Clinical Nuclear Medicine

Feed Category: 

A Machine-Learning Approach Using PET-Based Radiomics to Predict the Histological Subtypes of Lung Cancer

imagePurpose
We sought to distinguish lung adenocarcinoma (ADC) from squamous cell carcinoma using a machine-learning algorithm with PET-based radiomic features.
Methods
A total of 396 patients with 210 ADCs and 186 squamous cell carcinomas who underwent FDG PET/CT prior to treatment were retrospectively analyzed. Four clinical features (age, sex, tumor size, and smoking status) and 40 radiomic features were investigated in terms of lung ADC subtype prediction. Radiomic features were extracted from the PET images of segmented tumors using the LIFEx package. The clinical and radiomic features were ranked, and a subset of useful features was selected based on Gini coefficient scores in terms of associations with histological class. The areas under the receiver operating characteristic curves (AUCs) of classifications afforded by several machine-learning algorithms (random forest, neural network, naive Bayes, logistic regression, and a support vector machine) were compared and validated via random sampling.
Results
We developed and validated a PET-based radiomic model predicting the histological subtypes of lung cancer. Sex, SUVmax, gray-level zone length nonuniformity, gray-level nonuniformity for zone, and total lesion glycolysis were the 5 best predictors of lung ADC. The logistic regression model outperformed all other classifiers (AUC = 0.859, accuracy = 0.769, F1 score = 0.774, precision = 0.804, recall = 0.746) followed by the neural network model (AUC = 0.854, accuracy = 0.772, F1 score = 0.777, precision = 0.807, recall = 0.750).
Conclusions
A machine-learning approach successfully identified the histological subtypes of lung cancer. A PET-based radiomic features may help clinicians improve the histopathologic diagnosis in a noninvasive manner.

Feed Item Url: 
https://journals.lww.com/nuclearmed/Fulltext/2019/12000/A_Machine_Learning_Approach_Using_PET_Based.5.aspx

Granulomatous Myositis Showing Fluctuating “Leopard-Man” Sign: A Case Report on FDG PET/CT Imaging

imageA 57-year-old woman with a history of uterine endometrial carcinoma underwent PET/CT examinations for initial staging and posttreatment survey. Multiple patchy accumulations were noted in the muscles, particularly in both thighs. These accumulations resolved spontaneously 6 months after the follow-up examination. However, 3.5 years after the surgery, the multiple patchy accumulations reappeared in the muscle of the upper and lower extremities showing an increase in signal intensity from previous examination. A biopsy of the right thigh revealed epithelioid cell granuloma without necrosis. We therefore consider that this case might be “idiopathic” granulomatous myositis.

Feed Item Url: 
https://journals.lww.com/nuclearmed/Fulltext/2019/12000/Granulomatous_Myositis_Showing_Fluctuating.6.aspx

18F-Fluorocholine PET/CT Imaging of Brown Tumors in a Patient With Severe Primary Hyperparathyroidism

imageBrown tumors are rare skeletal anomalies occurring in patients with hyperparathyroidism and exposing patients to pathological fractures. We report the case of a 26-year-old woman with severe primary hyperparathyroidism (calcemia, 2.9 mmol/L; parathyroid hormone, 59 pmol/L) who underwent 18F-fluorocholine (FCH) PET/CT before parathyroidectomy. FCH PET localized the hyperfunctioning parathyroid gland and showed multiple foci in correspondence with bone lytic lesions on CT. Those lesions were not visible on the 99mTc-MIBI dual-phase scintigraphy. The pathology of one of the FCH-positive bone lesions corresponded to a brown tumor related to hyperparathyroidism.

Feed Item Url: 
https://journals.lww.com/nuclearmed/Fulltext/2019/12000/18F_Fluorocholine_PET_CT_Imaging_of_Brown_Tumors.10.aspx

18F-FDG PET Imaging Features of Patients With Autoimmune Lymphoproliferative Syndrome

imageIntroduction
Autoimmune lymphoproliferative syndrome (ALPS) is a rare immune dysregulatory condition, usually presenting in childhood with massive lymphadenopathy, splenomegaly, and an increased incidence of lymphoma. Methods to differentiate between benign ALPS adenopathy and lymphoma are needed. To this end, we evaluated the usefulness of FDG PET.
Methods
We prospectively evaluated 76 ALPS/ALPS-like patients including FS-7–associated surface antigen (FAS) germline mutation with (n = 4) and without lymphoma (n = 50), FAS-somatic (n = 6), ALPS-unknown (n = 6), and others (n = 10) who underwent FDG PET. Uptakes in 14 nodal sites, liver, and spleen were determined.
Results
In 76 ALPS patients, FDG PET showed uptake in multiple nodal sites in all but 1 patient. The highest SUVmax values in FAS mutation without lymphoma, FAS mutation with lymphoma, FAS somatic, ALPS-unknown, and other genetic mutations were a median (range) 9.2 (4.3–25), 16.2 (10.7–37.2), 7.6 (4.6–18.1), 11.5 (4.8–17.2), and 5.5 (0–15.3), respectively. Differences between uptake in the FAS group with and without lymphoma were statistically significant, but overlapped, making discrimination between individuals with/without lymphoma impossible. The spleen:liver uptake ratio was greater than 1 in 82% of patients.
Conclusions
While statistically significant differences were observed in FAS mutation ALPS with and without lymphoma, the significant overlap in FDG uptake and visual appearance in many patients prevents discrimination between patients with and without lymphoma. Similar patterns of FDG biodistribution were noted between the various ALPS subgroups.

Feed Item Url: 
https://journals.lww.com/nuclearmed/Fulltext/2019/12000/18F_FDG_PET_Imaging_Features_of_Patients_With.4.aspx