WIMSP 2026

 Overview

Artificial intelligence has become a transformative force in healthcare, enabling systems that can interpret medical images and biosignals with accuracy comparable to, and in some cases exceeding, that of clinical experts. However, the development of reliable, safe, and equitable AI tools for medical applications remains an open research challenge that demands dedicated interdisciplinary forums.

WIMSP 2026 (Workshop on Intelligent Medical Imaging and Signal Processing) aims to be an intimate and focused forum for researchers, clinicians, and engineers to present and discuss advances at the intersection of artificial intelligence and clinical data analysis. The workshop targets two complementary domains:

1. Medical imaging: automated analysis of radiological scans (X-ray, CT, MRI, ultrasound), histopathological slides, retinal fundus images, and endoscopic video.

2. Biomedical signal processing: AI-driven interpretation of physiological time series such as ECG, EEG, EMG, and photoplethysmography (PPG).

  Objetives

1. 1. Provide a dedicated venue within MICAI for work at the boundary of AI and biomedicine, a growing area that currently lacks a dedicated forum in Mexico's leading AI conference.

   2. Promote the exchange of methodological advances, including deep learning architectures, foundation models, graph neural networks, and uncertainty quantification, applied to medical data.

   3. Encourage interdisciplinary collaboration between computer scientists, biomedical engineers, and medical practitioners from Mexico, Latin America, and abroad.

   4. Highlight challenges specific to clinical deployment: limited annotated data, domain shift across hospitals and devices, regulatory requirements, and explainability.

   5. Foster the development of a research community around medical AI in Mexico, connecting students and early-career researchers with established international experts.

  Topics

Medical Image Analysis

• Segmentation, detection, and classification of anatomical structures and lesions

• Deep learning for radiology: X-ray, CT, MRI, and PET image analysis

• Computational histopathology and whole-slide image analysis

• Retinal fundus image analysis and ophthalmological AI

• Endoscopy and surgical video understanding

• 3D volumetric analysis and reconstruction

• Image registration and multi-modal fusion

• Generative models and synthetic data for medical imaging (GANs, diffusion models)

Biomedical Signal Processing

• Deep learning for ECG arrhythmia detection and cardiovascular monitoring

• EEG-based brain-computer interfaces (BCI) and neurological disorder diagnosis

• EMG signal analysis for prosthetics and rehabilitation

• Photoplethysmography (PPG) and wearable biosignal processing

• Multimodal physiological signal fusion

• Foundation models and self-supervised learning for biosignals

Methodological and Translational Challenges

• Transfer learning, domain adaptation, and domain generalization in clinical settings

• Federated learning and privacy-preserving methods for medical data

• Explainability, interpretability, and uncertainty quantification in medical AI

• Fairness and bias mitigation in clinical AI systems

• Annotation-efficient learning: semi-supervised, weakly supervised, and self-supervised approaches

• AI deployment in resource-limited healthcare environments

• Benchmarking, datasets, and evaluation protocols for medical AI

  Poster CFP