An open research lab dedicated to pioneering AI, Machine Learning, and Deep Learning - building transparent, ethical, and impactful intelligent systems.
DIL is an independent online research laboratory bringing together researchers, engineers, and innovators to solve problems that matter - at the intersection of theory and practice.
To push the boundaries of artificial intelligence, machine learning, and deep learning through rigorous open research - developing intelligent systems that are transparent, ethical, and impactful, bridging academic research and real-world applications.
To become a globally recognized research lab that pioneers transformative AI/ML solutions — fostering a new generation of researchers, innovators, and technologists who drive intelligent progress across industries, institutions, and communities worldwide.
To publish cutting-edge research, build open-source tools, create high-quality datasets, nurture undergraduate researchers, and collaborate globally with academia and industry to tackle real-world challenges using intelligent systems.
Our lab focuses on six interconnected research pillars that span fundamental theory and applied innovation.
Reasoning systems, knowledge representation, intelligent agents, and AI-driven decision-making frameworks for complex real-world problems.
Supervised, unsupervised, and reinforcement learning algorithms. Applied ML pipelines for prediction, classification, and anomaly detection.
Neural network architectures including CNNs, RNNs, Transformers, and generative models for solving high-dimensional learning tasks.
Language models, sentiment analysis, image recognition, object detection, and multimodal AI systems combining text and visual data.
AI applications in healthcare, biomedical imaging, agriculture (tea leaf, rice leaf), and environmental science for societal impact.
Peer-reviewed papers, conference proceedings, and journal articles published by Dipanchal Innovation Lab researchers.
We're preparing exciting research publications. Stay tuned for groundbreaking insights from our lab!
Current investigations underway at DIL - open for collaboration and contributions.
Developing deep learning models for early-stage detection of esophageal and oral cancers using endoscopic and CT scan imaging data.
A PRISMA-guided review assessing the clinical readiness of machine learning, deep learning, and large language models for medical disease diagnosis across validation, interpretability, privacy, fairness, and deployment challenges.
This project develops a hybrid deep learning framework to identify fish diseases, monitor environmental conditions, predict outbreak risks, and recommend treatments for sustainable aquaculture management.
This study demonstrates that improving melanoma detection is primarily a task-structuring problem rather than an architectural complexity problem.
We analyzed 13 years of Sylhet’s air pollution data and built a machine learning model to forecast air quality for public health and policy
A lightweight attention-based model that detects bone cancer from X-rays, explains its decisions visually, quantifies its own confidence, and remains reliable under real-world noisy conditions.
Curated and annotated datasets developed by DIL for the research community — some open access, some available on request.
A comprehensive dataset of 2,000+ Leading University student results - systematically organized and annotated to support predictive modeling, trend analysis, and AI-driven educational research.
The researchers, advisors, and innovators who make DIL's work possible - organized by their roles in the lab.
Organized knowledge base - PDFs, videos, and reference materials categorized by domain and sub-topic are coming soon.....
Have a collaboration idea, research query, or want to join the lab? We'd love to hear from you.
We typically respond within 48 hours.