arXiv subjects

Mohammad Eslami

Publications and source records attributed to Mohammad Eslami.

15 recordsLinked to original sources

arXiv2026

Cross-Layer Co-Optimized LSTM Accelerator for Real-Time Gait Analysis

Long Short-Term Memory (LSTM) neural networks have penetrated healthcare applications where real-time requirements and edge computing capabilities are essential. Gait analysis that detects abnormal steps to prevent patients from falling is a prominent problem for such applications. Given the extremely stringent design requirements in performance, power dissipation, and area, an Application-Specific Integrated Circuit (ASIC) enables an efficient real-time exploitation of LSTMs for gait analysis, achieving high accuracy. To the best of our knowledge, this work presents the first cross-layer co-optimized LSTM accelerator for real-time gait analysis, targeting an ASIC design. We conduct a comprehensive design space exploration from software down to layout design. We carry out a bit-width optimization at the software level with hardware-aware quantization to reduce the hardware complexity, explore various designs at the register-transfer level, and generate alternative layouts to find efficient realizations of the LSTM accelerator in terms of hardware complexity and accuracy. The physical synthesis results show that, using the 65 nm technology, the die size of the accelerator's layout optimized for the highest accuracy is 0.325 mm^2, while the alternative design optimized for hardware complexity with a slightly lower accuracy occupies 15.4% smaller area. Moreover, the designed accelerators achieve accurate gait abnormality detection 4.05x faster than the given application requirement.

cs.AR↗

arXiv2026

CataractSAM-2: A Domain-Adapted Model for Anterior Segment Surgery Segmentation and Scalable Ground-Truth Annotation

We present CataractSAM-2, a domain-adapted extension of Meta's Segment Anything Model 2, designed for real-time semantic segmentation of cataract ophthalmic surgery videos with high accuracy. Positioned at the intersection of computer vision and medical robotics, CataractSAM-2 enables precise intraoperative perception crucial for robotic-assisted and computer-guided surgical systems. Furthermore, to alleviate the burden of manual labeling, we introduce an interactive annotation framework that combines sparse prompts with video-based mask propagation. This tool significantly reduces annotation time and facilitates the scalable creation of high-quality ground-truth masks, accelerating dataset development for ocular anterior segment surgeries. We also demonstrate the model's strong zero-shot generalization to glaucoma trabeculectomy procedures, confirming its cross-procedural utility and potential for broader surgical applications. The trained model and annotation toolkit are released as open-source resources, establishing CataractSAM-2 as a foundation for expanding anterior ophthalmic surgical datasets and advancing real-time AI-driven solutions in medical robotics, as well as surgical video understanding.

cs.CV↗

arXiv2025

A Zero-overhead Flow for Security Closure

In the traditional Application-Specific Integrated Circuit (ASIC) design flow, the concept of timing closure implies to reach convergence during physical synthesis such that, under a given area and power budget, the design works at the targeted frequency. However, security has been largely neglected when evaluating the Quality of Results (QoR) from physical synthesis. In general, commercial place & route tools do not understand security goals. In this work, we propose a modified ASIC design flow that is security-aware and, differently from prior research, does not degrade QoR for the sake of security improvement. Therefore, we propose a first-of-its-kind zero-overhead flow for security closure. Our flow is concerned with two distinct threat models: (i) insertion of Hardware Trojans (HTs) and (ii) physical probing/fault injection. Importantly, the flow is entirely executed within a commercial place & route engine and is scalable. In several metrics, our security-aware flow achieves the best-known results for the ISPD`22 set of benchmark circuits while incurring negligible design overheads due to security-related strategies. Finally, we open source the entire methodology (as a set of scripts) and also share the protected circuits (as design databases) for the benefit of the hardware security community.

cs.CR↗

arXiv2024

Federated Learning for Diabetic Retinopathy Diagnosis: Enhancing Accuracy and Generalizability in Under-Resourced Regions

Diabetic retinopathy is the leading cause of vision loss in working-age adults worldwide, yet under-resourced regions lack ophthalmologists. Current state-of-the-art deep learning systems struggle at these institutions due to limited generalizability. This paper explores a novel federated learning system for diabetic retinopathy diagnosis with the EfficientNetB0 architecture to leverage fundus data from multiple institutions to improve diagnostic generalizability at under-resourced hospitals while preserving patient-privacy. The federated model achieved 93.21% accuracy in five-category classification on an unseen dataset and 91.05% on lower-quality images from a simulated under-resourced institution. The model was deployed onto two apps for quick and accurate diagnosis.

