Advanced Imaging Modalities
Expert-defined terms from the Professional Certificate in AI in Medical Imaging course at UK School of Management. Free to read, free to share, paired with a globally recognised certification pathway.
Advanced Imaging Modalities #
Innovative medical imaging techniques that offer enhanced diagnostic accuracy and precision compared to traditional methods. They include Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), and Single-Photon Emission Computed Tomography (SPECT).
Artificial Intelligence (AI) #
A branch of computer science that focuses on creating intelligent machines that can learn from data and make decisions like humans. In medical imaging, AI can be used to automate image analysis, detect abnormalities, and assist in diagnosis.
Computed Tomography (CT) #
A non-invasive imaging technique that uses X-rays to create detailed cross-sectional images of the body. CT scans can provide high-resolution 3D images of internal organs, bones, and soft tissues.
Deep Learning #
A subset of AI that uses artificial neural networks with multiple layers to learn and make decisions. Deep learning algorithms can analyze large datasets and extract complex features, making them ideal for medical image analysis.
Magnetic Resonance Imaging (MRI) #
A non-ionizing imaging technique that uses magnetic fields and radio waves to create detailed images of the body's internal structures. MRI is particularly useful for imaging soft tissues, such as the brain, spinal cord, and muscles.
Natural Language Processing (NLP) #
A field of AI that focuses on enabling computers to understand, interpret, and generate human language. In medical imaging, NLP can be used to extract information from radiology reports and assist in diagnosis.
Positron Emission Tomography (PET) #
A nuclear medicine imaging technique that uses radioactive tracers to visualize and measure metabolic activity in the body. PET scans can detect changes in cellular function, making them useful for detecting cancer, neurological disorders, and heart disease.
Radiomic Features #
Quantitative features extracted from medical images using data-characterization algorithms. Radiomic features can include texture, shape, and intensity, and can provide valuable information for diagnosis, prognosis, and treatment planning.
Radiomics #
The extraction and analysis of large amounts of quantitative features from medical images. Radiomics can provide valuable insights into disease characteristics and treatment response, and can assist in personalized medicine.
Radiology Report #
A document that summarizes the findings of a medical imaging examination. Radiology reports are typically written by radiologists and include information about the patient's medical history, the imaging technique used, and the results of the examination.
Single #
Photon Emission Computed Tomography (SPECT): A nuclear medicine imaging technique that uses gamma cameras to create 3D images of the body's internal structures. SPECT scans can provide functional information about organs and tissues, making them useful for detecting cancer, neurological disorders, and heart disease.
Supervised Learning #
A type of machine learning that uses labeled data to train algorithms. In supervised learning, the algorithm is provided with input data and the corresponding output labels, and learns to map inputs to outputs.
Texture Analysis #
A radiomic feature that describes the spatial arrangement of pixel intensities in an image. Texture analysis can provide information about tissue structure and composition, and can assist in diagnosis and treatment planning.
Unsupervised Learning #
A type of machine learning that uses unlabeled data to train algorithms. In unsupervised learning, the algorithm is not provided with output labels, and must learn to identify patterns and structure in the data on its own.
Volumetric Analysis #
A radiomic feature that describes the volume of a specific region of interest in a medical image. Volumetric analysis can provide information about the size and growth of tumors, and can assist in treatment planning and response assessment.
Convolutional Neural Networks (CNNs) #
A type of deep learning algorithm that is particularly well-suited for image analysis. CNNs use convolutional layers to extract features from images and fully connected layers to classify those features.
Fully Convolutional Networks (FCNs) #
A type of CNN that is designed for semantic segmentation of images. FCNs use transposed convolutional layers to upsample feature maps and generate pixel-wise classifications.
Generative Adversarial Networks (GANs) #
A type of deep learning algorithm that uses two neural networks - a generator and a discriminator - to generate realistic images. GANs have been used in medical imaging to generate synthetic images for training and augmentation.
Image Registration #
The process of aligning two or more medical images acquired at different times or with different imaging modalities. Image registration can improve the accuracy of diagnosis and treatment planning by enabling the comparison and fusion of image data.
Lesion Segmentation #
The process of identifying and delineating a lesion or tumor in a medical image. Lesion segmentation can provide valuable information for diagnosis, treatment planning, and response assessment.
Magnetic Resonance Spectroscopy (MRS) #
A non-invasive imaging technique that uses magnetic fields and radio waves to measure the concentration of metabolites in the body. MRS can provide information about tissue metabolism and has been used in cancer diagnosis and treatment monitoring.
Multimodal Imaging #
The integration of medical images acquired with different imaging modalities. Multimodal imaging can provide complementary information about the body's internal structures and functions, and can improve the accuracy of diagnosis and treatment planning.
Quantitative Imaging #
The extraction and analysis of quantitative data from medical images. Quantitative imaging can provide valuable information about tissue properties and disease characteristics, and can assist in diagnosis, prognosis, and treatment planning.
Radiogenomics #
The integration of radiomic features and genomic data to improve the accuracy of diagnosis and treatment planning. Radiogenomics can provide insights into the molecular mechanisms of disease and has been used in cancer research.
Transfer Learning #
The use of pre-trained deep learning models for medical image analysis. Transfer learning can save time and resources by leveraging the knowledge gained from training on large datasets in other domains.
U #
Net: A type of CNN that is widely used for biomedical image segmentation. U-Net uses a symmetric encoder-decoder architecture with skip connections between the encoder and decoder layers, enabling precise segmentation of small structures.
V #
Net: A type of 3D CNN that is widely used for medical image segmentation. V-Net uses a similar architecture to U-Net but with 3D convolutional layers, enabling the segmentation of volumetric data.
Z #
Score Normalization: A normalization technique that transforms medical images to have zero mean and unit variance. Z-score normalization can improve the accuracy of image analysis by reducing variability in image intensity.
Contrast Agent #
A substance that is administered to a patient to enhance the contrast of medical images. Contrast agents can improve the visibility of specific structures or tissues, and can assist in diagnosis and treatment planning.
Dose Exposure #
The amount of radiation that a patient is exposed to during a medical imaging examination. Dose exposure should be minimized to reduce the risk of radiation-induced damage.
Fusion Imaging #
The integration of medical images acquired with different imaging modalities to create a single composite image. Fusion imaging can provide complementary information about the body's internal structures and functions, and can improve the accuracy of diagnosis and treatment planning.
Image Reconstruction #
The process of generating a medical image from raw data. Image reconstruction algorithms can improve the quality and accuracy of medical images by reducing noise and artifacts.
Intravenous Contrast #
A contrast agent that is administered intravenously to enhance the contrast of medical images. Intravenous contrast is commonly used in CT and MRI examinations.
Ionizing Radiation #
Radiation that is capable of ionizing atoms and molecules. Ionizing radiation is used in medical imaging techniques such as CT and X-ray examinations.
Medical Image Analysis #
The process of extracting quantitative data and features from medical images. Medical image analysis can provide valuable information for diagnosis, prognosis, and treatment planning.
Non #
I