Adnan Saeed

Adnan Saeed is a deep learning researcher working on medical image analysis, with a focus on multimodal architectures, graph neural networks, and evidential deep learning for clinical imaging tasks. His peer reviewed research has appeared in journals across machine learning and biomedical signal processing. At AI Trend Blend he turns recent papers into clear, practical explainers, with an emphasis on what a method actually does and where it holds up, written for readers who want depth without the hype.

ADAPTS Classifies Concept Drift Before It Decides How To Adapt.

ADAPTS Classifies Concept Drift Before It Decides How To Adapt

Analysis by the aitrendblend editorial team · Continual Learning and Concept Drift Adaptation · 15 min read Concept Drift Anomaly Detection Continual Learning Time Series Unsupervised Learning A conceptual illustration of drift aware pool based adaptation, not an original figure from the paper. A sensor in an industrial plant fails overnight and its readings jump

ADAPTS Classifies Concept Drift Before It Decides How To Adapt Read More »

How Stochastic Transport Fixes Composite Image Restoration.

How Stochastic Transport Fixes Composite Image Restoration

Analysis by the aitrendblend editorial team · Pillar 3, Generative AI and diffusion models · Published in Knowledge-Based Systems, volume 349, 2026, DOI 10.1016/j.knosys.2026.116411 stochastic transport flow matching mixture of experts composite degradation image restoration F2D-Net factorizes the restoration flow into a shared backbone plus pixel gated experts, driven by noise that shrinks to zero

How Stochastic Transport Fixes Composite Image Restoration Read More »

Graph Neural Networks Bring Coherent Forecasts to Retail

Graph Neural Networks Bring Coherent Forecasts to Retail

Analysis by the aitrendblend editorial team · Pillar 5, Graph neural networks · Reading time about 14 minutes graph neural networks hierarchical forecasting retail demand GCN and GAT forecast reconciliation A retail sales hierarchy reimagined as a graph, where store totals, brand groups and individual items all learn from each other before a forecast ever

Graph Neural Networks Bring Coherent Forecasts to Retail Read More »

Random Shuffle RWKV Fixes Directional Bias In Image Fusion.

Random Shuffle RWKV Fixes Directional Bias In Image Fusion

Analysis by the aitrendblend editorial team · Pillar 4, Vision transformers and attention · Published in Information Fusion, volume 136, 2026, DOI 10.1016/j.inffus.2026.104545 RWKV attention pan sharpening random shuffle scanning linear attention remote sensing fusion Random shuffle plus inverse shuffle removes fixed scan order bias from vision RWKV attention. Source, Zhou et al., 2026. Ask

Random Shuffle RWKV Fixes Directional Bias In Image Fusion Read More »

How ProtoSig Uses Clustering to Make Signature Verification Faster, Fairer, and More Stable.

How ProtoSig Uses Clustering to Make Signature Verification Faster, Fairer, and More Stable

ProtoSig replaces thousands of random forgeries with 50 clustered prototype signatures, cutting training compute by over 98% while matching verification accuracy — and making signature verification fairer and more stable.

How ProtoSig Uses Clustering to Make Signature Verification Faster, Fairer, and More Stable Read More »