ResnDevs Lab
Applied research that ships.
Seven research threads from the studio behind CMS and DrChamber — pursued with a bias toward results that hold up in production.
Research areas
Seven areas, one question.
Publications
Writing from the Lab.
- SubmittedNature Scientific Reports
Hybrid Retrieval-Augmented Framework for Low-Resource Bangla Visual Question Answering
A hybrid retrieval-augmented approach to visual question answering for low-resource Bangla — submitted to Nature Scientific Reports.
- SubmittedNature Scientific Reports
Deep Vision for Rail Safety: Real-Time Object Detection at Level Crossings
Computer-vision work on real-time object detection for railway level crossings — submitted to Nature Scientific Reports.
- SubmittedSpringer Nature Pattern Analysis and Applications (Q2)
HybridADNet: A Multi-Paradigm Deep Learning Framework for Alzheimer's Disease Diagnosis from Brain MRI
A multi-paradigm deep learning framework for Alzheimer's disease diagnosis from brain MRI — submitted to Springer Nature Pattern Analysis and Applications (special invitation from MedPRAI).
- SubmittedSpringer Nature Journal of Biomedical Science
HybridVirusNet: A Hybrid Detection–Classification Framework for Multi-Class Virus Identification in TEM Images
A hybrid detection–classification framework for multi-class virus identification in transmission electron microscopy images — submitted to Springer Nature Journal of Biomedical Science.
- SubmittedMedical Image Analysis (Q1)
HyConDViT-Net: A Vision-Based Hybrid Deep Learning Approach Integrating CNNs, Vision Transformers, and Detection Backbones for PCOS Detection from Ultrasound Imaging
A hybrid CNN, Vision Transformer, and detection-backbone approach for PCOS detection from ultrasound imaging — submitted to Medical Image Analysis.
- Writing
Why the n matters
We named the company after a conjunction. The argument for why the space between research and development is the interesting place to stand.