Clinical Evaluation of Deep Learning for Tumor Delineation on 18F-FDG PET/CT of Head and Neck Cancer.

Artificial intelligence (AI) may decrease 18F-FDG PET/CT-based gross tumor volume (GTV) delineation variability and automate tumor-volume-derived image biomarker extraction. Hence, we aimed to identify and evaluate promising state-of-the-art deep learning methods for head and neck cancer (HNC) PET GTV delineation. Methods: We trained and evaluated deep learning methods using retrospectively included scans of HNC patients referred for radiotherapy between January 2014 and December 2019 (ISRCTN16907234). We used 3 test datasets: an internal set to compare methods, another internal set to compare AI-to-expert variability and expert interobserver variability (IOV), and an external set to compare internal and external AI-to-expert variability. Expert PET GTVs were used as the reference standard. Our benchmark IOV was measured using the PET GTV of 6 experts. The primary outcome was the Dice similarity coefficient (DSC). ANOVA was used to compare methods, a paired t test was used to compare AI-to-expert variability and expert IOV, an unpaired t test was used to compare internal and external AI-to-expert variability, and post hoc Bland-Altman analysis was used to evaluate biomarker agreement. Results: In total, 1,220 18F-FDG PET/CT scans of 1,190 patients (mean age ± SD, 63 ± 10 y; 858 men) were included, and 5 deep learning methods were trained using 5-fold cross-validation (n = 805). The nnU-Net method achieved the highest similarity (DSC, 0.80 [95% CI, 0.77-0.86]; n = 196). We found no evidence of a difference between expert IOV and AI-to-expert variability (DSC, 0.78 for AI vs. 0.82 for experts; mean difference of 0.04 [95% CI, -0.01 to 0.09]; P = 0.12; n = 64). We found no evidence of a difference between the internal and external AI-to-expert variability (DSC, 0.80 internally vs. 0.81 externally; mean difference of 0.004 [95% CI, -0.05 to 0.04]; P = 0.87; n = 125). PET GTV-derived biomarkers of AI were in good agreement with experts. Conclusion: Deep learning can be used to automate 18F-FDG PET/CT tumor-volume-derived imaging biomarkers, and the deep-learning-based volumes have the potential to assist clinical tumor volume delineation in radiation oncology.

Is Erythrocyte Sedimentation Rate Necessary for the Initial Diagnosis of Giant Cell Arteritis?

Giant cell arteritis (GCA) is an ophthalmological emergency that can be difficult to diagnose and prompt treatment is vital. We investigated the sequential diagnostic value for patients with suspected GCA using three biochemical measures as they arrive to the clinician: first, platelet count, then C-reactive protein (CRP), and lastly, erythrocyte sedimentation rate (ESR). This retrospective cross-sectional study of consecutive patients with suspected GCA investigated platelet count, CRP, and ESR using diagnostic test accuracy statistics and odds ratios (ORs) in a sequential fashion. The diagnosis was established by experts at follow-up, considering clinical findings and tests including temporal artery biopsy. A total of 94 patients were included, of which 37 (40%) were diagnosed with GCA. Compared with those without GCA, patients with GCA had a higher platelet count (p < 0.001), CRP (p < 0.001), and ESR (p < 0.001). Platelet count demonstrated a low sensitivity (38%) and high specificity (88%); CRP, a high sensitivity (86%) and low specificity (56%); routine ESR, a high sensitivity (89%) and low specificity (47%); and age-adjusted ESR, a moderate sensitivity (65%) and moderate specificity (65%). Sequential analysis revealed that ESR did not provide additional value in evaluating risk of GCA. Initial biochemical evaluation can be based on platelet count and CRP, without waiting for ESR, which allows faster initial decision-making in GCA.

Validation of non-invasive haemodynamic methods in patients with liver disease: the Finometer and the Task Force Monitor.

Patients with advanced cirrhosis often present a hyperdynamic circulation characterized by a decrease in systolic and diastolic blood pressure (SBP and DBP), and an increase in heart rate (HR) and cardiac output (CO). Accurate assessment of the altered circulation can be performed invasively; however, due to the disadvantages of this approach, non-invasive methods are warranted. The purpose of this study was to compare continuous non-invasive measurements of haemodynamic variables by the Finometer and the Task Force Monitor with simultaneous invasive measurements. In 25 patients with cirrhosis, SBP, DBP and HR were measured non-invasively and by femoral artery catheterization. CO was measured non-invasively and by indicator dilution technique. The non-invasive pressure monitoring was considered acceptable with a bias (accuracy) and a SD (precision) not exceeding 5 and 8 mmHg, respectively, as recommended by the Association for the Advancement of Medical Instrumentation. The accuracy and precision of the Finometer and the Task Force Monitor were as follows: SBP: -3·6 ± 17·9 and -8·9 ± 17·5 mmHg, respectively; DBP: 4·2 ± 9·6 and 1·9 ± 8·6 mmHg, respectively; HR: 2·0 ± 6·9 and 2·2 ± 6·2 bpm, respectively; and CO: 0·1 ± 1·6 and -1·0 ± 2·0 L min-1 , respectively. The study demonstrates that the overall performances of the Finometer and the Task Force Monitor in estimating absolute values of SBP, DBP, HR and CO in patients with cirrhosis are not equivalent to the gold standard, but may have an acceptable performance with repeated measurements.