Hybrid COVID-19 segmentation and recognition framework (HMB-HCF) using deep learning and genetic algorithms.

COVID-19 (Coronavirus) went through a rapid escalation until it became a pandemic disease. The normal and manual medical infection discovery may take few days and therefore computer science engineers can share in the development of the automatic diagnosis for fast detection of that disease. The study suggests a hybrid COVID-19 framework (named HMB-HCF) based on deep learning (DL), genetic algorithm (GA), weighted sum (WS), and majority voting principles in nine phases. Its segmentation phase suggests a lung segmentation algorithm using X-Ray images (named HMB-LSAXI) for extracting lungs. Its classification phase is built from a hybrid convolutional neural network (CNN) architecture using an abstractly-designed CNN (named HMB1-COVID19) and transfer learning (TL) pre-trained models (VGG16, VGG19, ResNet50, ResNet101, Xception, DenseNet121, DenseNet169, MobileNet, and MobileNetV2). The hybrid CNN architecture is used for learning, classification, and parameters optimization while GA is used to optimize the hyperparameters. This hybrid working mechanism is combined in an overall algorithm named HMB-DLGA. The study experiments implemented the WS approach to evaluate the models' performance using the loss, accuracy, F1-score, precision, recall, and area under curve (AUC) metrics with different pre-defined ratios. A collected, combined, and unified X-Ray dataset from 8 different public datasets was used alongside the regularization, dropout, and data augmentation techniques to limit the overall overfitting. The applied experiments reported state-of-the-art metrics. VGG16 reported 100% WS metric (i.e., 0.0097, 99.78%, 0.9984, 99.89%, 99.78%, and 0.9996 for the loss, accuracy, F1, precision, recall, and AUC respectively) concerning the highest WS. It also reported a 99.92% WS metric (i.e., 0.0099, 99.84%, 0.9984, 99.84%, 99.84%, and 0.9996 for the loss, accuracy, F1, precision, recall, and AUC respectively) concerning the last reported WS result. HMB-HCF was validated on 13 different public datasets to verify its generalization. The best-achieved metrics were compared with 13 related studies. These extensive experiments' target was the applicability verification and generalization.

English language proficiency and academic performance: A study of a medical preparatory year program in Saudi Arabia.

INTRODUCTION: All medical schools in Saudi Arabia have English as the primary official medium of instruction. Most of the high school education, however, is delivered in Arabic and hence the transition to an English based learning environment tends to be difficult for some students. Our study aims to correlate English language proficiency with academic performance among medical students in their preparatory year. METHODS: A cross-sectional study design was used. Test scores of 103 preparatory year students (54 female and 49 male) were analyzed after the students completed an English language course and medical introductory course in their preparatory year. The total score obtained in the English course assessment was compared to each component of the medical content assessment. RESULTS: A significantly positive correlation (Spearman's Rho, at 0.01 levels) was seen between the scores of the English exam and the written exam (P <0.001) and the oral exam (P = -0.003) parts respectively of the medical examination. Significant correlation with the English exam score was not obtained for the other components of the medical assessment, namely; student assignments, presentations and portfolios. CONCLUSION: English language proficiency is an important factor in determining academic proficiency of medical students in our college at the preparatory year level.

A framework for a competency based medical curriculum in saudi arabia.

BACKGROUND: We recently adopted a competency based curriculum based on the CanMEDs model. This shift required the cross-mapping of all key CanMEDs competencies with the competencies for higher education in Saudi Arabia as per the Saudi National Commission for Academic Accreditation & Assessment (NCAAA) guidelines. OBJECTIVES: To formulate competencies for our curriculum and to create a framework aligned with NCAAA, CanMEDs and Saudi Meds. METHODS: After finalization of program outcomes, the program goals were cross-mapped with CanMEDs and Saudi Meds competencies and then the CanMEDs competencies were reverse mapped with our outcomes. Finally benchmarking of outcomes with the programs of the Universities of Manitoba and Toronto was done. RESULTS: We were able to cross-map and match major outcomes of our program with both the CanMEDs and the Saudi Meds frameworks, ensuring that the outcomes are in line with NCAAA, CanMEDs and Saud Meds. Also, our program objectives were bench marked with two of the Canadian medical schools. CONCLUSION: We propose that our framework can be a model for other universities in Saudi Arabia to consider when shifting to a competency based curriculum.

Assessment of learning style in a sample of saudi medical students.

CONFLICT OF INTEREST: NONE DECLARED. BACKGROUND: By knowing the different students' learning styles, teachers can plan their instruction carefully in ways that are capitalized on student preferences. The current research is done to determine specific learning styles of students. METHOD: This cross sectional study was conducted in Al Ahsa College of Medicine from 2011 to 2012. A sample of 518 students completed a questionnaire based on Kolb inventory (LSI 2) to determine their learning style. A spreadsheet was prepared to compute all the information to get the cumulative scores of learning abilities and identify the learning styles. RESULTS: The mean values of the learning abilities; active experimentation (AE), reflective observation (RO), abstract conceptualizing (AC) or concrete experience (CE) for male students were 35, 28, 30 and 26 respectively while they were 31, 30, 31 and 29 respectively for female students. There were significant difference between male and female students regarding the mean values of AE-RO (6.7 vs 1.5) and AC-CE (4.1 vs 2.1). This indicated that the style of male students were more convergent and accommodating than those of female students. The female had more assimilating and divergent styles. CONCLUSION: Learning style in Saudi medical students showed difference between males and females in the early college years. Most male students had convergent and accommodating learning styles, while the female dominant learning styles were divergent and assimilating. Planning and implementation of instruction need to consider these findings.

