Search

BACKGROUND. Deep learning abdominal organ segmentation algorithms have shown excellent results in adults; validation in children is sparse. OBJECTIVE. The purpose of this article is to develop and validate deep learning models for liver, spleen, and pancreas segmentation on pediatric CT examinations. METHODS. This retrospective study developed and validated deep learning models for liver, spleen, and pancreas segmentation using 1731 CT examinations (1504 training, 221 testing), derived from three internal institutional pediatric (age ≤ 18 years) datasets ( n = 483) and three public datasets comprising pediatric and adult examinations with various pathologies ( n = 1248). Three deep learning model architectures (SegResNet, DynUNet, and SwinUNETR) from the Medical Open Network for Artificial Intelligence (MONAI) framework underwent training using native training (NT), relying solely on institutional datasets, and transfer learning (TL), incorporating pretraining on public datasets. For comparison, TotalSegmentator, a publicly available segmentation model, was applied to test data without further training. Segmentation performance was evaluated using mean Dice similarity coefficient (DSC), with manual segmentations as reference. RESULTS. For internal pediatric data, the DSC for TotalSegmentator, NT models, and TL models for normal liver was 0.953, 0.964-0.965, and 0.965-0.966, respectively; for normal spleen, 0.914, 0.942-0.945, and 0.937-0.945; for normal pancreas, 0.733, 0.774-0.785, and 0.775-0.786; and for pancreas with pancreatitis, 0.703, 0.590-0.640, and 0.667-0.711. For public pediatric data, the DSC for TotalSegmentator, NT models, and TL models for liver was 0.952, 0.871-0.908, and 0.941-0.946, respectively; for spleen, 0.905, 0.771-0.827, and 0.897-0.926; and for pancreas, 0.700, 0.577-0.648, and 0.693-0.736. For public primarily adult data, the DSC for TotalSegmentator, NT models, and TL models for liver was 0.991, 0.633-0.750, and 0.926-0.952, respectively; for spleen, 0.983, 0.569-0.604, and 0.923-0.947; and for pancreas, 0.909, 0.148-0.241, and 0.699-0.775. The DynUNet TL model was selected as the best-performing NT or TL model considering DSC values across organs and test datasets and was made available as an open-source MONAI bundle (https://github.com/cchmc-dll/pediatric_abdominal_segmentation_bundle.git). CONCLUSION. TL models trained on heterogeneous public datasets and fine-tuned using institutional pediatric data outperformed internal NT models and Total-Segmentator across internal and external pediatric test data. Segmentation performance was better in liver and spleen than in pancreas. CLINICAL IMPACT. The selected model may be used for various volumetry applications in pediatric imaging.

The purpose of this study was to assess relationships between liver-corrected T1 (cT1) values (adjusted for T2* effect, MRI system manufacturer, and field strength) and histologic inflammation and fibrosis in 35 participants (15 women and girls, 20 boys and men; median age, 16.0 years) with autoimmune liver disease. At multivariable analysis, inflammation score (β = 15.5) and sex (β = 56.0 [female]) were independent predictors of cT1, and fibrosis score (β = 32.3) and age (β = 5.5) were independent predictors of cT1 IQR. Liver T1 may have relevance for assessing liver inflammatory activity and fibrosis stage.

BACKGROUND. The Mayo risk score and SCOPE (Sclerosing Cholangitis Outcomes in Pediatrics) index are clinical risk scores for monitoring the progression of primary sclerosing cholangitis (PSC) and predicting clinically important endpoints. OBJECTIVE. The purpose of this study was to evaluate relationships between quantitative MRI measures of liver disease and clinical risk scores in children and young adults with autoimmune liver disease (AILD). METHODS. This prospective study included 58 patients (35 male and 23 female patients; mean age, 15.1 ± 1.1 [SD] years [range, 6-24 years]) with AILD (16 with PSC, 30 with autoimmune hepatitis, and 12 with autoimmune sclerosing cholangitis) who underwent research liver MRI examinations including MR elastography, T2*-corrected T1 (cT1), and quantitative MRCP measurements. Associations between quantitative MRI metrics and clinical risk scores were evaluated using Spearman rank-order correlation coefficients and multivariable regression analyses. RESULTS. The mean Mayo risk score was -1.1 ± 0.9 (SD) (range, -2.9 to 1.1); the mean SCOPE index was 2.7 ± 2.2 (range, 0-9). Mean liver stiffness was 2.8 ± 1.1 kPa, whole-liver mean cT1 was 874.7 ± 77.7 ms, and whole-liver cT1 interquartile range (IQR) was 150.8 ± 55.6 ms. The Mayo risk score was significantly correlated with liver stiffness (ρ = 0.56; p < .001), whole-liver mean cT1 (ρ = 0.31; p = .02), whole-liver cT1 IQR (ρ = 0.58; p < .001), and multiple quantitative MRCP measures (ρ = 0.36-0.45, p < .001). SCOPE index was significantly correlated with liver stiffness (ρ = 0.68; p < .001), whole-liver cT1 IQR (ρ = 0.51; p < .001), and multiple quantitative MRCP measures (ρ = 0.47-0.49; p < .001). Multivariable linear regression analyses identified liver stiffness, whole-liver cT1 IQR, and left hepatic duct maximum diameter as significant independent predictors of the Mayo risk score (adjusted R 2 = 0.45), and liver stiffness, whole-liver cT1 IQR, maximum common bile duct (CBD) diameter, and median CBD diameter as significant independent predictors of the SCOPE index (adjusted R 2 = 0.69). On logistic regression analysis, greater than low risk by SCOPE index was best predicted by liver stiffness [odds ratio [OR] = 49.6; 95% CI, 3.1-793.6) and maximum CBD diameter (OR = 2.5; 95% CI, 1.3-4.7). CONCLUSION. Increased liver stiffness, increased cT1 IQR, and larger bile duct diameters are independently associated with higher (worse) Mayo risk score and SCOPE index among children and young adults with AILD and may be surrogate markers of clinically meaningful endpoints. CLINICAL IMPACT. Multiparametric MRI of the liver incorporating quantitative metrics may serve as a noninvasive diagnostic and prognostic tool in pediatric AILD. TRIAL REGISTRATION. ClinicalTrials.gov: NCT03175471.