Description
This book constitutes the refereed proceedings of the First International Medical Image Computing in Resource Constrained Settings Workshop & Knowledge Interchange, MIRASOL 2025, held in Conjunction with MICCAI 2025, Daejeon, South Korea on September 27, 2025.
The 31 full papers included in this book were carefully reviewed and selected from 62 submissions. They are organized around three themes that describe insights into the latest advances in machine learning, medical imaging analysis, image segmentation, disease detection, diagnosis, and prognosis, with a focus on global health applications and innovations in the region for real-world clinical impact.
Bidirectional Prototype Contrastive Learning for Domain Adaptive Medical Segmentation.- Are ECGs enough? Deep learning classification of pulmonary embolism using electrocardiograms.- Enhancing and Accelerating Vessel Annotations in Medical Imaging.- AST-n: A Fast Sampling Approach for Low-Dose CT Reconstruction using Diffusion Models.- Deep Ensemble Approach for Enhancing Brain Tumor Segmentation in Resource-Limited Settings.- Non-invasive mean Pulmonary Artery Pressure Prediction using Multi-Modal Feature Fusion of Chest X-ray and ECG.- Large Scale DICOM Compliance Evaluation of Medical Image Data Elements in Low-Resource Settings.- EDGE-KD: Explainability-Driven Guidance for Efficient Knowledge Distillation in Chest X-Ray Classification.- Contour-Guided Segmentation and X-ray Image Validation for Pneumonia Detection using Mobile Deep Learning in Low-Resource Settings.- From Development to Deployment of AI-assisted Telehealth and Screening for vision- and hearing-threatening diseases in resource-constrained settings: Field Observations, Challenges and Way Forward.- Brain Tumor Segmentation in Sub-Sahara Africa with Advanced Transformer and ConvNet Methods: Fine-Tuning, Data Mixing and Ensembling.- Multimodal Fusion for Melanoma Classification Using Dermoscopic Images and Clinical Metadata.- An Empirical Study on Liver Volumetry: Deep Learning-Based Estimation of Tumor and Remnant Liver Volumes for Preoperative Planning.- Towards Trustworthy Breast Ultrasound Segmentation using Monte Carlo Dropout and Deep Ensembles for Epistemic Uncertainty Estimation.- Lightweight 3D U-Net for Brain Tumor Segmentation on CPUs: Enabling Deep Learning in Low-Resource Environments.- Fetal Abdomen-Guided Ultrasound Reconstruction via Autoencoder For Optimal Frame Selection and Segmentation.- Recovering Diagnostic Value: Super-Resolution Aided Echocardiographic Classification in Resource-Constrained Imaging.- SharpXR: Structure-Aware Denoising for Pediatric Chest X-Rays.- Uncertainty-Aware Evaluation of Deep Learning Object Detectors under Scarce and Evolving Test Datasets.- Advancing Fetal Ultrasound Image Quality Assessment in Low-Resource Settings.- LightDenoise-HL: A Compact Attention-Based Denoising Framework with Dual-Frequency Filtering for Endoscopic Imaging.- Accessible Skin Analysis: A Low-Cost Multispectral Imaging System with Skin-Mimicking Phantoms.- Development and Evaluation of an AI-Driven Telemedicine System for Prenatal Healthcare.- CoMViT: An Efficient Vision Backbone for Supervised Classification in Medical Imaging.- Designing AI Algorithms to Suit Local Context.- Global South Health Practitioners' Awareness and Perceptions of Integrating Artificial Intelligence in Radiological Workflows: A Quantitative Nationwide Study.- Attention-Enhanced Deep Learning for Multi-Class Alzheimer s Disease Classification Using Macular OCT Images in Low-Resource Settings.- Resource-Efficient Glioma Segmentation on Sub-Saharan MR.- TB Screening App: Smartphone Imaging of Tuberculin Skin Test Indurations for Latent Tuberculosis Screening in Low-Resource Settings.- Empowering Medical Equipment Sustainability in Low-Resource Settings: An AI-Powered Diagnostic and Support Platform for Biomedical Technicians.- SepsiGraph: A Graph-Based Multimodal Approach for Early Sepsis Prediction in Dynamic Resource-Constrained Clinical Settings.- Clinically-Informed Preprocessing Improves Stroke Segmentation in Low-Resource Settings.