eess.IV↗

arXiv2024

SCARF: Securing Chips with a Robust Framework against Fabrication-time Hardware Trojans

The globalization of the semiconductor industry has introduced security challenges to Integrated Circuits (ICs), particularly those related to the threat of Hardware Trojans (HTs) - malicious logic that can be introduced during IC fabrication. While significant efforts are directed towards verifying the correctness and reliability of ICs, their security is often overlooked. In this paper, we propose a comprehensive approach to enhance IC security from the front-end to back-end stages of design. Initially, we outline a systematic method to transform existing verification assets into potent security checkers by repurposing verification assertions. To further improve security, we introduce an innovative technique for integrating online monitors during physical synthesis - a back-end insertion providing an additional layer of defense. Experimental results demonstrate a significant increase in security, measured by our introduced metric, Security Coverage (SC), with a marginal rise in area and power consumption, typically under 20%. The insertion of online monitors during physical synthesis enhances security metrics by up to 33.5%. This holistic approach offers a comprehensive and resilient defense mechanism across the entire spectrum of IC design.

cs.CR↗

arXiv2023

SALSy: Security-Aware Layout Synthesis

Integrated Circuits (ICs) are the target of diverse attacks during their lifetime. Fabrication-time attacks, such as the insertion of Hardware Trojans, can give an adversary access to privileged data and/or the means to corrupt the IC's internal computation. Post-fabrication attacks, where the end-user takes a malicious role, also attempt to obtain privileged information through means such as fault injection and probing. Taking these threats into account and at the same time, this paper proposes a methodology for Security-Aware Layout Synthesis (SALSy), such that ICs can be designed with security in mind in the same manner as power-performance-area (PPA) metrics are considered today, a concept known as security closure. Furthermore, the trade-offs between PPA and security are considered and a chip is fabricated in a 65nm CMOS commercial technology for validation purposes - a feature not seen in previous research on security closure. Measurements on the fabricated ICs indicate that SALSy promotes a modest increase in power in order to achieve significantly improved security metrics.

cs.CR↗

arXiv2023

Reusing Verification Assertions as Security Checkers for Hardware Trojan Detection

Globalization in the semiconductor industry enables fabless design houses to reduce their costs, save time, and make use of newer technologies. However, the offshoring of Integrated Circuit (IC) fabrication has negative sides, including threats such as Hardware Trojans (HTs) - a type of malicious logic that is not trivial to detect. One aspect of IC design that is not affected by globalization is the need for thorough verification. Verification engineers devise complex assets to make sure designs are bug-free, including assertions. This knowledge is typically not reused once verification is over. The premise of this paper is that verification assets that already exist can be turned into effective security checkers for HT detection. For this purpose, we show how assertions can be used as online monitors. To this end, we propose a security metric and an assertion selection flow that leverages Cadence JasperGold Security Path Verification (SPV). The experimental results show that our approach scales for industry-size circuits by analyzing more than 100 assertions for different Intellectual Properties (IPs) of the OpenTitan System-on-Chip (SoC). Moreover, our detection solution is pragmatic since it does not rely on the HT activation mechanism.

cs.CR↗

arXiv2022

Perception-Aligned AI Outputs: End-to-End Visual Prediction for Uncertainty Communication in Clinical Decision-Making

Explainable Artificial Intelligence (XAI) is essential for trustworthy AI in healthcare, yet many existing methods rely on technical explanations that are difficult for clinicians and patients to interpret. We introduce Visualized Learning for Machine Learning (VL4ML), a human-centered explainability framework that communicates model predictions and uncertainty through intuitive visual representations rather than numerical or post-hoc explanations. By encoding diagnostic information in colors, patterns, and spatial structures, VL4ML enables users to interpret predictions without requiring knowledge of model internals or statistical expertise. We demonstrate the framework across multiple clinical tasks, including classification, regression, longitudinal prediction, and multimodal analysis. Its effectiveness was evaluated through a human-centered study involving 158 participants (39.2% clinical professionals) and an expert interpretability assessment. More than 79% of participants positively rated the visual explanations across evaluation dimensions, 84.0% found them more memorable than numeric outputs, and 76.9% reported faster decision-making. Over 82% successfully perceived uncertainty embedded in the visual representations without prior statistical training. No significant differences were observed between clinicians and non-clinicians or between male and female participants, indicating broad accessibility. These results suggest that VL4ML complements existing XAI and uncertainty quantification methods by providing intuitive, universally interpretable visual explanations that support transparent and trustworthy clinical decision-making.

