arXiv subjects

Utsab Saha

Publications and source records attributed to Utsab Saha.

10 recordsLinked to original sources

arXiv2026

Comprehensive Dataset and Signal Processing Framework for Phonocardiogram-Based Heart Rate and Blood Pressure Estimation

Cardiovascular diseases (CVDs) represent significant global health challenges today, necessitating regular and reliable monitoring to enable early intervention. Phonocardiogram (PCG) signals present a promising non-invasive method for assessing cardiovascular health. While recent studies have focused on estimating heart rate (HR) from PCG signals and blood pressure (BP) through multimodal combinations with other physiological data, reliable and cost-effective systems that can predict both HR and BP using only PCG signals remain largely unexplored. In this study, we proposed and developed a lab-scale cost-effective Phonocardiogram Tracking (PhonoTrack) system that can measure both HR and BP using only the PCG signal. We also introduced a corresponding dataset collected from 15 participants to evaluate the effectiveness of the proposed system. HR was determined using several peak detection methods, such as Hilbert Transform (HT), Shannon Entropy (SE), and WES, achieving notable Pearson correlation coefficients of 0.965, 0.973, and 0.955, respectively. The corresponding root mean square errors (RMSEs) were 2.467 bpm, 1.688 bpm, and 1.992 bpm for HT, SE, and WES, respectively. Additionally, we developed an advanced semi-empirical model based on multiple regression techniques to estimate systolic blood pressure (SBP) and diastolic blood pressure (DBP). This model demonstrated standard deviations of 2.10 mmHg for SBP and 3.20 mmHg for DBP across all subjects, with Pearson correlation coefficients of 0.89 and 0.70, respectively. These findings pave the way for developing a non-invasive, low-cost, and portable PhonoTrack device, positioning it as a promising solution for continuous cardiovascular monitoring settings.

eess.SP↗

arXiv2026

Writing Blog Posts Helps Students Connect Experiential Learning to the Workplace

Undergraduates in work-based learning experiences often produce meaningful contributions as viewed by their supervisors, yet report a negative perception of their contributions because they struggled during the process or produced only a few lines of code change. As a result, many omit these contributions from their resumes and job interviews, losing a meaningful signal of technical ability. This study examines how guided blog posts help CS students in work based learning experiences reflect on what they learned and contextualize their experiences. It also evaluates the depth of reflection produced. The study included twenty-five juniors and seniors studying CS at CTCs and other affordable local colleges. All participated in one cohort during Fall 2024. Each student was assigned a simple open source issue to solve from a popular open source project over the course of several weeks with the help of an industry mentor. While working on the project, students drafted a LinkedIn blog post using a five-section outline covering project mission, assigned issue, technical architecture, challenges faced, and submitted solution. We conducted a thematic analysis of the published posts and measured reflection depth using Mejia and Turns's Knowledge Gain instrument. Four themes emerged from the posts: identifying problem solving techniques, growth mindset, the challenges and benefits of collaborative development, and the impacts of their contribution on users. Additionally, students demonstrated deep reflection across all four Knowledge Gain constructs. Structured blog posts offer a low-cost addition to experiential CS learning such as capstones, micro-internships, internships, and apprenticeships. This study is descriptive; future work should compare outcomes against a control group.

cs.CY↗

arXiv2025

The Open Source Resume: How Open Source Contributions Help Students Demonstrate Alignment with Employer Needs

Computer science educators are increasingly integrating open source contributions into classes to prepare students for higher expectations due to GenAI, and to improve employment outcomes in an increasingly competitive job market. However, little is known about how employers view student open source contributions. This paper addresses two research questions qualitatively: what traits do employers desire for entry-level hires in 2025, and how can they be demonstrated through open source contributions? It also tests quantitatively the hypothesis that student knowledge of employers' expectations will improve their motivation to work on open source projects. To answer our qualitative questions, we conducted interviews with US hiring managers. We collaborated with each interviewee to create a "hiring manager agreement," which listed desirable traits and specific ways to demonstrate them through open source, along with a promise to interview some students meeting the criteria. To evaluate our quantitative hypothesis, we surveyed 650 undergraduates attending public universities in the US using an instrument based on expectancy-value theory. Hiring managers wanted many non-technical traits that are difficult to teach in traditional CS classes, such as initiative. There were many commonalities in how employers wanted to see these traits demonstrated in open source contributions. Viewing hiring manager agreements improved student motivation to contribute to open source projects. Our findings suggest that open source contributions may help CS undergraduates get hired, but this requires sustained engagement in multiple areas. Educators can motivate students by sharing employer expectations, but further work is required to determine if this changes their behavior.

