Comparison of Radiologists and Deep Learning for US Grading of Hepatic Steatosis.

Background Screening for nonalcoholic fatty liver disease (NAFLD) is suboptimal due to the subjective interpretation of US images. Purpose To evaluate the agreement and diagnostic performance of radiologists and a deep learning model in grading hepatic steatosis in NAFLD at US, with biopsy as the reference standard. Materials and Methods This retrospective study included patients with NAFLD and control patients without hepatic steatosis who underwent abdominal US and contemporaneous liver biopsy from September 2010 to October 2019. Six readers visually graded steatosis on US images twice, 2 weeks apart. Reader agreement was assessed with use of κ statistics. Three deep learning techniques applied to B-mode US images were used to classify dichotomized steatosis grades. Classification performance of human radiologists and the deep learning model for dichotomized steatosis grades (S0, S1, S2, and S3) was assessed with area under the receiver operating characteristic curve (AUC) on a separate test set. Results The study included 199 patients (mean age, 53 years ± 13 [SD]; 101 men). On the test set (n = 52), radiologists had fair interreader agreement (0.34 [95% CI: 0.31, 0.37]) for classifying steatosis grades S0 versus S1 or higher, while AUCs were between 0.49 and 0.84 for radiologists and 0.85 (95% CI: 0.83, 0.87) for the deep learning model. For S0 or S1 versus S2 or S3, radiologists had fair interreader agreement (0.30 [95% CI: 0.27, 0.33]), while AUCs were between 0.57 and 0.76 for radiologists and 0.73 (95% CI: 0.71, 0.75) for the deep learning model. For S2 or lower versus S3, radiologists had fair interreader agreement (0.37 [95% CI: 0.33, 0.40]), while AUCs were between 0.52 and 0.81 for radiologists and 0.67 (95% CI: 0.64, 0.69) for the deep learning model. Conclusion Deep learning approaches applied to B-mode US images provided comparable performance with human readers for detection and grading of hepatic steatosis. Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Tuthill in this issue.

Differentiation of focal nodular hyperplasia and hepatocellular adenoma using qualitative and quantitative imaging features and classification and regression tree analysis.

PURPOSE: To assess qualitative and quantitative analysis of gadoxetate disodium-enhanced hepatobiliary phase MR imaging (MRI) and assess the performance of classification and regression tree analysis for the differentiation of focal nodular hyperplasia (FNH) and hepatocellular adenoma (HCA). MATERIALS AND METHODS: This retrospective study was approved by our local ethics committee. One hundred seventy patients suspected of having FNH or HCA underwent gadoxetate disodium-enhanced MRI. The reference standard was either pathology or follow-up imaging. Two readers reviewed images to identify qualitative imaging features and measure signal intensity on unenhanced, dynamic, and hepatobiliary phase images. For quantitative analysis, contrast enhancement ratio (CER), lesion-to-liver contrast (LLC), signal intensity ratio (SIR), and relative signal enhancement ratio (RSER) were calculated. A classification and regression tree (CART) analysis was developed. RESULTS: Eighty-five patients met the inclusion criteria, with a total of 97 FNHs and 43 HCAs. For qualitative analysis, the T1 signal intensity on the hepatobiliary phase provided the highest overall classification performance (91.9% sensitivity, 90.1% specificity, and 90.9% accuracy). For quantitative analysis, RSER in the hepatobiliary phase with a threshold of 0.723 provided the highest classification performance (92.6% sensitivity and 89.4% specificity) to differentiate FNHs from HCAs. A CART model based on five qualitative imaging features provided an accuracy of 94.4% (95% confidence interval 90.0-98.9%). CONCLUSION: Gadoxetate disodium-enhanced hepatobiliary phase provides high diagnostic performance as demonstrated in quantitative and qualitative analysis in differentiation of FNH and HCA, supported by a CART decision model.

An artificial intelligence model predicts the survival of solid tumour patients from imaging and clinical data.

BACKGROUND: The need for developing new biomarkers is increasing with the emergence of many targeted therapies. Artificial Intelligence (AI) algorithms have shown great promise in the medical imaging field to build predictive models. We developed a prognostic model for solid tumour patients using AI on multimodal data. PATIENTS AND METHODS: Our retrospective study included examinations of patients with seven different cancer types performed between 2003 and 2017 in 17 different hospitals. Radiologists annotated all metastases on baseline computed tomography (CT) and ultrasound (US) images. Imaging features were extracted using AI models and used along with the patients' and treatments' metadata. A Cox regression was fitted to predict prognosis. Performance was assessed on a left-out test set with 1000 bootstraps. RESULTS: The model was built on 436 patients and tested on 196 patients (mean age 59, IQR: 51-6, 411 men out of 616 patients). On the whole, 1147 US images were annotated with lesions delineation, and 632 thorax-abdomen-pelvis CTs (total of 301,975 slices) were fully annotated with a total of 9516 lesions. The developed model reaches an average concordance index of 0.71 (0.67-0.76, 95% CI). Using the median predicted risk as a threshold value, the model is able to significantly (log-rank test P value < 0.001) isolate high-risk patients from low-risk patients (respective median OS of 11 and 31 months) with a hazard ratio of 3.5 (2.4-5.2, 95% CI). CONCLUSION: AI was able to extract prognostic features from imaging data, and along with clinical data, allows an accurate stratification of patients' prognoses.

Laparoscopic Conversion of Omega Loop Gastric Bypass into Roux-en-Y Gastric Bypass.

ᅟ.

Effects on duration of post-operative ischemia and patterns of blood flow recovery in different conditions of mouse hind limb ischemia.

BACKGROUND: Current limitations to the experimentation on patients with peripheral arterial disease push the development of different preclinical strategies. We investigated both duration of ischemia and blood flow recovery in mouse models of partial femoral artery ligation. METHODS: Male BALB/c mice were used. The ligation over needle method involved placing a suture needle over the femoral artery, ligating over it and then removing the needle. The transfixation method involved transfixing the approximate center of the femoral artery and then tying the suture. Laser Doppler Perfusion Imaging was used to assess perfusion every 3rd day until 42 days after the procedure. RESULTS: Ligation over needle method: Immediately post procedure, mean perfusion was -71.87% ± 4.43. Then mean difference in perfusion remained below the base line reading on days 3, 6, 9, and 12. From day 15 on wards mean perfusion progressively improved remaining near base line. Transfixation Method: Immediately post procedure mean perfusion was -70.82% ± 4.73. Mean perfusion improved following the procedure on days 3 and 6; a plateau followed this on days 9, 12 and 15. From day 15 onwards perfusion progressively improved remaining well below base line until crossing it on day 36. CONCLUSION: The currently described models do not pose major improvements over previously described methods.