Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2020 Oct;9(19):7172-7182.
doi: 10.1002/cam4.3386. Epub 2020 Aug 18.

Variability in accuracy of prostate cancer segmentation among radiologists, urologists, and scientists

Michael Y Chen  1   2   3   4 , Maria A Woodruff  1 , Prokar Dasgupta  5 , Nicholas J Rukin  1   2   3   4
Affiliations
Comparative Study

Variability in accuracy of prostate cancer segmentation among radiologists, urologists, and scientists

Michael Y Chen et al. Cancer Med. 2020 Oct.

Abstract

Background: There is increasing research in using segmentation of prostate cancer to create a digital 3D model from magnetic resonance imaging (MRI) scans for purposes of education or surgical planning. However, the variation in segmentation of prostate cancer among users and potential inaccuracy has not been studied.

Methods: Four consultant radiologists, four consultant urologists, four urology trainees, and four nonclinician segmentation scientists were asked to segment a single slice of a lateral T3 prostate tumor on MRI ("Prostate 1"), an anterior zone prostate tumor MRI ("Prostate 2"), and a kidney tumor computed tomography (CT) scan ("Kidney"). Time taken and self-rated subjective accuracy out of a maximum score of 10 were recorded. Root mean square error, Dice coefficient, Matthews correlation coefficient, Jaccard index, specificity, and sensitivity were calculated using the radiologists as the ground truth.

Results: There was high variance among the radiologists in segmentation of Prostate 1 and 2 tumors with mean Dice coefficients of 0.81 and 0.58, respectively, compared to 0.96 for the kidney tumor. Urologists and urology trainees had similar accuracy, while nonclinicians had the lowest accuracy scores for Prostate 1 and 2 tumors (0.60 and 0.47) but similar for kidney tumor (0.95). Mean sensitivity in Prostate 1 (0.63) and Prostate 2 (0.61) was lower than specificity (0.92 and 0.93) suggesting under-segmentation of tumors in the non-radiologist groups. Participants spent less time on the kidney tumor segmentation and self-rated accuracy was higher than both prostate tumors.

Conclusion: Segmentation of prostate cancers is more difficult than other anatomy such as kidney tumors. Less experienced participants appear to under-segment models and underestimate the size of prostate tumors. Segmentation of prostate cancer is highly variable even among radiologists, and 3D modeling for clinical use must be performed with caution. Further work to develop a methodology to maximize segmentation accuracy is needed.

Keywords: 3D model; 3D printing; MRI; prostate; segmentation.

PubMed Disclaimer

Conflict of interest statement

The authors wish to declare no conflict of interest.

Figures

FIGURE 1
FIGURE 1
A comparison of using threshold density to initiate the segmentation process. On CT imaging of bone (A) the high density allows for rapid segmentation from surrounding soft tissue. To segment the kidney on CT (B), other soft tissues such as spleen are included but the kidney is separated from surrounding adipose tissue. To segment prostate on MRI (C) the rectum and capsular tissue are included, requiring additional manual segmentation. Screenshots taken on Mimics 21.0 (Materialise, Leuven, Belgium). The threshold for bone is predefined by the software, while the others were manually selected
FIGURE 2
FIGURE 2
An example of how the point comparison method works to calculate the distance between two points which is then averaged across the whole model using the root mean square (RMS) calculation method
FIGURE 3
FIGURE 3
Segmentation results of Prostate 1 from 16 participants with time taken and self‐rated accuracy
FIGURE 4
FIGURE 4
Segmentation results of Prostate 2 from 16 participants with time taken and self‐rated accuracy
FIGURE 5
FIGURE 5
Segmentation results of Kidney from 16 participants with time taken and self‐rated accuracy
See this image and copyright information in PMC

References

    1. Cacciamani GE, Okhunov Z, Meneses AD, et al. Impact of three‐dimensional printing in urology: state of the art and future perspectives. a systematic review by ESUT‐YAUWP group. Eur Urol. 2019;76(2):209‐221. - PubMed
    1. Chen MY, Skewes J, Desselle M, et al. Current applications of three‐dimensional printing in urology. BJU Int. 2020;125(1):17‐27. - PubMed
    1. Ebbing J, Jäderling F, Collins JW, et al. Comparison of 3D printed prostate models with standard radiological information to aid understanding of the precise location of prostate cancer: a construct validation study. PLoS One. 2018;13(6):e0199477. - PMC - PubMed
    1. Porpiglia F, Bertolo R, Checcucci E, et al. Development and validation of 3D printed virtual models for robot‐assisted radical prostatectomy and partial nephrectomy: urologists' and patients' perception. World J Urol. 2018;36(2):201‐207. - PubMed
    1. Priester A, Natarajan S, Le JD, et al. A system for evaluating magnetic resonance imaging of prostate cancer using patient‐specific 3D printed molds. Am J Clin Exp Urol. 2014;2(2):127‐135. - PMC - PubMed

Publication types