cs.LG↗

arXiv2021

Deep Variational Clustering Framework for Self-labeling of Large-scale Medical Images

We propose a Deep Variational Clustering (DVC) framework for unsupervised representation learning and clustering of large-scale medical images. DVC simultaneously learns the multivariate Gaussian posterior through the probabilistic convolutional encoder and the likelihood distribution with the probabilistic convolutional decoder; and optimizes cluster labels assignment. Here, the learned multivariate Gaussian posterior captures the latent distribution of a large set of unlabeled images. Then, we perform unsupervised clustering on top of the variational latent space using a clustering loss. In this approach, the probabilistic decoder helps to prevent the distortion of data points in the latent space and to preserve the local structure of data generating distribution. The training process can be considered as a self-training process to refine the latent space and simultaneously optimizing cluster assignments iteratively. We evaluated our proposed framework on three public datasets that represented different medical imaging modalities. Our experimental results show that our proposed framework generalizes better across different datasets. It achieves compelling results on several medical imaging benchmarks. Thus, our approach offers potential advantages over conventional deep unsupervised learning in real-world applications. The source code of the method and all the experiments are available publicly at: https://github.com/csfarzin/DVC

cs.CV↗

arXiv2021

Feasibility Assessment of Multitasking in MRI Neuroimaging Analysis: Tissue Segmentation, Cross-Modality Conversion and Bias correction

Neuroimaging is essential in brain studies for the diagnosis and identification of disease, structure, and function of the brain in its healthy and disease states. Literature shows that there are advantages of multitasking with some deep learning (DL) schemes in challenging neuroimaging applications. This study examines the feasibility of using multitasking in three different applications, including tissue segmentation, cross-modality conversion, and bias-field correction. These applications reflect five different scenarios in which multitasking is explored and 280 training and testing sessions conducted for empirical evaluations. Two well-known networks, U-Net as a well-known convolutional neural network architecture, and a closed architecture based on the conditional generative adversarial network are implemented. Different metrics such as the normalized cross-correlation coefficient and Dice scores are used for comparison of methods and results of the different experiments. Statistical analysis is also provided by paired t-test. The present study explores the pros and cons of these methods and their practical impacts on multitasking in different implementation scenarios. This investigation shows that bias correction and cross-modality conversion applications are significantly easier than the segmentation application, and having multitasking with segmentation is not reasonable if one of them is identified as the main target application. However, when the main application is the segmentation of tissues, multitasking with cross-modality conversion is beneficial, especially for the U-net architecture.

eess.IV↗

arXiv2020

Automatic vocal tract landmark localization from midsagittal MRI data

The various speech sounds of a language are obtained by varying the shape and position of the articulators surrounding the vocal tract. Analyzing their variations is crucial for understanding speech production, diagnosing speech disorders and planning therapy. Identifying key anatomical landmarks of these structures on medical images is a pre-requisite for any quantitative analysis and the rising amount of data generated in the field calls for an automatic solution. The challenge lies in the high inter- and intra-speaker variability, the mutual interaction between the articulators and the moderate quality of the images. This study addresses this issue for the first time and tackles it by means by means of Deep Learning. It proposes a dedicated network architecture named Flat-net and its performance are evaluated and compared with eleven state-of-the-art methods from the literature. The dataset contains midsagittal anatomical Magnetic Resonance Images for 9 speakers sustaining 62 articulations with 21 annotated anatomical landmarks per image. Results show that the Flat-net approach outperforms the former methods, leading to an overall Root Mean Square Error of 3.6 pixels/0.36 cm obtained in a leave-one-out procedure over the speakers. The implementation codes are also shared publicly on GitHub.

eess.IV↗

arXiv2019

Image to Images Translation for Multi-Task Organ Segmentation and Bone Suppression in Chest X-Ray Radiography

