Year: 2026
Type: book-chapter
Abstract: Nuclear medicine shows function, not structure. A radioactive tracer is given to the patient and the scanner maps where it goes, and hybrid machines add CT or MRI so that the uptake can be placed on the anatomy. The images are built from a limited number of detected photons, so they are noisy and blurred, and every step from reconstruction to measurement involves correction and estimation. These are tasks to which artificial intelligence is well suited. This chapter follows the examination from injection to report. It explains image reconstruction, attenuation correction and the role of the CT or MRI component, the use of deep learning to lower the dose or shorten the scan, and the measurement of uptake with its main source of error, the partial-volume effect. Applications are reviewed in oncology, cardiology, neurology and bone imaging, followed by radionuclide therapy and dosimetry. Simulations computed for the chapter on artificial phantoms illustrate the main ideas, and the final sections deal with pitfalls and the part played by the technologist.
Parent work: Artificial Intelligence in Medical Imaging
Language: English
Subfield: Radiology, Nuclear Medicine and Imaging
Field: Medicine
Domain: Health Sciences
Open Access status: gold
OpenAlex ID: W7223674988