arXiv subjects

Md Azim Ullah

Publications and source records attributed to Md Azim Ullah.

4 recordsLinked to original sources

arXiv2023

AutoML Systems For Medical Imaging

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.

cs.AI↗

arXiv2023

Introduction of Medical Imaging Modalities

The diagnosis and treatment of various diseases had been expedited with the help of medical imaging. Different medical imaging modalities, including X-ray, Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Nuclear Imaging, Ultrasound, Electrical Impedance Tomography (EIT), and Emerging Technologies for in vivo imaging modalities is presented in this chapter, in addition to these modalities, some advanced techniques such as contrast-enhanced MRI, MR approaches for osteoarthritis, Cardiovascular Imaging, and Medical Imaging data mining and search. Despite its important role and potential effectiveness as a diagnostic tool, reading and interpreting medical images by radiologists is often tedious and difficult due to the large heterogeneity of diseases and the limitation of image quality or resolution. Besides the introduction and discussion of the basic principles, typical clinical applications, advantages, and limitations of each modality used in current clinical practice, this chapter also highlights the importance of emerging technologies in medical imaging and the role of data mining and search aiming to support translational clinical research, improve patient care, and increase the efficiency of the healthcare system.

eess.IV↗

arXiv2023

Active Learning on Medical Image

The development of medical science greatly depends on the increased utilization of machine learning algorithms. By incorporating machine learning, the medical imaging field can significantly improve in terms of the speed and accuracy of the diagnostic process. Computed tomography (CT), magnetic resonance imaging (MRI), X-ray imaging, ultrasound imaging, and positron emission tomography (PET) are the most commonly used types of imaging data in the diagnosis process, and machine learning can aid in detecting diseases at an early stage. However, training machine learning models with limited annotated medical image data poses a challenge. The majority of medical image datasets have limited data, which can impede the pattern-learning process of machine-learning algorithms. Additionally, the lack of labeled data is another critical issue for machine learning. In this context, active learning techniques can be employed to address the challenge of limited annotated medical image data. Active learning involves iteratively selecting the most informative samples from a large pool of unlabeled data for annotation by experts. By actively selecting the most relevant and informative samples, active learning reduces the reliance on large amounts of labeled data and maximizes the model's learning capacity with minimal human labeling effort. By incorporating active learning into the training process, medical imaging machine learning models can make more efficient use of the available labeled data, improving their accuracy and performance. This approach allows medical professionals to focus their efforts on annotating the most critical cases, while the machine learning model actively learns from these annotated samples to improve its diagnostic capabilities.

eess.IV↗

arXiv2023

Scalable Algorithmic Infrastructure for Computation of Social Crowding and Viral Disease Encounters -- mContain Case Study

mContain was developed (and sparsely deployed) by MD2K center at University of Memphis in the early stages of COVID-19 pandemic to help reduce community transmission in Shelby County and Memphis metropolitan area. The application counts and displays the number of daily proximity encounters with other app users. To reduce the chances of entering crowded places, users can see the level of crowding at busy places on a map. If a user and their COVID-19 test provider both agree to share the results of their test, the app can notify other users about possible exposures to COVID-19. The smartphone application collects location and Bluetooth data and sends it to cloud for near real time processing and decisions to be sent back for visualization and interface with the user. The backend algorithmic infrastructure responsible for real time crowd estimation and contact tracing from streaming batch data use open-source cloud analytics platform Cerebral-Cortex. This project concerns about presenting the authors contributions in the algorithmic development, design and implementation of mContain application as part of the entire collaborative project. We describe the mcontain algorithmic infrastructure and major computational challenges encountered when developing and deploying this application for real-life usage. Details of the app can be found in https://mcontain.md2k.org/

cs.SI↗