METHODS: The NFS was calculated and LSM obtained for consecutive adult NAFLD patients scheduled for liver biopsy. The accuracy of predicting advanced fibrosis using either modality and in combination were assessed. An algorithm combining the NFS and LSM was developed from a training cohort and subsequently tested in a validation cohort.

RESULTS: There were 101 and 46 patients in the training and validation cohort, respectively. In the training cohort, the percentages of misclassifications using the NFS alone, LSM alone, LSM alone (with grey zone), both tests for all patients and a 2-step approach using LSM only for patients with indeterminate and high NFS were 5.0, 28.7, 2.0, 2.0 and 4.0 %, respectively. The percentages of patients requiring liver biopsy were 30.7, 0, 36.6, 36.6 and 18.8 %, respectively. In the validation cohort, the percentages of misclassifications were 8.7, 28.3, 2.2, 2.2 and 8.7 %, respectively. The percentages of patients requiring liver biopsy were 28.3, 0, 41.3, 43.5 and 19.6 %, respectively.

METHODS: The proposed method uses a 2D contourlet transform and a set of texture features that are efficiently extracted from the transformed image. Then, the combination of a kernel discriminant analysis (KDA)-based feature reduction technique and analysis of variance (ANOVA)-based feature ranking technique was used, and the images were then classified into various stages of liver fibrosis.

RESULTS: Our 2D contourlet transform and texture feature analysis approach achieved a 91.46% accuracy using only four features input to the probabilistic neural network classifier, to classify the five stages of liver fibrosis. It also achieved a 92.16% sensitivity and 88.92% specificity for the same model. The evaluation was done on a database of 762 ultrasound images belonging to five different stages of liver fibrosis.