Liver Tumor Segmentation in CT Scans Using Modified SegNet.

The main cause of death related to cancer worldwide is from hepatic cancer. Detection of hepatic cancer early using computed tomography (CT) could prevent millions of patients' death every year. However, reading hundreds or even tens of those CT scans is an enormous burden for radiologists. Therefore, there is an immediate need is to read, detect, and evaluate CT scans automatically, quickly, and accurately. However, liver segmentation and extraction from the CT scans is a bottleneck for any system, and is still a challenging problem. In this work, a deep learning-based technique that was proposed for semantic pixel-wise classification of road scenes is adopted and modified to fit liver CT segmentation and classification. The architecture of the deep convolutional encoder-decoder is named SegNet, and consists of a hierarchical correspondence of encode-decoder layers. The proposed architecture was tested on a standard dataset for liver CT scans and achieved tumor accuracy of up to 99.9% in the training phase.

New criteria for functions to be in a class of p-valent alpha convex functions.

We obtain certain simple sufficiency criteria for a class of p-valent alpha convex functions. Many known results appear as special consequences of our work. Some applications of our work to the generalized integral operator are also given.

Relation of the left ventricular geometric patten to coronary artery disease in hypertensive patients using a 320-detector-row CT scanner.

BACKGROUND: It was estimated that about 1.3 billion people were diagnosed to be hypertensive in 2015. All countries consistently show this high prevalence. Ischemic heart disease stands as the most common cause of systolic blood pressure-related deaths per year. Left ventricular hypertrophy determined by echocardiography can predict cardiovascular morbidity and mortality. The question of whether the LV geometric pattern has an additional prognostic value is still not clearly answered. Currently, coronary computed tomography is widely used in clinical practice with a great capability of simultaneous evaluation of the LV mass and the coronary arterial tree. Our study aims to examine the relationship between LV mass and geometry and coronary artery disease using an ECG-gated 320-detector- row CT scanner. RESULTS: Two hundred ninety-eight hypertensive Egyptian individuals were enrolled in our study, the mean age was 57.5 ± 10.5, and males comprised 76.5% of the study population. The mean LV mass and LV mass index were 193 ± 60 gm and 95.2 ± 27.5 g/m2 respectively. One-fifth of the patient had CAD luminal stenosis ≥ 50%. Normal LV geometric pattern was observed in about 37% of the study population. About one-third of the patients showed concentric remodeling. Patients with increased LV mass index represented one-third of the study population with a greater percentage of the concentric hypertrophy pattern than the eccentric hypertrophy pattern. Patients with high CAD-RADS showed statistically significant higher LV mass, LV mass index, and septal wall thickness. Patients with high CAD-RADS showed a greater percentage of concentric and eccentric hypertrophy. The LV geometric pattern was the only independent predictor of the high CAD-RADS. The LV geometric patterns associated with high RADS ordered from the highest to the lowest, were concentric LVH, Eccentric LV, and concentric remodeling. CONCLUSIONS: LV geometric pattern is the only independent predictor of high CAD-RADS after adjustment for LV mass index and septal wall thickness. Among abnormal LV geometric patterns, concentric hypertrophy stands as the most important predictor of high CAD-RADS.