Machine learning in the differentiation of follicular lymphoma from diffuse large B-cell lymphoma with radiomic [18F]FDG PET/CT features

Jesus, Filipe Montes de; Yin, Yunchao; Mantzorou-Kyriaki, E; Kahle, X U; Haas, Robbert J. de; Yakar, Derya; Glaudemans, Andor W J M; Noordzij, Walter; Kwee, Thomas C.; Nijland, Marcel

doi:10.1007/s00259-021-05626-3

Cited by 14 publications

(16 citation statements)

References 50 publications

(71 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[23,24] Meanwhile, it is found that the radiomics based on PET-CT has a certain application value in the diagnosis of lymphoma subtypes, and it could preliminarily distinguish FL from DLBCL. [25] What surprised us is that radiomics features in our study didn't show a satisfactory result, with the AUC of 0.666, and sensitivity, speci city, accuracy of 74.1%, 55.7%, 59.7%, respectively.…”

Section: Discussioncontrasting

confidence: 55%

Differential diagnosis value of ultrasound and radiomics features in B-cell and T-cell lymphoma

Zhou

Yuan

Deng

et al. 2022

Preprint

View full text Add to dashboard Cite

Objectives One of the challenges in the diagnosis of composite lymphoma is the selection of the lymph node adequately reflect heterogeneity for biopsy. This study evaluated whether ultrasound and radiomics may help differentiating B-cell and T-cell lymphoma. Materials and Methods This is a retrospective study and a total of 124 patients with B-cell lymphoma(n = 97) or T-cell lymphoma(n = 27) were enrolled in. Each patient went through ultrasound examinations including two dimensional ultrasound (2DUS) and contrast-enhanced ultrasound (CEUS). Radiomic features were extracted from gray-scale ultrasound images and then selected. A radiomics score (Rad-score) was calculated via a linear combination. Performances of differential diagnosis of ultrasound, radiomics and the combination were compared with each other. Results Three ultrasound features (hilum, echo characteristic and PI of TIC analysis) and three radiomics features (texture features) were finally chosen as statistically significant. The AUC of ultrasound based on the three features was 0.829, with a sensitivity, specificity, and accuracy of 92.6%,63.9%, and 70.2%, respectively. The AUC of radiomics based on Rad-score was 0.666, with a sensitivity, specificity, and accuracy of 74.1%, 55.7%, and 59.7%, respectively. By comparison, ultrasound combined with radiomics showed greatest diagnostic value in distinguishing B-cell and T-cell lymphoma, with the AUC of 0.858, and a sensitivity, specificity, accuracy of 92.6%, 74.2%, 78.2%. Conclusions Ultrasound combined with radiomics are of great diagnostic value for differentiating B-cell and T-cell lymphoma and enables us to select a lymph node that adequately reflect heterogeneity for biopsy and avoid missed diagnosis.

show abstract

Section: Discussioncontrasting

confidence: 55%

Differential diagnosis value of ultrasound and radiomics features in B-cell and T-cell lymphoma

Zhou

Yuan

Deng

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…A total of 24 articles on lymphomas were included in this review [ 122 , 123 , 124 , 125 , 126 , 127 , 128 , 129 , 130 , 131 , 132 , 133 , 134 , 135 , 136 , 137 , 138 , 139 , 140 , 141 , 142 , 143 , 144 , 145 ], 13 of which studying diffuse large B-cell lymphoma (including 2 studies on gastro-intestinal lymphoma), 3 on follicular lymphoma, 3 on Hodgkin’s lymphoma, 2 on mantle cell lymphoma and 3 on other sub-types of lymphoma. 18F-FDG was the only tracer employed and all studies built radiomic models on baseline, pre-treatment PET images, often including clinical parameters and international prognostic indices.…”

Section: Resultsmentioning

confidence: 99%

“…Several studies evaluated the role of PET radiomics and AI analysis in blood malignancies [ 122 , 123 , 124 , 125 , 126 , 127 , 128 , 129 , 130 , 131 , 132 , 133 , 134 , 135 , 136 , 137 , 138 , 139 , 140 , 141 , 142 , 143 , 144 , 145 ]. Lymphoma radiomics on 18F-FDG PET/CT was more often assessed on patients with DLBCL, in retrospective cohorts and for prognostic purposes.…”

Section: Discussionmentioning

confidence: 99%

Radiomics in Oncological PET Imaging: A Systematic Review—Part 2, Infradiaphragmatic Cancers, Blood Malignancies, Melanoma and Musculoskeletal Cancers

et al. 2022

View full text Add to dashboard Cite

The objective of this review was to summarize published radiomics studies dealing with infradiaphragmatic cancers, blood malignancies, melanoma, and musculoskeletal cancers, and assess their quality. PubMed database was searched from January 1990 to February 2022 for articles performing radiomics on PET imaging of at least 1 specified tumor type. Exclusion criteria includd: non-oncological studies; supradiaphragmatic tumors; reviews, comments, cases reports; phantom or animal studies; technical articles without a clinically oriented question; studies including <30 patients in the training cohort. The review database contained PMID, first author, year of publication, cancer type, number of patients, study design, independent validation cohort and objective. This database was completed twice by the same person; discrepant results were resolved by a third reading of the articles. A total of 162 studies met inclusion criteria; 61 (37.7%) studies included >100 patients, 13 (8.0%) were prospective and 61 (37.7%) used an independent validation set. The most represented cancers were esophagus, lymphoma, and cervical cancer (n = 24, n = 24 and n = 19 articles, respectively). Most studies focused on 18F-FDG, and prognostic and response to treatment objectives. Although radiomics and artificial intelligence are technically challenging, new contributions and guidelines help improving research quality over the years and pave the way toward personalized medicine.

