MsgeCNN: Multiscale geometric embedded convolutional neural network for ONFH segmentation and grading.

BACKGROUND: The incidence of osteonecrosis of the femoral head (ONFH) is increasing gradually, rapid and accurate grading of ONFH is critical. The existing Steinberg staging criteria grades ONFH according to the proportion of necrosis area to femoral head area. PURPOSE: In the clinical practice, the necrosis region and femoral head region are mainly estimated by the observation and experience of doctor. This paper proposes a two-stage segmentation and grading framework, which can be used to segment the femoral head and necrosis, as well as to diagnosis. METHODS: The core of the proposed two-stage framework is the multiscale geometric embedded convolutional neural network (MsgeCNN), which integrates geometric information into the training process and accurately segments the femoral head region. Then, the necrosis regions are segmented by the adaptive threshold method taking femoral head as the background. The area and proportion of the two are calculated to determine the grade. RESULTS: The accuracy of the proposed MsgeCNN for femoral head segmentation is 97.73%, sensitivity is 91.17%, specificity is 99.40%, dice score is 93.34%. And the segmentation performance is better than the existing five segmentation algorithms. The diagnostic accuracy of the overall framework is 90.80%. CONCLUSIONS: The proposed framework can accurately segment the femoral head region and the necrosis region. The area, proportion, and other pathological information of the framework output provide auxiliary strategies for subsequent clinical treatment.

Identification of low-overpressure interval and its implication to hydrocarbon migration: Case study in the Yanan sag of the Qiongdongnan Basin, South China Sea.

The Qiongdongnan Basin is a strongly overpressured basin with the maximum pressure coefficient (the ratio of the actual pore pressure versus hydrostatic pressure at the same depth) over 2.27. However, there exists a widespread low-overpressure interval between the strong overpressure intervals in the Yanan Sag of western basin. The mechanisms of the low-overpressure interval are not well understood. Three main approaches, pore pressure test data and well-log analysis, pressure prediction based on the relationship between the deviation of the velocity and the pressure coefficients, and numerical modeling, were employed to illustrate the distribution and evolution of the low-overpressure interval. And we analyzed and explained the phenomenon of the low-overpressure interval that is both underlain and overlain by high overpressure internal. The low-overpressure interval between the strong overpressure intervals can be identified and modelled by drilling data of P-wave sonic and the mud weight, and the numerical modeling using the PetroMod software. Results show that the low-overpressure interval is mainly composed of sandstone sediments. The porosities of sandstone in the low-overpressure interval primarily range from 15%-20%, and the permeabilities range from 10-100 md. Analysis of the geochemical parameters of C1, iC4/nC4, ΔR3, and numerical modeling shows that oil and gas migrated upward into the sandstone in the low-overpressure interval, and then migrated along the sandstone of low-overpressure interval into the Yacheng uplift. The low-overpressure both underlain and overlain by overpressure resulted from the fluids migrating along the sandstones in the low-overpressure interval into the Yacheng uplift since 1.9Ma. The mudstone in the strong overpressure interval is good cap overlain the sandstone of low-overpressure interval, therefore up-dip pinchouts or isolated sandstone in the low-overpressure interval locating the migration path of oil and gas are good plays for hydrocarbon exploration.