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.

A medical AI system using YOLOv8 and hyperparameter optimization to detect coronary artery stenosis in invasive coronary angiography images.

Hyperparameter Optimization of YOLO Models for Invasive Coronary Angiography Lesion Detection

4 Comments / Machine Learning /

Revolutionizing Cardiac Care: How Hyperparameter Optimization Boosts YOLO Accuracy in Coronary Lesion Detection Cardiovascular diseases remain the leading cause of death worldwide, with coronary artery disease (CAD) at the forefront. Early and accurate detection of coronary stenosis—narrowing of the arteries supplying the heart—is critical for timely intervention and improved patient outcomes. While invasive coronary angiography […]

Hyperparameter Optimization of YOLO Models for Invasive Coronary Angiography Lesion Detection Read More »

Diagram illustrating the FRIES framework for estimating inconsistency in saliency metrics across deep learning models and perturbations.

FRIES: A Groundbreaking Framework for Inconsistency Estimation of Saliency Metrics

1 Comment / Machine Learning /

Unlocking Trust in AI: Introducing FRIES – The First Framework for Inconsistency Estimation of Saliency Metrics As artificial intelligence (AI) becomes increasingly embedded in high-stakes domains like healthcare, finance, and autonomous systems, the need for explainable AI (XAI) has never been greater. One of the most widely used tools in XAI is the saliency map,

FRIES: A Groundbreaking Framework for Inconsistency Estimation of Saliency Metrics Read More »

Discover RETTA: the first retrieval-enhanced test-time adaptation framework for zero-shot video captioning.

RETTA: Retrieval-Enhanced Test-Time Adaptation for Zero-Shot Video Captioning

2 Comments / Machine Learning /

RETTA: Revolutionizing Zero-Shot Video Captioning with Retrieval-Enhanced Test-Time Adaptation In the rapidly evolving field of vision-language modeling, the ability to automatically generate accurate and contextually relevant descriptions of video content—known as video captioning—has become a cornerstone for applications ranging from assistive technology for the visually impaired to intelligent video search engines. While supervised models have

RETTA: Retrieval-Enhanced Test-Time Adaptation for Zero-Shot Video Captioning Read More »

Discover DeepSPV—the first deep learning pipeline to estimate 3D spleen volume from 2D ultrasound

DeepSPV: Revolutionizing 3D Spleen Volume Estimation from 2D Ultrasound with AI

7 Comments / Machine Learning /

In the rapidly evolving field of medical imaging, accurate and non-invasive assessment of organ size is critical—especially when managing chronic conditions like sickle cell disease (SCD) and liver disorders, where splenomegaly (enlarged spleen) is a common clinical indicator. Traditionally, clinicians rely on manual measurements from 2D ultrasound (US) images, which are quick and accessible but

DeepSPV: Revolutionizing 3D Spleen Volume Estimation from 2D Ultrasound with AI Read More »

Visual comparison of feature clustering with CE vs. SuperCM using t-SNE plots on CIFAR-10, SVHN, and MNIST datasets—showing tighter, more separated clusters with SuperCM."

7 Shocking Ways SuperCM Boosts Accuracy (And 1 Fatal Flaw You Must Avoid)

Leave a Comment / Machine Learning /

In the world of machine learning, semi-supervised learning (SSL) and unsupervised domain adaptation (UDA) are game-changers—especially when labeled data is scarce or expensive to obtain. But what if you could supercharge these models with a simple yet powerful technique? Enter SuperCM, a novel framework introduced in a groundbreaking 2025 Pattern Recognition paper that’s turning heads

7 Shocking Ways SuperCM Boosts Accuracy (And 1 Fatal Flaw You Must Avoid) Read More »

Infographic showing Gauging-β algorithm workflow: border detection, hierarchical clustering, and reassignment of points for superior data separation.

7 Revolutionary Clustering Breakthroughs: Why Gauging-β Outperforms (And When It Fails)

Leave a Comment / Machine Learning /

In the rapidly evolving world of machine learning and data science, clustering algorithms are the backbone of unsupervised learning. Yet, despite decades of research, many algorithms still struggle with non-convex shapes, overlapping clusters, and sensitivity to parameters. Enter Gauging-β — a powerful new algorithm that redefines how we approach data clustering by intelligently identifying and

7 Revolutionary Clustering Breakthroughs: Why Gauging-β Outperforms (And When It Fails) Read More »

Diagram showing transported velocity fields transforming cell shape sequences into Euclidean time series for advanced analysis.

7 Revolutionary Breakthroughs in Cell Shape Analysis: How a Powerful New Model Outshines Old Methods

2 Comments / Machine Learning /

In the fast-evolving world of biomedical research and artificial intelligence, understanding cell motility—how cells move and change shape—is critical for unlocking secrets behind cancer metastasis, immune responses, and developmental biology. Yet, traditional methods have long struggled to accurately model the complex dynamics of cellular shapes over time. Now, a groundbreaking study titled “Time-series analysis of

7 Revolutionary Breakthroughs in Cell Shape Analysis: How a Powerful New Model Outshines Old Methods Read More »

Illustration showing a futuristic AI-powered medical imaging analyzing a brain MRI, with digital neural network pathways glowing in blue, symbolizing the Recurrent Inference Image Registration (RIIR) process.

7 Revolutionary Breakthroughs in AI Medical Imaging: The Good, the Bad, and the Future of RIIR

1 Comment / Machine Learning /

In the rapidly evolving world of medical imaging, a groundbreaking new technology is emerging that promises to redefine how doctors align and analyze patient scans. Meet the Recurrent Inference Image Registration (RIIR) network—a revolutionary deep learning framework that’s not only faster and more accurate than traditional methods but also works with dramatically less data. This

7 Revolutionary Breakthroughs in AI Medical Imaging: The Good, the Bad, and the Future of RIIR Read More »

Illustration of CONFIDERAI score function analyzing overlapping decision rules in a 2D feature space, highlighting high-risk prediction zones and conformal critical sets for trustworthy AI.

5 Revolutionary Breakthroughs in AI Safety: How CONFIDERAI Eliminates Prediction Failures While Boosting Trust (But Watch Out for Hidden Risks)

Leave a Comment / Machine Learning /

In the rapidly evolving world of artificial intelligence, one question looms larger than ever: Can we truly trust AI systems when lives are on the line? From detecting DNS tunneling attacks to predicting cardiovascular disease, the stakes have never been higher. While explainable AI (XAI) has made strides in transparency, a critical gap remains —

5 Revolutionary Breakthroughs in AI Safety: How CONFIDERAI Eliminates Prediction Failures While Boosting Trust (But Watch Out for Hidden Risks) Read More »

Submillimeter Diffusion MRI With DnSPIRiT Reconstruction

Submillimeter Diffusion MRI With DnSPIRiT Reconstruction

3 Comments / Machine Learning, Medical AI /

Analysis by the aitrendblend editorial team  ·  Pillar, AI for medical imaging and healthcare  ·  Reading time about 14 min submillimeter dMRI DnSPIRiT 3D multislab EPI gyral bias U fibers tractography Pulseq Submillimeter Diffusion MRI: Higher spatial resolution changes which white matter pathways a tractography algorithm can even see. Replace this feature image before publishing.

Submillimeter Diffusion MRI With DnSPIRiT Reconstruction Read More »