Deep convolutional neural network for differentiating between sarcoidosis and lymphoma based on [18F]FDG maximum-intensity projection images

OBJECTIVES

To compare the [18F]FDG PET/CT findings of untreated sarcoidosis and malignant lymphoma (ML) and develop convolutional neural network (CNN) models to differentiate between these diseases using maximum intensity projection (MIP) [18F]FDG PET images.

METHODS

We retrospectively collected data on consecutive patients newly diagnosed with sarcoidosis and ML who underwent [18F]FDG PET/CT before treatment. Two nuclear radiologists reviewed the images. CNN models were created using MIP PET images and evaluated with k-fold cross-validation. The points of interest were visualized using gradient-weighted class activation mapping (Grad-CAM).

RESULTS

A total of 56 patients with sarcoidosis and 62 patients with ML were included. Patients with sarcoidosis had more prominent FDG accumulation in the mediastinal lymph nodes and lung lesions, while those with ML had more prominent accumulation in the cervical lymph nodes (all p < 0.001). For the mediastinal lymph nodes, sarcoidosis patients had significant FDG accumulation in the level 2, 4, 7, and 10 lymph nodes (all p < 0.01). Otherwise, the accumulation in ML patients tended to be in the level 1 lymph nodes (p = 0.08). The CNN model using frontal and lateral MIP images achieved an average accuracy of 0.890 (95% CI: 0.804-0.977), a sensitivity of 0.898 (95% CI: 0.782-1.000), a specificity of 0.907 (95% CI: 0.799-1.000), and an area under the curve of 0.963 (95% CI: 0.899-1.000). Grad-CAM showed that the model focused on the sites of abnormal FDG accumulation.

CONCLUSIONS

CNN models based on differences in FDG accumulation sites archive high performance in differentiating between sarcoidosis and ML.

CLINICAL RELEVANCE STATEMENT

We developed a CNN model using MIP images of [18F]FDG PET/CT to distinguish between sarcoidosis and malignant lymphoma. It achieved high performance and could be useful in diagnosing diseases with involvement across organs and lymph nodes.

KEY POINTS

• There are differences in FDG distribution when comparing whole-body [18F]FDG PET/CT findings in patients with sarcoidosis and malignant lymphoma before treatment. • Convolutional neural networks, a type of deep learning technique, trained with maximum-intensity projection PET images from two angles showed high performance. • A deep learning model that utilizes differences in FDG distribution may be helpful in differentiating between diseases with lesions that are characteristically widespread among organs and lymph nodes.

Survival and acute exacerbation for patients with idiopathic pulmonary fibrosis (IPF) or non-IPF idiopathic interstitial pneumonias: 5-year follow-up analysis of a prospective multi-institutional patient registry

OBJECTIVE

Few prospective cohort studies with relatively large numbers of patients with non-idiopathic pulmonary fibrosis (non-IPF) of idiopathic interstitial pneumonia (IIP) have been described. We aimed to assess disease progression and cause of death for patients with non-IPF IIPs or IPF under real-life conditions.

METHODS

Data were analysed for a prospective multi-institutional cohort of 528 IIP patients enrolled in Japan between September 2013 and April 2016. Diagnosis of IPF versus non-IPF IIPs was based on central multidisciplinary discussion, and follow-up surveillance was performed for up to 5 years after patient registration. Survival and acute exacerbation (AE) were assessed.

RESULTS

IPF was the most common diagnosis (58.0%), followed by unclassifiable IIPs (35.8%) and others (6.2%). The 5-year survival rate for non-IPF IIP and IPF groups was 72.8% and 53.7%, respectively, with chronic respiratory failure being the primary cause of death in both groups. AE was the second most common cause of death for both non-IPF IIP (24.1%) and IPF (23.5%) patients. The cumulative incidence of AE did not differ significantly between the two groups (p=0.36), with a 1-year incidence rate of 7.4% and 9.0% in non-IPF IIP and IPF patients, respectively. We found that 30.2% and 39.4% of non-IPF IIP and IPF patients, respectively, who experienced AE died within 3 months after an AE event, whereas 55.8% and 66.7% of such patients, respectively, died within 5 years after registration.

CONCLUSION

Closer monitoring of disease progression and palliative care interventions after AE are important for non-IPF IIP patients as well as for IPF patients.