), a novel artificial intelligence framework that diagnoses 118 different oral diseases by computationally modeling the cognitive processes of an expert clinician. Our core hypothesis is that moving beyond simple pattern matching to emulate expert reasoning is critical to building clinically useful diagnostic aids. CSI's architecture integrates a fine-tuned multimodal CLIP model with a specialized ChatGLM-6B language model. This system executes a Hierarchical Diagnostic Reasoning Tree (HDRT), a structured framework that distills the systematic, multi-step logic of differential diagnosis. The framework operates in two modes: a Fast Mode for rapid screening and a Standard Mode that leverages the full HDRT for an interactive and in-depth diagnostic workup. To train and validate our system, we curated a primary dataset of 4,310 images, supplemented by an external hold-out set of 176 images for final validation. A clinically-informed augmentation strategy expanded our training data to over 30,000 image-text pairs. On a 431-image internal test set, CSI's Fast Mode achieved an accuracy of 73.4%, which increased to 89.5% with the HDRT-driven Standard Mode. The performance gain is directly attributable to the hierarchical reasoning process. Herein, we detail the architectural philosophy, development, and rigorous evaluation of the CSI framework.

cs.AI↗

Radious: Unveiling the Enigma of Dental Radiology with BEIT Adaptor and Mask2Former in Semantic Segmentation

X-ray images are the first steps for diagnosing and further treating dental problems. So, early diagnosis prevents the development and increase of oral and dental diseases. In this paper, we developed a semantic segmentation algorithm based on BEIT adaptor and Mask2Former to detect and identify teeth, roots, and multiple dental diseases and abnormalities such as pulp chamber, restoration, endodontics, crown, decay, pin, composite, bridge, pulpitis, orthodontics, radicular cyst, periapical cyst, cyst, implant, and bone graft material in panoramic, periapical, and bitewing X-ray images. We compared the result of our algorithm to two state-of-the-art algorithms in image segmentation named: Deeplabv3 and Segformer on our own data set. We discovered that Radious outperformed those algorithms by increasing the mIoU scores by 9% and 33% in Deeplabv3+ and Segformer, respectively.

cs.CV↗