Lilly Kauffman, BA
Johns Hopkins Hospital

Click here to view this module as a video lecture.

Wilhelm Roentgen was a German physicist who was studying emissions generated by electrical currents. It was by accident that on November 8, 1895, he discovered what he called the X-ray, which passed through substances like tissue and blood but left bone visible. He famously used his wife’s hand, with her wedding ring, as a demonstration.

The discoveries were shared publicly in January 1896 and by the next month, X-rays made their debut in medicine.

The X-ray was a huge discovery that changed medicine forever. But did it change it enough?

X-rays are limited to bone and do not get good visuals of soft tissue. Ultrasound was invented in 1956 but doctors still had to perform surgery to see different pathologies. Doctors wanted a less invasive way to see inside patients.

Thus the idea behind the CT scan emerged. Research was done by many people since the early 1900s but nothing was sufficient.

Funding was from Central Research Laboratories of EMI, Ltd. and the British Department of Health and Social Security.

EMI, Ltd. was also a music label that made a lot of money from The Beatles’ record sales. Many like to say that The Beatles funded the first CT Scanner, but it is not completely clear if there was a direct link.

The scanner was developed at Atkinson Morley Hospital in London (now closed). Godfrey Hounsfield, an electrical engineer, is most credited with the creation of the CT scan but he was also helped by Dr. Allan Cormack, a physicist, and Dr. Jamie Ambrose, a neuroradiologist.

Many other people were involved with smaller contributions.

The trio met in 1969 and a working model of the world’s first CT scanner was completed in 1971.

On October 1, 1971, the first living human patient was scanned with the CT scanner – and it was a success. Though pixelated, doctors could see a clear tumor on the scan (which ended up being a cystic astrocytoma).

The findings were shared in April 1972 and CT scanners quickly moved into hospitals around the globe.

Left: a prototype scan on a cadaver
Right: a scan image obtained from the living patient on October 1, 1971

“In New et al’s 1974 description of the first Electric and Musical Industries (EMI) scanner in the United States, the acquisition of one 80 × 80 transverse section required 5 minutes of scanning (1). Current CT scanners can acquire 1200 512 × 512 transverse sections in 1 second, representing an increase in efficiency of 1.5 billion percent.”
Rubin GD. Computed tomography: revolutionizing the practice of medicine for 40 years. Radiology. 2014 Nov;273(2 Suppl):S45-74. doi: 10.1148/radiol.14141356. PMID: 25340438.

Spiral (or helical) CT was invented in 1990. It allows the patient to pass continuously through the CT scanner while images are obtained uninterrupted. Before, scans had a stop-and-start method of obtaining images, which took time and allowed for patient movement to decrease the quality of the scan.

The advent of spiral CT made scans faster and more accurate. At its advent, scans were 4-slice and quickly increased to 16, 32, 64, and to number 256 and beyond.

Dual Energy CT was first introduced in 2006. It uses a second energy field to obtain an additional attenuation measurement. This allows two different materials – such as iodinated blood, plaque, bone, tissue, etc – to be more easy distinguished on a CT scan.

In 1987, the first article on Volume Rendering was published by Dr. Elliot Fishman et al. Pixar’s 3D imaging software, developed by Lucasfilm, was combined with CT scans to create 3D representations of the scans. This technique helps radiologists (and others) better visualize the patient’s anatomy and pathology.

Cinematic Rendering was introduced under the same principal in 2016. In addition to its 3D presentation, Cinematic Rendering also has the ability to show shadows and different organ textures that can help aid the radiologist in a diagnosis.

Photon-counting CT is one of the newest developments in CT. It converts X-ray photons into electrical signals, which can measure photon energy. PCCT can create images with improved spatial resolution, iodine signal, and artifact and noise reduction, while allowing for multi-energy imaging and radiation dose efficiency.

Artificial Intelligence has already made its way into CT scanning but still has a ways to go before it is standard.

There are many uses of AI in CT scanning, but one example is the earlier detection of cancer. Using automated organ segmentation, AI can detect abnormalities of specific organs (pancreas, liver, etc.), such as cysts and textural changes, which may not be yet visible or noticeable to the radiologist’s eye.

Earlier detection means earlier resection.

Three-dimensional photo-realistic rendering of the manual annotation (a) and deep convolutional neural networks prediction (b).



Park S, Chu LC, Fishman EK, Yuille AL, Vogelstein B, Kinzler KW, Horton KM, Hruban RH, Zinreich ES, Fouladi DF, Shayesteh S, Graves J, Kawamoto S. Annotated normal CT data of the abdomen for deep learning: Challenges and strategies for implementation. Diagn Interv Imaging. 2020 Jan;101(1):35-44. doi: 10.1016/j.diii.2019.05.008. Epub 2019 Jul 26. Erratum in: Diagn Interv Imaging. 2020 Jun;101(6):427. PMID: 31358460.

Artificial Intelligence also has the potential to aid radiologists in triaging Emergency Room patients based on the severity of their CT scan findings.

ChatGPT, along with other Large Language Models, has seen an explosion in use with radiology over the past few months.

Uses include generating a differential diagnosis and final diagnosis based on imaging findings and patient history; study recommendations (including modalities and radiation dose); generating radiology reports; and more.