Evaluation of a convolution neural network for baseline total tumor metabolic volume on [18F]FDG PET in diffuse large B cell lymphoma.

OBJECTIVES: New PET data-processing tools allow for automatic lesion selection and segmentation by a convolution neural network using artificial intelligence (AI) to obtain total metabolic tumor volume (TMTV) and total lesion glycolysis (TLG) routinely at the clinical workstation. Our objective was to evaluate an AI implemented in a new version of commercial software to verify reproducibility of results and time savings in a daily workflow. METHODS: Using the software to obtain TMTV and TLG, two nuclear physicians applied five methods to retrospectively analyze data for 51 patients. Methods 1 and 2 were fully automated with exclusion of lesions ≤ 0.5 mL and ≤ 0.1 mL, respectively. Methods 3 and 4 were fully automated with physician review. Method 5 was semi-automated and used as reference. Time and number of clicks to complete the measurement were recorded for each method. Inter-instrument and inter-observer variation was assessed by the intra-class coefficient (ICC) and Bland-Altman plots. RESULTS: Between methods 3 and 5, for the main user, the ICC was 0.99 for TMTV and 1.0 for TLG. Between the two users applying method 3, ICC was 0.97 for TMTV and 0.99 for TLG. Mean processing time (± standard deviation) was 20 s ± 9.0 for method 1, 178 s ± 125.7 for method 3, and 326 s ± 188.6 for method 5 (p < 0.05). CONCLUSION: AI-enabled lesion detection software offers an automated, fast, reliable, and consistently performing tool for obtaining TMTV and TLG in a daily workflow. KEY POINTS: • Our study shows that artificial intelligence lesion detection software is an automated, fast, reliable, and consistently performing tool for obtaining total metabolic tumor volume and total lesion glycolysis in a daily workflow.

Impact of 18F-FDG PET/MR based tumor delineation in radiotherapy planning for cholangiocarcinoma.

PURPOSE: Radiation therapy (RT) is an effective treatment for unresectable cholangiocarcinoma (CC). Accurate tumor volume delineation is critical in achieving high rates of local control while minimizing treatment-related toxicity. This study compares 18F-FDG PET/MR to MR and CT for target volume delineation for RT planning. METHODS: We retrospectively included 22 patients with newly diagnosed unresectable primary CC who underwent 18F-FDG PET/MR for initial staging. Gross tumor volume (GTV) of the primary mass (GTVM) and lymph nodes (GTVLN) were contoured on CT images, MR images, and PET/MR fused images and compared among modalities. The dice similarity coefficient (DSC) was calculated to assess spatial coverage between different modalities. RESULTS: GTV MPET/MR (median: 94 ml, range 16-655 ml) was significantly greater than GTV MMR (69 ml, 11-635 ml) (p = 0.0001) and GTV MCT (96 ml, 4-564 ml) (p = 0.035). There was no significant difference between GTV MCT and GTV MMR (p = 0.078). Subgroup analysis of intrahepatic and extrahepatic tumors showed that the median GTV MPET/MR was significantly greater than GTV MMR in both groups (117.5 ml, 22-655 ml vs. 102.5 ml, 22-635 ml, p = 0.004 and 37 ml, 16-303 ml vs. 34 ml, 11-207 ml, p = 0.042, respectively). The GTV LNPET/MR (8.5 ml, 1-27 ml) was significantly higher than GTV LNCT (5 ml, 4-16 ml) (p = 0.026). GTVPET/MR had the highest similarity to the GTVMR, i.e., DSCPET/MR-MR (0.82, 0.25-1.00), compared to DSC PET/MR-CT of 0.58 (0.22-0.87) and DSCMR-CT of 0.58 (0.03-0.83). CONCLUSION: 18F-FDG PET/MR-based CC delineation yields greater GTVs and detected a higher number of positive lymph nodes compared to CT or MR, potentially improving RT planning by reducing the risk of geographic misses.

Prognostic value of the metabolic bulk volume in patients with diffuse large B-cell lymphoma on baseline 18F-FDG PET-CT.

We retrospectively investigated the prognostic value of the metabolic bulk volume (MBV), defined as the metabolic volume of the largest lesion, in 106 patients with diffuse large B-cell lymphoma who underwent baseline 18FDG PET-CT. Semi-automatically segmented (41% SUVmax) total metabolic tumor volume (TMTV) and MBV underwent receiver operating characteristic analysis, identifying optimal thresholds of 147 cm3 for the TMTV and 41.5 cm3 for the MBV. A low TMTV and a low MBV were significantly associated with longer progression-free survival (PFS) and overall survival (OS). According to Cox multivariate analysis, the TMTV and MBV were independent predictors, respectively, for PFS (HR = 3.501), and OS (HR = 7.242). The TMTV and MBV classification were discordant in 18.9% of patients with a 5-year PFS rate of 100% for patients with low MBV/high TMTV and 74% for high MBV/low TMTV. The baseline MBV can be an efficient tool for the risk stratification of aggressive lymphoma.