arXiv subjects

Shadman Sobhan

Publications and source records attributed to Shadman Sobhan.

4 recordsLinked to original sources

arXiv2026

LightVesselNet: An Ultra-Lightweight Sub-100K Parameter Network for Retinal Blood Vessel Segmentation

Retinal blood vessel segmentation plays a vital role in the early detection of diabetic retinopathy and glaucoma. While recent deep learning models have achieved great segmentation accuracy, they typically require heavy computational resources, making real-world deployment on edge devices difficult. In this paper, we propose LightVesselNet, an efficient neural network designed for retinal vessel segmentation in a resource-constrained environment. Despite containing only 75K parameters, LightVesselNet performs competitively with much larger models. The network employs a compact encoder decoder architecture enhanced with channel and spatial attention mechanisms, a multi-scale feature aggregation module at the bottleneck, and a subpixel upsampling strategy in the decoder. A dedicated edge residual connection preserves fine vessel detail throughout decoding. Extensive experiments on five publicly available datasets: DRIVE, STARE, CHASEDB1, FIVES, and HRF, yield sensitivity scores of 0.8189, 0.8499, 0.8640, 0.8634, 0.8096, and Dice coefficients of 0.8070, 0.8072, 0.8181, 0.8649, and 0.7686, respectively. LightVesselNet shows improved efficiency (Performance vs Parameter or GFlops) compared to State-of-the-Art models. Cross-dataset evaluation confirms the model's generalisation capability. Overall, LightVesselNet is a strong candidate for deployment in low-resource clinical settings and mobile screening tools.

cs.CV↗

arXiv2025

PromptForSegCXR: Prompt-Driven Multi-Organ and Multi-Disease Segmentation in Chest X-rays using a Multi-stage Fusion Mechanism

Image segmentation is central to automated medical image analysis, enabling precise identification of anatomical structures and pathological regions. Conventional segmentation models typically target a single organ or disease, limiting their adaptability across clinical scenarios. While multi-organ and multi-disease segmentation has been explored, building such datasets requires extensive manual annotation by medical experts. Prompt-driven segmentation offers a flexible, user-guided alternative that speeds up annotation, yet no prior work has addressed prompt-based interactive segmentation across multiple organs and diseases in chest X-rays. This study makes two main contributions. First, we introduce a novel dataset of expert-designed doodle prompts spanning 23 classes (six organs and seventeen diseases), curated from multiple public chest X-ray datasets for prompt-driven segmentation. Second, we propose PromptForSegCXR, a lightweight dual-input segmentation framework that combines the chest X-ray with user-provided doodle prompts to accurately segment diverse anatomical and pathological regions. The model uses a multi-stage feature fusion strategy to integrate spatial and semantic representations, along with a depthwise-pointwise-residual convolution block with squeeze-and-excitation attention for efficient hierarchical feature extraction and adaptive recalibration. Experimental results show the model achieves a Dice score of 81.62 percent on the full dataset, outperforming SAM-based prompt segmentation models by up to 10 percent and conventional segmentation architectures by up to 23 percent, while remaining lightweight. These results demonstrate the effectiveness of the proposed approach for accurate, flexible, prompt-driven chest X-ray segmentation.

eess.IV↗

arXiv2025

LLM-Assisted Question-Answering on Technical Documents Using Structured Data-Aware Retrieval Augmented Generation

Large Language Models (LLMs) are capable of natural language understanding and generation. But they face challenges such as hallucination and outdated knowledge. Fine-tuning is one possible solution, but it is resource-intensive and must be repeated with every data update. Retrieval-Augmented Generation (RAG) offers an efficient solution by allowing LLMs to access external knowledge sources. However, traditional RAG pipelines struggle with retrieving information from complex technical documents with structured data such as tables and images. In this work, we propose a RAG pipeline, capable of handling tables and images in documents, for technical documents that support both scanned and searchable formats. Its retrieval process combines vector similarity search with a fine-tuned reranker based on Gemma-2-9b-it. The reranker is trained using RAFT (Retrieval-Augmented Fine-Tuning) on a custom dataset designed to improve context identification for question answering. Our evaluation demonstrates that the proposed pipeline achieves a high faithfulness score of 94% (RAGas) and 96% (DeepEval), and an answer relevancy score of 87% (RAGas) and 93% (DeepEval). Comparative analysis demonstrates that the proposed architecture is superior to general RAG pipelines in terms of table-based questions and handling questions outside context.

cs.CL↗

arXiv2025

MedPrompt: LLM-CNN Fusion with Weight Routing for Medical Image Segmentation and Classification

Current medical image analysis systems are typically task-specific, requiring separate models for classification and segmentation, and lack the flexibility to support user-defined workflows. To address these challenges, we introduce MedPrompt, a unified framework that combines a few-shot prompted Large Language Model (Llama-4-17B) for high-level task planning with a modular Convolutional Neural Network (DeepFusionLab) for low-level image processing. The LLM interprets user instructions and generates structured output to dynamically route task-specific pretrained weights. This weight routing approach avoids retraining the entire framework when adding new tasks-only task-specific weights are required, enhancing scalability and deployment. We evaluated MedPrompt across 19 public datasets, covering 12 tasks spanning 5 imaging modalities. The system achieves a 97% end-to-end correctness in interpreting and executing prompt-driven instructions, with an average inference latency of 2.5 seconds, making it suitable for near real-time applications. DeepFusionLab achieves competitive segmentation accuracy (e.g., Dice 0.9856 on lungs) and strong classification performance (F1 0.9744 on tuberculosis). Overall, MedPrompt enables scalable, prompt-driven medical imaging by combining the interpretability of LLMs with the efficiency of modular CNNs.

cs.CV↗