AutoML Systems For Medical Imaging
Tasmia Tahmida Jidney, Angona Biswas, MD Abdullah Al Nasim, Ismail Hossain, Md Jahangir Alam, Sajedul Talukder, Mofazzal Hossain, Md Azim Ullah
TL;DR
This work surveys AutoML for medical imaging, addressing the challenge of rapidly building ML models given the explosion of imaging data. It analyzes techniques such as automated feature engineering, automated hyperparameter optimization, and Neural Architecture Search, and maps them to imaging tasks like diagnosis, segmentation, registration, synthesis, augmentation, and GAN-based generation. Key contributions include a taxonomy of applications, a synthesis of challenges (data quality, privacy, heterogeneity, interpretability, evaluation, and transparency), and a discussion of future directions for clinical integration. The study highlights the potential of AutoML to accelerate clinical decision-making and broaden access to AI-powered imaging, while underscoring the need for robust validation and careful workflow integration to ensure safe, effective deployment.
Abstract
The integration of machine learning in medical image analysis can greatly enhance the quality of healthcare provided by physicians. The combination of human expertise and computerized systems can result in improved diagnostic accuracy. An automated machine learning approach simplifies the creation of custom image recognition models by utilizing neural architecture search and transfer learning techniques. Medical imaging techniques are used to non-invasively create images of internal organs and body parts for diagnostic and procedural purposes. This article aims to highlight the potential applications, strategies, and techniques of AutoML in medical imaging through theoretical and empirical evidence.