Projects
A First Order Filter for the Detection of Potentially Habitable Exoplanets
June 2025-September 2025
Status: Completed
ArXiv ID: arXiv:2512.00899
Developed a first-order geometric filter for the preliminary identification of potentially habitable exoplanets, benchmarked to Earth–Sun system ratios of orbital distance to stellar diameter and stellar to planetary diameter. The method enables rapid, interpretable prioritization of exoplanet candidates across G, K, and M-type stars, reducing follow-up resource requirements.
Interpretable ECG-Based Prediction of Myocardial Infarction Using the PRANALI Pipeline
May 2025-October 2025
Status: Completed
SSRN ID: ssrn.5734913
Developed a robust and interpretable machine learning pipeline — PRANALI — for detecting myocardial infarction using extracted features from 12-lead ECG signals. The project involved large-scale ECG data preprocessing, statistical and spectral feature engineering, class imbalance handling (SMOTE), and XGBoost-based classification. Model interpretability was achieved using SHAP analysis to identify critical ECG markers associated with myocardial infarction. Achieved a high ROC-AUC score of 0.84, demonstrating clinical relevance.
A Dynamical Scalar Field Model for Dark Energy: Addressing the Hubble Tension and Cosmic Evolution
January 2025-September 2025
Status: Completed
ArXiv ID: arXiv:2511.10317
Developing a dynamical dark energy model where a scalar field with a time-dependent potential replaces the cosmological constant, allowing it to act as both Early Dark Energy (EDE) and late-time dark energy. This model aims to address the Hubble tension while maintaining consistency with CMBR and Large-Scale Structure (LSS) data. The research involves comparing Lambda-CDM vs. DDE fits, analyzing observational constraints, and refining the potential evolution to align with cosmological data.
A Streamlined Framework for Simulating and Fitting Galactic Rotation Curves
November 2024-August 2025
Status: Completed
ASCL ID: ascl:2510.004
Developed a Python-based framework for simulating and fitting galactic rotation curves using physically motivated bulge, disk, and halo models (Hernquist, de Vaucouleurs, exponential disk, NFW, Burkert). Implemented three fitting techniques—non-linear least squares, bootstrap resampling, and MCMC—with full uncertainty propagation, diagnostic visualizations, and support for both tabulated and raw FITS data. Designed for modularity, reproducibility, and adaptability, enabling both research applications and pedagogical use in galactic dynamics and dark matter studies.
Investigating the Correlation Between Galactic Rotation Curves, Ages, and M/L Ratios in Spiral Galaxies
December 2023-July 2025
Status: Completed
ArXiv ID: arXiv:2512.00823
Analyzed 16 nearby spiral galaxies to investigate links between stellar age, dark matter content, and I-band mass-to-light ratios. Modeled rotation curve data with Hernquist bulge, exponential disk, and NFW halo profiles using Monte Carlo fitting, and identified dark matter–dominated regions to derive masses and luminosities. Found strong correlations between age and both dark matter mass and density, suggesting galaxies with higher dark matter content tend to be older. Results support scenarios where early dark matter build-up shapes long-term galactic evolution.