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

Abstract

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.

The role of artificial intelligence based on PET/CT radiomics in NSCLC: Disease management, opportunities, and challenges

“…A study differentiates sarcoidosis and lymphomas with seven different feature selection approaches and four different ML classifiers. At the lesion level, they found highly accurate signatures to create models to differentiate cancer vs. sarcoidosis (AUC 0.94) and Hodgkin vs. diffuse large B-cell lymphoma (AUC 0.95) ( 36 ). In a more extensive cohort retrospective study ( 37 ), the researchers used a CNN to automatically locate and classify PET uptake patterns in lung cancer and lymphoma.…”

Sarcoidosis and lymphoma mortality risk: An observational study from the Spanish National Registry

“…However, both sarcoid and lymphoma are FDG avid and the distinction between benign and malignant disorders on the basis of a PET scan is difficult. Current reports have shown that this imaging technique can correctly characterize sarcoidosis and lymphoma lesions with very good performance, sometimes supported by machine learning combined with radiomics [ 25 ]. Despite the above, histopathology remains to date the sarcoidosis diagnosis cornerstone, given that correct classification based on imaging can create a diagnostic challenge and misclassification it is still not infrequent and has important consequences [ 1 ].…”

Clinical application of 18F-fluorodeoxyglucose positron emission tomography/computed tomography radiomics-based machine learning analyses in the field of oncology

“…de Jesus et al [ 77 ] reported that the ML model with the gradient boosting algorithm using PET/CT radiomics had a significantly higher AUC in distinguishing DLBCL and follicular lymphoma according to SUVmax (AUC:0.86 vs. 0.79, p < 0.01). Lovinfosse et al [ 78 ] also showed that the ML model with the RF algorithm using clinical data and PET-radiomics had good performance in differentiating DLBCL from HD, with an AUC of 0.95. Further, the authors showed that the constructed ML model with the RF algorithm had good performance in differentiating malignant lymphoma and sarcoidosis, with an AUC of 0.94.…”