Paper Title : Hardware Requirements for 2D Cylindrical-High Pass Ladder Coil Design enabling Homogeneous Excitation in Ultra High-Field MRI
Venue and Year : 2023 IEEE Symposium on Industrial Electronics & Applications (ISIEA)
Instrumentation Science & Technology
The Challenge: At 7T ultra-high-field MRI, standard RF coils suffer from signal shading (inhomogeneity) and high energy absorption (SAR) due to short wavelengths.
Our Solution: We developed a volumetric 2D cylindrical high-pass ladder (c-HPL) coil architecture that uses an inductively coupled network of resonators to distribute radiofrequency fields smoothly.
Project Overview: Hardware design files, analytical models, and bench-testing scripts for a 2D cylindrical high-pass ladder (c-HPL) MRI volume coil built for 7T ultra-high-field applications.
The Impact: This design achieves exceptional transmit homogeneity, paving the way for clearer, safer neuroimaging and high-resolution clinical diagnostics below the 1mm threshold.
Work was conducted by Dr. Shadeeb Hossain along with the listed co-authors while serving as a Postdoctoral Research Fellow at Weill Cornell Medicine (in collaboration with GE Healthcare).
Please cite the following publications if you are using these resources:
[1]Hossain, S., Taracila, V., Robb, F. J., Moore, J., & Winkler, S. A. (2023, July). Hardware Requirements for 2D Cylindrical-High Pass Ladder Coil Design enabling Homogeneous Excitation in Ultra High-Field MRI. In 2023 IEEE Symposium on Industrial Electronics & Applications (ISIEA) (pp. 1-4). IEEE. [Paper] [GitHub Repository] [Conference Presentation]
[2] Hossain, S., Taracila, V., Robb, F. J., Moore, J., & Winkler, S. A. (2023). Design of a volumetric cylindrical coil-tuned at 298 MHz for 7 T imaging. Instrumentation Science & Technology, 1-23. [Paper] [GitHub Repository]
Paper Title : Artificial Intelligence to Analyze Auscultation
Signals to Detect PDA Heart Defect
Venue and Year : 2025 IEEE 4th International Conference on Computing and Machine Intelligence (ICMI)
The Challenge: Accurate early screening for congenital heart defects like Patent Ductus Arteriosus (PDA) typically requires specialized pediatric cardiologists and expert auscultation interpretation.
Our Solution: We developed an automated machine learning framework utilizing Convolutional Neural Networks (CNNs) and auditory-inspired signal processing to classify phonocardiogram (PCG) spectrograms.
The Impact: By applying Butterworth and gammatone filtering alongside Short-Time Fourier Transform (STFT) pre-processing, our optimized CNN architecture achieves a validation accuracy of 96%, providing a robust proof-of-concept for affordable, non-invasive digital health screenings.
Please cite the following publications if you are using these resources:
[1]S. Hossain, "Artificial Intelligence to Analyze Auscultation Signals to Detect PDA Heart Defect," 2025 IEEE 4th International Conference on Computing and Machine Intelligence (ICMI), MI, USA, 2025, pp. 1-4, doi: 10.1109/ICMI65310.2025.11141338. [Paper]