Chest X-ray radiography is one of the earliest medical imaging technologies and remains one of the most widely-used for diagnosis, screening, and treatment follow up of diseases related to lungs and heart. The literature in this field of research reports many interesting studies dealing with the challenging tasks of bone suppression and organ segmentation but performed separately, limiting any learning that comes with the consolidation of parameters that could optimize both processes. This study, and for the first time, introduces a multitask deep learning model that generates simultaneously the bone-suppressed image and the organ-segmented image, enhancing the accuracy of tasks, minimizing the number of parameters needed by the model and optimizing the processing time, all by exploiting the interplay between the network parameters to benefit the performance of both tasks. The architectural design of this model, which relies on a conditional generative adversarial network, reveals the process on how the well-established pix2pix network (image-to-image network) is modified to fit the need for multitasking and extending it to the new image-to-images architecture. The developed source code of this multitask model is shared publicly on Github as the first attempt for providing the two-task pix2pix extension, a supervised/paired/aligned/registered image-to-images translation which would be useful in many multitask applications. Dilated convolutions are also used to improve the results through a more effective receptive field assessment. The comparison with state-of-the-art algorithms along with ablation study and a demonstration video are provided to evaluate efficacy and gauge the merits of the proposed approach.

eess.IV↗

arXiv2019

SignCol: Open-Source Software for Collecting Sign Language Gestures

Sign(ed) languages use gestures, such as hand or head movements, for communication. Sign language recognition is an assistive technology for individuals with hearing disability and its goal is to improve such individuals' life quality by facilitating their social involvement. Since sign languages are vastly varied in alphabets, as known as signs, a sign recognition software should be capable of handling eight different types of sign combinations, e.g. numbers, letters, words and sentences. Due to the intrinsic complexity and diversity of symbolic gestures, recognition algorithms need a comprehensive visual dataset to learn by. In this paper, we describe the design and implementation of a Microsoft Kinect-based open source software, called SignCol, for capturing and saving the gestures used in sign languages. Our work supports a multi-language database and reports the recorded items statistics. SignCol can capture and store colored(RGB) frames, depth frames, infrared frames, body index frames, coordinate mapped color-body frames, skeleton information of each frame and camera parameters simultaneously.

cs.HC↗

arXiv2017

A Compressive Method for Centralized PSD Map Construction with Imperfect Reporting Channel

Spectrum resources management of growing demands is a challenging problem and Cognitive Radio (CR) known to be capable of improving the spectrum utilization. Recently, Power Spectral Density (PSD) map is defined to enable the CR to reuse the frequency resources regarding to the area. For this reason, the sensed PSDs are collected by the distributed sensors in the area and fused by a Fusion Center (FC). But, for a given zone, the sensed PSDs by neighbor CR sensors may contain a shared common component for a while. This component can be exploited in the theory of the Distributed Source Coding (DSC) to make the sensors transmission data more compressed. However, uncertain channel fading and random shadowing would lead to varying signal strength at different CRs, even placed close to each other. Hence, existence of some perturbations in the transmission procedure yields to some imperfection in the reporting channel and as a result it degrades the performance remarkably. The main focus of this paper is to be able to reconstruct the PSDs of sensors \textit{robustly} based on the Distributed Compressive Sensing (DCS) when the data transmission is slightly imperfect. Simulation results verify the robustness of the proposed scheme.

cs.IT↗

arXiv2016

A Compressive Method for Centralized PSD Map Construction

Spectrum resources are facing huge demands and cognitive radio (CR) can improve the spectrum utilization. Recently, power spectral density (PSD) map is defined to enable the CR to reuse the frequency resources regarding to the area. For this reason, the sensed PSDs are fused by a Fusion Center (FC) which the sensed PSDs are collected by the distributed sensors in the area. But, for a given zone, the sensed PSD by neighbor CR sensors may contain a shared common component for a while. This component can be exploited in the theory of the distributed source coding (DSC) to compress sensing data more. In this paper based on the distributed compressive sensing (DCS) a method is proposed to compress and reconstruct the PSDs of the sensors when the data transmission is slightly imperfect. Simulation results show the advantages of using proposed method in compressing, reducing overhead and also recovering PSDs. % Proposed method can be used to develop a framework when the holding times of the users are large in comparison with the rate of the spectrum sensing.

cs.IT↗