cs.CY↗

arXiv2025

AudioFuse: Unified Spectral-Temporal Learning via a Hybrid ViT-1D CNN Architecture for Robust Phonocardiogram Classification

Biomedical audio signals, such as phonocardiograms (PCG), are inherently rhythmic and contain diagnostic information in both their spectral (tonal) and temporal domains. Standard 2D spectrograms provide rich spectral features but compromise the phase information and temporal precision of the 1D waveform. We propose AudioFuse, an architecture that simultaneously learns from both complementary representations to classify PCGs. To mitigate the overfitting risk common in fusion models, we integrate a custom, wide-and-shallow Vision Transformer (ViT) for spectrograms with a shallow 1D CNN for raw waveforms. On the PhysioNet 2016 dataset, AudioFuse achieves a state-of-the-art competitive ROC-AUC of 0.8608 when trained from scratch, outperforming its spectrogram (0.8066) and waveform (0.8223) baselines. Moreover, it demonstrates superior robustness to domain shift on the challenging PASCAL dataset, maintaining an ROC-AUC of 0.7181 while the spectrogram baseline collapses (0.4873). Fusing complementary representations thus provides a strong inductive bias, enabling the creation of efficient, generalizable classifiers without requiring large-scale pre-training. The complete source code for our baseline and AudioFuse models is publicly available on GitHub at: https://github.com/Saiful185/AudioFuse.

eess.AS↗

arXiv2025

Differentially Private Decentralized Dataset Synthesis Through Randomized Mixing with Correlated Noise

In this work, we explore differentially private synthetic data generation in a decentralized-data setting by building on the recently proposed Differentially Private Class-Centric Data Aggregation (DP-CDA). DP-CDA synthesizes data in a centralized setting by mixing multiple randomly-selected samples from the same class and injecting carefully calibrated Gaussian noise, ensuring ({\epsilon}, {\delta})-differential privacy. When deployed in a decentralized or federated setting, where each client holds only a small partition of the data, DP-CDA faces new challenges. The limited sample size per client increases the sensitivity of local computations, requiring higher noise injection to maintain the differential privacy guarantee. This, in turn, leads to a noticeable degradation in the utility compared to the centralized setting. To mitigate this issue, we integrate the Correlation-Assisted Private Estimation (CAPE) protocol into the federated DP-CDA framework and propose CAPE Assisted Federated DP-CDA algorithm. CAPE enables limited collaboration among the clients by allowing them to generate jointly distributed (anti-correlated) noise that cancels out in aggregate, while preserving privacy at the individual level. This technique significantly improves the privacy-utility trade-off in the federated setting. Extensive experiments on MNIST and FashionMNIST datasets demonstrate that the proposed CAPE Assisted Federated DP-CDA approach can achieve utility comparable to its centralized counterpart under some parameter regime, while maintaining rigorous differential privacy guarantees.

stat.ML↗

arXiv2025

CAE-Net: Generalized Deepfake Image Detection using Convolution and Attention Mechanisms with Spatial and Frequency Domain Features

The spread of deepfakes poses significant security concerns, demanding reliable detection methods. However, diverse generation techniques and class imbalance in datasets create challenges. We propose CAE-Net, a Convolution- and Attention-based weighted Ensemble network combining spatial and frequency-domain features for effective deepfake detection. The architecture integrates EfficientNet, Data-Efficient Image Transformer (DeiT), and ConvNeXt with wavelet features to learn complementary representations. We evaluated CAE-Net on the diverse IEEE Signal Processing Cup 2025 (DF-Wild Cup) dataset, which has a 5:1 fake-to-real class imbalance. To address this, we introduce a multistage disjoint-subset training strategy, sequentially training the model on non-overlapping subsets of the fake class while retaining knowledge across stages. Our approach achieved $94.46\%$ accuracy and a $97.60\%$ AUC, outperforming conventional class-balancing methods. Visualizations confirm the network focuses on meaningful facial regions, and our ensemble design demonstrates robustness against adversarial attacks, positioning CAE-Net as a dependable and generalized deepfake detection framework.

cs.CV↗

arXiv2024

DP-CDA: An Algorithm for Enhanced Privacy Preservation in Dataset Synthesis Through Randomized Mixing

