Union is strength: the combination of radiomics features and 3D-deep learning in a sole model increases diagnostic accuracy in demented patients: a whole brain 18FDG PET-CT analysis.

OBJECTIVE: FDG PET imaging plays a crucial role in the evaluation of demented patients by assessing regional cerebral glucose metabolism. In recent years, both radiomics and deep learning techniques have emerged as powerful tools for extracting valuable information from medical images. This article aims to provide a comparative analysis of radiomics features, 3D-deep learning convolutional neural network (CNN) and the fusion of them, in the evaluation of 18F-FDG PET whole brain images in patients with dementia and normal controls. METHODS: 18F-FDG brain PET and clinical score were collected in 85 patients with dementia and 125 healthy controls (HC). Patients were assigned to various form of dementia on the basis of clinical evaluation, follow-up and voxels comparison with HC using a two-sample Student's t -test, to determine the regions of brain involved. Radiomics analysis was performed on the whole brain after normalization to an optimized template. After selection using the minimum redundancy maximum relevance method and Pearson's correlation coefficients, the features obtained were added to a neural network model to find the accuracy in classifying HC and demented patients. Forty subjects not included in the training were used to test the models. The results of the three models (radiomics, 3D-CNN, combined model) were compared with each other. RESULTS: Four radiomics features were selected. The sensitivity was 100% for the three models, but the specificity was higher with radiomics and combined one (100% vs. 85%). Moreover, the classification scores were significantly higher using the combined model in both normal and demented subjects. CONCLUSION: The combination of radiomics features and 3D-CNN in a single model, applied to the whole brain 18FDG PET study, increases the accuracy in demented patients.

(123) I-2ß-carbomethoxy-3ß-(4-iodophenyl)-N-(3-fluoropropyl) nortropane single photon emission computed tomography and (123) I-metaiodobenzylguanidine myocardial scintigraphy in differentiating dementia with lewy bodies from other dementias: A comparative study.

OBJECTIVE: To compare the diagnostic value of striatal (123) I-2ß-carbomethoxy-3ß-(4-iodophenyl)-N-(3-fluoropropyl) nortropane ((123) I-FP-CIT) single photon emission computed tomography (SPECT) and (123) I-metaiodobenzylguanidine ((123) I-MIBG) myocardial scintigraphy in differentiating dementia with Lewy bodies (DLB) from other dementia types. METHODS: This prospective longitudinal study included 30 patients with a clinical diagnosis of DLB and 29 patients with non-DLB dementia (Alzheimer disease, n = 16; behavioral variant frontotemporal dementia, n = 13). All patients underwent (123) I-FP-CIT SPECT and (123) I-MIBG myocardial scintigraphy within a few weeks of clinical diagnosis. All diagnoses at each center were agreed upon by the local clinician and an independent expert, both unaware of imaging data, and re-evaluated after 12 months. Each image was visually classified as either normal or abnormal by 3 independent nuclear physicians blinded to patients' clinical data. RESULTS: Overall, sensitivity and specificity to DLB were respectively 93% and 100% for (123) I-MIBG myocardial scintigraphy, and 90% and 76% for (123) I-FP-CIT SPECT. Lower specificity of striatal compared to myocardial imaging was due to decreased (123) I-FP-CIT uptake in 7 non-DLB subjects (3 with concomitant parkinsonism) who had normal (123) I-MIBG myocardial uptake. Notably, in our non-DLB group, myocardial imaging gave no false-positive readings even in those subjects (n = 7) with concurrent medical illnesses (diabetes and/or heart disease) supposed to potentially interfere with (123) I-MIBG uptake. INTERPRETATION: (123) I-FP-CIT SPECT and (123) I-MIBG myocardial scintigraphy have similar sensitivity for detecting DLB, but the latter appears to be more specific for excluding non-DLB dementias, especially when parkinsonism is the only "core feature" exhibited by the patient. Our data also indicate that the potential confounding effects of diabetes and heart disease on (123) I-MIBG myocardial scintigraphy results might have been overestimated. Ann Neurol 2016;80:368-378.

Unsuspected Active Sarcoidosis Diagnosed by 18F-FDG PET/CT During the Search for a Primary Tumour in a Patient with Bone Lesions.

Sarcoidosis is a systemic chronic inflammatory disease of unknown aetiology, characterised by granulomatous lesions with heterogeneous clinical manifestations affecting multiple organs and tissues. Although the respiratory system is most commonly affected, the disease may also present with bone lesions. We report the case of a 31-year-old woman who presented with low back pain and no history of cancer and who was found to have suspicious lesions involving the entire spine on magnetic resonance imaging (MRI). The patient underwent 18F-fluorodeoxyglucose (FDG) PET/CT to search for a primary tumour and for staging purposes. 18F-FDG PET/CT revealed a pattern of hypermetabolic activity in widespread skeletal lesions and in a single left cervical lymph node. The primary tumour was not found, thus suggesting a haematologic disorder. Subsequent biopsies of a cervical lymph node and of bone tissue from L4 revealed active sarcoidosis with no evidence of cancer. This underlines the importance of considering all alternatives when hypermetabolic lesions are found on 18F-FDG PET/CT. Furthermore, 18F-FDG PET can be very useful to indicate accessible sites for guiding fine-needle aspiration cytology (FNAC).