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
Comment
. 2023 Jan;50(2):352-375.
doi: 10.1007/s00259-022-06001-6. Epub 2022 Nov 3.

Joint EANM/SNMMI guideline on radiomics in nuclear medicine : Jointly supported by the EANM Physics Committee and the SNMMI Physics, Instrumentation and Data Sciences Council

M Hatt  1 , A K Krizsan  2 , A Rahmim  3 , T J Bradshaw  4 , P F Costa  5 , A Forgacs  2 , R Seifert  6   7 , A Zwanenburg  8   9   10 , I El Naqa  11 , P E Kinahan  12 , F Tixier  1 , A K Jha  13 , D Visvikis  1
Affiliations
Comment

Joint EANM/SNMMI guideline on radiomics in nuclear medicine : Jointly supported by the EANM Physics Committee and the SNMMI Physics, Instrumentation and Data Sciences Council

M Hatt et al. Eur J Nucl Med Mol Imaging. 2023 Jan.

Abstract

Purpose: The purpose of this guideline is to provide comprehensive information on best practices for robust radiomics analyses for both hand-crafted and deep learning-based approaches.

Methods: In a cooperative effort between the EANM and SNMMI, we agreed upon current best practices and recommendations for relevant aspects of radiomics analyses, including study design, quality assurance, data collection, impact of acquisition and reconstruction, detection and segmentation, feature standardization and implementation, as well as appropriate modelling schemes, model evaluation, and interpretation. We also offer an outlook for future perspectives.

Conclusion: Radiomics is a very quickly evolving field of research. The present guideline focused on established findings as well as recommendations based on the state of the art. Though this guideline recognizes both hand-crafted and deep learning-based radiomics approaches, it primarily focuses on the former as this field is more mature. This guideline will be updated once more studies and results have contributed to improved consensus regarding the application of deep learning methods for radiomics. Although methodological recommendations in the present document are valid for most medical image modalities, we focus here on nuclear medicine, and specific recommendations when necessary are made for PET/CT, PET/MR, and quantitative SPECT.

Keywords: Deep learning; Machine learning; Nuclear medicine; Radiomics; Recommendations.

PubMed Disclaimer

Conflict of interest statement

M. Hatt declares that he has no conflict of interest, A. K. Krizsan works as Medical Physicist at Scanomed Nuclear Medicine Center Debrecen, Hungary, a subsidiary company of Mediso Medical Imaging Systems, Budapest, Hungary, A. Rahmim declares that he has no conflict of interest, T. J. Bradshaw receives research support from GE Healthcare, P. F. Costa declares that he has no conflict of interest, A. Forgacs holds Product Manager position at Mediso Medical Imaging Systems, Budapest, Hungary. In addition, he works as Chief Medical Phyicist at Scanomed Nuclear Medicine Center Debrecen, Hungary, a subsidiary company of Mediso Medical Imaging Systems, Budapest, Hungary, R. Seifert declares that he has no conflict of interest, A. Zwanenburg declares that he has no conflict of interest, I. El Naqa is on the Advisory board of Endectra LLC and received the following NIH grants/contracts: R01 CA233487, R37 CA222215, 5N92020D00018/75N92020F0001, and DOD: W81XWH2210277, P. Kinahan received the following grant: NCI grant P50 CA228944, F. Tixier declares that he has no conflict of interest, A.K. Jha declares that he has no conflict of interest, D. Visvikis declares that he has no conflict of interest.

Figures

Fig. 1
Fig. 1
Top part illustrates the typical standard radiomics workflow, whereas the bottom part illustrates two different (among a myriad of possibilities) use of deep neural networks: direct training of a network using the input images or using a pre-trained network for extracting additional/alternative features from segmented tumor
See this image and copyright information in PMC

Comment on

References

    1. Doi K. Computer-aided diagnosis in medical imaging: historical review, current status and future potential. Comput Med Imaging Graph. 2007;31:198–211. doi: 10.1016/j.compmedimag.2007.02.002. - DOI - PMC - PubMed
    1. Finette S, Bleier A, Swindell W. Breast tissue classification using diagnostic ultrasound and pattern recognition techniques: I. Methods of pattern recognition. Ultrason Imaging. 1983;5:55–70. doi: 10.1177/016173468300500106. - DOI - PubMed
    1. Kuo MD, Gollub J, Sirlin CB, Ooi C, Chen X. Radiogenomic analysis to identify imaging phenotypes associated with drug response gene expression programs in hepatocellular carcinoma. J Vasc Interv Radiol. 2007;18:821–830. doi: 10.1016/j.jvir.2007.04.031. - DOI - PubMed
    1. Segal E, Sirlin CB, Ooi C, Adler AS, Gollub J, Chen X, et al. Decoding global gene expression programs in liver cancer by noninvasive imaging. Nat Biotechnol Nature Publishing Group. 2007;25:675–680. doi: 10.1038/nbt1306. - DOI - PubMed
    1. El Naqa I, Grigsby P, Apte A, Kidd E, Donnelly E, Khullar D, et al. Exploring feature-based approaches in PET images for predicting cancer treatment outcomes. Pattern Recognit. 2009;42:1162–1171. doi: 10.1016/j.patcog.2008.08.011. - DOI - PMC - PubMed

LinkOut - more resources