Helicobacter pylori (H. pylori) can cause gastritis, peptic ulcer diseases and gastric carcinoma. Endoscopy as the gold standard method of diagnosis is an invasive procedure that might not be suitable in all scenarios. Therefore, this first study in Jordan aimed to assess the non-invasive ¹³C urea breath test (UBT) and stool antigen test for diagnosis of H. pylori infection and the successfulness of eradication therapy as alternatives for endoscopy. Hence, a total of 30 patients attending the endoscopy units at Alkarak teaching hospital were asked to complete a questionnaire with demographic and clinical data. They were then tested for H. pylori using ¹³C UBT and H. pylori stool antigen before having endoscopy. Another 30 patients who were positive for H. pylori by endoscopy were tested using both tests 6 weeks post eradication therapy. Results showed that the rate of H. pylori detection using endoscopy was 56.7% (17/30). Heartburns (82.3%, p value = 0.019), epigastric pain (88.2%, p value = 0.007) and vomiting (70.5%, p value = 0.02) were the most significant symptoms. Family history of peptic ulcer diseases was significantly associated with an increased risk for having a H. pylori positive result (p value = 0.02). Compared to endoscopy, the sensitivity of ¹³C UBT for the diagnosis of H. pylori was 94.1% (16/17), while it was 76.5% (13/17) for the stool antigen test. The specificity of both tests was equal (76.9%). However, the positive predictive and negative predictive values (84.2% and 90.9%) for ¹³C UBT were higher than those (81.3% and 71.4%) for the stool antigen test. The accuracy of ¹³C UBT was 86.7% compared to 76.7% for the stool antigen test. There was an 87% agreement (20 patients out of 23) between both tests when used to assess success of the eradication therapy. In conclusion, the ¹³C UBT was found to be more sensitive and accurate than the stool antigen test when used for diagnosis; furthermore, it has a comparable outcome to the stool antigen test in assessing the successfulness of the eradication treatment.

The burden of stroke can be substantially studied by establishing the functional consequences of stroke and its predictors on the population, economy and to guide rehabilitation efforts. This study aims to determine the subtypes, risk factors, and epidemiology of stroke in Jordan. Methods: A retrospective cohort study design was carried out to determine the risk factors and subtypes of stroke during 2017–2018. The study sample included 176 ischemic stroke patients of the King Abdullah University Hospital. Data was collected through medical records, which was then statistically analysed through frequencies and percentages. Results: Total 176 cases were identified out of which 101 (57.38%) were males and 75 (42.61%) were females and male to female ratio was 1.9:1. Hypertension was the commonest risk factor identified (50.56%), followed by diabetes mellitus (19.88%), hyperlipidaemia (15.34%), coronary artery disease (6.25%), atrial fibrillation (4.54%), and past history of stroke (1.13%). Risk factors such as hypertension (p = 0.007), diabetes (p = 0.000), coronary artery disease (p = 0.000) were significantly associated with subtypes of ischemic strokes. Conclusion: The study concludes that mean age of men was higher as compared to women in small vessel occlusion. The risk of ischemic stroke in patients with dyslipidaemia, diabetes mellitus, and hypertension was higher in middle-aged and old patients.

The right azygos lobe is a rare anatomical variant of the upper lung lobe that can be misdiagnosed as a neoplasm, a lung abscess, or a bulla. The aim of this study was to assess the prevalence of right azygos lobe and to evaluate the ability of postgraduate doctors to correctly identify right azygos lobe. We analyzed a total of 1709 axial thoracic multi-detector computed tomography (CT) images for the presence of an azygos lobe. Additionally, a paper-based survey was distributed among a sample of intern doctors and radiology and surgery residents, asking them to identify the right azygos lobe in a CT image and in an anatomy figure. Results showed that the prevalence of the right azygos lobe in the study sample was 0.88%. Men have more right azygos lobes than women. None of the intern doctors or surgery residents identified the right azygos lobe correctly, whereas more than half (57.1%) of the radiology residents did. Most of the incorrect answers about the CT scan were related to the bronchi (25.0%). The apex of the lung (17.7%) and the superior vena cava (17.7%) were the most common incorrect answers about the anatomy figure. In conclusion, the prevalence of the right azygos lobe in the current study is within the range of previously published literature. More education should be given for the identification of the right azygos lobe during anatomy and clinical teaching.

Computer-aided diagnosis (CAD) systems have been the focus of many researchers in both computer and medical fields. In this paper, we build two convolutional neural network (CNN) based CAD systems for diagnosing lumbar disk herniation from Magnetic Resonance Imaging (MRI) axial scans. The first one is a disk herniation detection CAD system which is a binary CAD system that determines whether the case image contains disk herniation or not. The second system is a disk herniation type classification CAD system which can determine the type of the disk herniation in the image if one exists. To train and test the proposed systems, an image set is built and annotated with the help of a radiologist. In order to get rid of the “noisy” parts of the input images and reduce their complexity, we experiment with different ROI extraction methods. The image set is also preprocessed and enlarged using augmentation techniques to make it suitable to be used with CNN. There are many novel aspects of this work. First, the problems of disk herniation detection and recognition from axial scans are not well-studied in the literature. Second, we use deep learning techniques which produces ground-breaking results in many image processing tasks, but are yet to reach their full potential with medical image processing in general. Finally, we explore the use of many innovative techniques such as data augmentation and transfer learning, which greatly improves the accuracy of our models. The results of our systems are impressive with a 95.65% accuracy for the detection problem and a 91.38% accuracy for the recognition problem.