{"doi":"10.1002/mp.16099","title":"The emergence of PET/CT: Engineering, innovation, and usage","abstract":"The first prototype combined positron emission tomography/x-ray computed tomography (PET/CT) scanner was built at the University of Pittsburgh Medical Center in 1998. This had a major impact on nuclear medicine and its role in radiology. It is possible to look at the emergence of PET/CT in three ways: engineering, innovation, and usage. The role of medical physics was present in all of these. The high signal-to-noise ratio of many solid cancers imaged with 18F-FDG PET leads to advantages in the conspicuity of tumors compared to other modalities. However, it was recognized in the early 1990s1 that the location could be difficult to precisely determine due to the relatively low resolution and high noise inherent in PET imaging as compared to CT or magnetic resonance imaging (MRI). Small shifts in the estimated location, for example, if a lesion is separate from, or invading into, an organ or tissue boundary, can cause major changes in the diagnosis, staging, treatment decisions, and expected outcomes. For this reason, considerable attention was focused on the development of image registration methods where the anatomic detail of CT or MRI was fused with the functional information uniquely available from PET. Although this technique can be used successfully in the brain, it is logistically challenging in the body for multiple reasons, including differences in relative anatomic positions between PET and CT scans from external positioning differences, differences in bowel gas distribution, changes in hydration, and other reasons. During this time in the early 1990s, David Townsend was working at the University Hospital Geneva in Switzerland in collaboration with CTI Inc. of Knoxville, an early manufacturer of PET scanners, and was developing a prototype of a low-cost PET scanner called the Advanced Rotating Tomograph (ART)2 (Figure 1). The ART scanner reduced costs by having only two partial segments of the detector ring. The two opposed banks of detectors covering approximately one-third of a conventional full ring, which rotated to acquire tomographic data as illustrated in Figure 1. The so-called “fully-3D” acquisition, that is, without slice-separating septa, gave this tomograph sensitivity comparable to a conventional full ring tomograph operated in 2D mode. The transition from 2D imaging to higher sensitivity, but more complex, fully-3D imaging in PET was another transition that was underway during this period. The 3D-Reprojection (3DRP) algorithm3 became the industry standard in the mid-1990s, and also led to my visiting David Townsend in Geneva to see the ART scanner, which implemented the 3DRP algorithm. I visited independently using my own funds while I was still a graduate student at the University of Pennsylvania. My PhD advisor, Joel Karp, later graciously agreed to reimburse my expenses. In retrospect, I would be challenged to supervise the graduate student that I once was, and I am grateful for the flexibility and patience that Joel had, in addition to his mentorship in research. These interactions eventually led to my first job as a medical physicist with David Townsend when he relocated to the University of Pittsburgh. Prior to departing University Hospital Geneva, an encounter that David Townsend later related was that when he was showing the ART scanner with the covers off to another faculty member, they remarked that there was enough space between the PET detector blocks to install a CT scanner. At the University of Pittsburgh, we, along with our then graduate student Thomas Beyer, submitted an R01 grant to NIH in 1995 to build a prototype rotating PET/CT. On the second try, the grant was funded, and while the R01 grant was not enough to pay for all the costs, it was enough to start the project in partnership with CTI. Figure 1 shows one of the early concept drawings. The PET and CT configuration turned out to be impractical for two major reasons discussed below. However, the figure was a useful conceptual tool ","journal":"Medical Physics","year":2022,"id":287847,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":3,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.953,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2022-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":65783,"name":"Paul E. Kinahan","orcid":"0000-0001-6461-3306","position":0,"is_corresponding":true}],"reference_count":9,"raw_metadata":null,"created_at":"2026-07-19T00:30:11.069895Z","pmid":"36378931","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}