show abstract

“…Lippi et al related good performance of ML to discriminate different types of lymphomas from each other, especially HL, but in a small population of patients (11). Recently, de Jesus et al showed very promising results to differentiate follicular lymphoma and DLBCL using radiomics and ML classifier in a population of 120 patients, which could have important clinical use when monitoring patients for aggressive transformation (14). Their best performing model showed an AUC of 0.86 significantly higher than the performance of the SUVmaxbased model (AUC 0.79).…”

Section: Discussionmentioning

confidence: 99%

“…The development of artificial intelligence and machine learning (ML) combined with radiomics has gained popularity in different medical imaging tasks, including lesion identification and characterization. In lymphoma, some studies showed the potential of [ 18 F]FDG PET/CT radiomics to differentiate lymphoma from other types of cancers and to differentiate different types of lymphoma (10)(11)(12)(13)(14). To the best of our knowledge, no study has yet explored the use of 18 F-FDG PET/CT radiomics to characterize sarcoidosis lesions, except one for the diagnosis of cardiac involvement (15).…”

Section: Introductionmentioning

confidence: 99%

Distinction of Lymphoma from Sarcoidosis on¹⁸F-FDG PET/CT: Evaluation of Radiomics-Feature–Guided Machine Learning Versus Human Reader Performance

et al. 2022

View full text Add to dashboard Cite

Sarcoidosis and lymphoma often share common features on 18 F-FDG PET/CT, such as intense hypermetabolic lesions of lymph nodes and multiple organs. We aimed at developing and validating radiomics signatures to differentiate sarcoidosis from Hodgkin (HL) and diffuse large B-cell (DLBCL) lymphoma. Methods: We retrospectively collected 420 patients (169 sarcoidosis, 140 HL and 111 DLBCL) who underwent a pretreatment 18 F-FDG PET/CT at the University Hospital of Liege. The studies were randomly distributed to 4 physicians who gave their diagnostic suggestion between the 3 diseases. Individual and pooled performances of physicians were then calculated. The inter-observer variability was evaluated using a sample of 34 studies interpreted by all physicians. Volumes of interest (VOI) were delineated over the lesions and the liver using MIM software, and 215 radiomic features were extracted using Radiomics toolbox. Models were developed combining clinical data (age, gender and weight) and radiomics (original and tumor-to-liver TLR radiomics), with 7 different feature selection approaches and 4 different machine learning (ML) classifiers, to differentiate sarcoidosis and lymphomas on both lesion-based and patient-based approaches. Results: For identifying lymphoma vs. sarcoidosis, physicians' pooled sensitivity, specificity, area under the curve (AUC) and accuracy were 0.

show abstract

Machine learning in the differentiation of follicular lymphoma from diffuse large B-cell lymphoma with radiomic [18F]FDG PET/CT features

Abstract: M. (2022). Machine learning in the differentiation of follicular lymphoma from diffuse large B-cell lymphoma with radiomic [F-18]FDG PET/CT features. European Journal of Nuclear Medicine and Molecular Imaging.

Cited by 14 publications

References 50 publications

Differential diagnosis value of ultrasound and radiomics features in B-cell and T-cell lymphoma

Differential diagnosis value of ultrasound and radiomics features in B-cell and T-cell lymphoma

Radiomics in Oncological PET Imaging: A Systematic Review—Part 2, Infradiaphragmatic Cancers, Blood Malignancies, Melanoma and Musculoskeletal Cancers

Distinction of Lymphoma from Sarcoidosis on¹⁸F-FDG PET/CT: Evaluation of Radiomics-Feature–Guided Machine Learning Versus Human Reader Performance

Contact Info

Product

Resources

About

Machine learning in the differentiation of follicular lymphoma from diffuse large B-cell lymphoma with radiomic [18F]FDG PET/CT features

Abstract: M. (2022). Machine learning in the differentiation of follicular lymphoma from diffuse large B-cell lymphoma with radiomic [F-18]FDG PET/CT features. European Journal of Nuclear Medicine and Molecular Imaging.

Cited by 14 publications

References 50 publications

Differential diagnosis value of ultrasound and radiomics features in B-cell and T-cell lymphoma

Differential diagnosis value of ultrasound and radiomics features in B-cell and T-cell lymphoma

Radiomics in Oncological PET Imaging: A Systematic Review—Part 2, Infradiaphragmatic Cancers, Blood Malignancies, Melanoma and Musculoskeletal Cancers

Distinction of Lymphoma from Sarcoidosis on18F-FDG PET/CT: Evaluation of Radiomics-Feature–Guided Machine Learning Versus Human Reader Performance

Contact Info

Product

Resources

About

Distinction of Lymphoma from Sarcoidosis on¹⁸F-FDG PET/CT: Evaluation of Radiomics-Feature–Guided Machine Learning Versus Human Reader Performance