In recent years, the growth of data across various sectors, including healthcare, security, finance, and education, has created significant opportunities for analysis and informed decision-making. However, these datasets often contain sensitive and personal information, which raises serious privacy concerns. It has been shown in multiple works that a person's identity is intertwined with their data, even if the data is anonymized. Due to this lack of separation between a person's identity and their information, the patterns associated with an individual's information can uniquely identify them. Protecting individual privacy is crucial, yet many existing machine learning and data publishing algorithms struggle with high-dimensional data, facing challenges related to the trade-off between computational efficiency and privacy. To address these challenges, we introduce an effective data publishing algorithm \emph{DP-CDA}. Our proposed algorithm generates synthetic data by randomly mixing the privacy-sensitive data in a class-specific manner and inducing carefully tuned randomness to ensure formal privacy guarantees. Our comprehensive privacy accounting shows that the proposed DP-CDA provides a stronger privacy guarantee compared to existing methods, allowing for better utility while maintaining a stricter level of privacy. To evaluate the effectiveness of DP-CDA, we examine the accuracy of predictive models trained on the synthetic data, which serves as a measure of dataset utility. Importantly, we identify an optimal order of mixing that balances privacy-utility trade-off. Our results indicate that synthetic datasets produced using the DP-CDA can achieve superior utility compared to those generated by conventional data publishing algorithms, even when subject to the same privacy requirements.

stat.ML↗

arXiv2024

Npix2Cpix: A GAN-Based Image-to-Image Translation Network With Retrieval- Classification Integration for Watermark Retrieval From Historical Document Images

The identification and restoration of ancient watermarks have long been a major topic in codicology and history. Classifying historical documents based on watermarks is challenging due to their diversity, noisy samples, multiple representation modes, and minor distinctions between classes and intra-class variations. This paper proposes a modified U-net-based conditional generative adversarial network (GAN) named Npix2Cpix to translate noisy raw historical watermarked images into clean, handwriting-free watermarked images by performing image translation from degraded (noisy) pixels to clean pixels. Using image-to-image translation and adversarial learning, the network creates clutter-free images for watermark restoration and categorization. The generator and discriminator of the proposed GAN are trained using two separate loss functions, each based on the distance between images, to learn the mapping from the input noisy image to the output clean image. After using the proposed GAN to pre-process noisy watermarked images, Siamese-based one-shot learning is employed for watermark classification. Experimental results on a large-scale historical watermark dataset demonstrate that cleaning the noisy watermarked images can help to achieve high one-shot classification accuracy. The qualitative and quantitative evaluation of the retrieved watermarked image highlights the effectiveness of the proposed approach.

cs.CV↗

arXiv2023

Proximal Policy Optimization-Based Reinforcement Learning Approach for DC-DC Boost Converter Control: A Comparative Evaluation Against Traditional Control Techniques

This article proposes a proximal policy optimization (PPO)-based reinforcement learning (RL) approach for DC-DC boost converter control that is compared with traditional control methods. The performance of the PPO algorithm is evaluated using MATLAB Simulink co-simulation, and the results demonstrate that the most efficient approach for achieving short settling time and stability is to combine the PPO algorithm with a reinforcement learning-based control method. The simulation results show that the control method based on RL with the PPO algorithm pro vides step response characteristics that outperform traditional control approaches, thereby enhancing DC-DC boost converter control. This research also highlights the inherent capability of the reinforcement learning method to enhance the performance of boost converter control.

eess.SY↗

arXiv2023

Decoding Human Activities: Analyzing Wearable Accelerometer and Gyroscope Data for Activity Recognition

A person's movement or relative positioning can be effectively captured by different types of sensors and corresponding sensor output can be utilized in various manipulative techniques for the classification of different human activities. This letter proposes an effective scheme for human activity recognition, which introduces two unique approaches within a multi-structural architecture, named FusionActNet. The first approach aims to capture the static and dynamic behavior of a particular action by using two dedicated residual networks and the second approach facilitates the final decision-making process by introducing a guidance module. A two-stage training process is designed where at the first stage, residual networks are pre-trained separately by using static (where the human body is immobile) and dynamic (involving movement of the human body) data. In the next stage, the guidance module along with the pre-trained static or dynamic models are used to train the given sensor data. Here the guidance module learns to emphasize the most relevant prediction vector obtained from the static or dynamic models, which helps to effectively classify different human activities. The proposed scheme is evaluated using two benchmark datasets and compared with state-of-the-art methods. The results clearly demonstrate that our method outperforms existing approaches in terms of accuracy, precision, recall, and F1 score, achieving 97.35% and 95.35% accuracy on the UCI HAR and Motion-Sense datasets, respectively which highlights both the effectiveness and stability of the proposed scheme.

cs.CV↗