Maternal obesity predict isolated birth defects in live births in Eastern Province of Saudi Arabia.

OBJECTIVE: There is no research on the predictors of birth defects in Al Ahsa Governorate in the Eastern Province of Saudi Arabia. The aim of this research was to detect the predictors of isolated structural birth defects in live births. METHODS: We conducted this study from April 2006 to 2010. Live births with isolated birth defects represented our sample for this retrospective case control study. Univariate analysis was done for all possible risk factors. Logistic regression analysis was done for all predictors in relation to different birth defects. RESULTS: Out of 37168 live births, isolated structural birth defects were found in 318 cases. Obesity ( body mass index > 30) was a significant predictor for increased nervous system anomalies ( odds ratio (OR): 7.83, CI: 3.9-15.4), facial defects (OR: 5.92, CI: 2.8-12.4), genitourinary anomalies (OR: 4.6 CI: 1.9-11.1), and cardiac malformations (OR: 2.7 CI: 1.3-5.7). Consanguinity increased the risk for cardiac malformations (OR: 3.32, CI: 1.54-7.17). Low socio-economic status increased the risk for nervous system anomalies (OR: 2.09, CI: 1.18-3.7), facial defects (OR: 2.33, CI: 1.25-4.33) and musculoskeletal anomalies (OR: 2.3, CI: 1.29-4.09). CONCLUSION: Maternal obesity represented the most common predictor for certain categories of isolated structural birth defects including nervous system, facial, genitourinary and cardiac.

Cytological pattern of cervical Papanicolaou smear in eastern region of Saudi Arabia.

BACKGROUND: Cancer of the cervix has been considered as one of the preventable cancers. This study is the first published research addressing the screening of cancer of the cervix in the eastern region of Saudi Arabia. AIM: This study aims to detect the prevalence of abnormal epithelial changes and its types in the eastern region of Saudi Arabia. SETTINGS AND STUDY DESIGN: A retrospective study was designed to evaluate all previously conducted cervical smears examined at a secondary care maternity hospital in Saudi Arabia, during the period from 2003 to 2010. During this period, a total of 1171 smears were reported. MATERIALS AND METHODS: We analyzed the records of all patients who had undergone Papanicolaou (Pap) smear during this period. After data collection, all cases were recorded as per Bethesda nomenclature. RESULTS: A total of 624 (53.3%) abnormal Pap smears were found, with only 58 cases reported to have epithelial pathological diagnosis (SIL). They represented 4.95% of total taken smears. A majority of the SIL diagnoses in our population were ASCUS, representing 60% of SIL cases. The prevalence of squamous cervical carcinoma was 0.34%. CONCLUSION: The study has shown a relatively high prevalence of epithelial abnormalities in cervical smears in the studied population. The squamous cell carcinoma represented a higher than the overall prevalence compared to World Health Organization (WHO) factsheets about Saudi Arabia. The mean age of epithelial abnormalities and squamous cell carcinoma was in the reproductive years.

Risk factors and birth prevalence of birth defects and inborn errors of metabolism in Al Ahsa, Saudi Arabia.

BACKGROUND: Birth defects and inborn errors of metabolism are related to variable poor perinatal and neonatal outcomes. Our aim was to explore the pattern and prevalence of birth defects and metabolic birth errors in Al-Ahsa Governorate in the Eastern Province of Saudi Arabia. METHODS: This retrospective case control study was done from April 2006 to 2009. Children with any birth defect or metabolic errors of metabolism at birth or in the neonatology section were our sample for study. Control group was randomly selected from the cases with normal live births. Blood tests were performed for children suspected to suffer from genetic blood disorders. The principal BD as per the International Classification of Diseases-10 (ICD-10) code was also noted. RESULTS: Out of 38001 live births, birth defects were found in 1.14% and errors of metabolism were detected in 0.17%. The most common birth defects were craniofacial malformations. The 3-methylcrotonyl-CoA carboxylase deficiency was the most common inborn errors of metabolism. Consanguinity, rural residence and prematurity were associated with significant rise in birth defects. On the other hand, consanguinity and low birth weight were associated with significant rise in metabolic errors. CONCLUSION: First cousins consanguinity represented the most significant risk factor for birth defects and inborn errors of metabolism. High degree of inbreeding, consanguinity may exacerbate underlying recessive genetic risk factors.

The phenomenology of premenstrual syndrome in female medical students: a cross sectional study.

BACKGROUND: The premenstrual syndrome (PMS) is particularly common in the younger age groups and, therefore represents a significant public health problem in young girls. This study aims to estimate the prevalence, severity, determinants of premenstrual syndrome (PMS) and its impact among the female medical students in Al-Ahsa, Saudi Arabia. METHODS: This study was performed at the College of Medicine, King Faisal University, Saudi Arabia, from June through December 2009. It included 250 medical students. They filled different questionnaires covering American College of Obstetrics and Gynecology (ACOG) criteria to diagnose PMS, demographic & reproductive factors, physical activity and mental condition. Regression analysis was conducted for all the predictors. RESULTS: PMS was diagnosed in 35.6% of cases, distributed as 45% mild, 32.6% moderate and 22.4% severe. There were significant trends for older age, rural residence, family income and family history of PMS. The dominant limited activity was concentration in class (48.3%). Limitations of activities were significantly more frequent among severe cases. The prevalence of anxiety and depression was statistically more evident in the PMS group. Regression analysis revealed that, PMS was significantly associated with older age groups, rural residence, lower age at menarche, regularity of menses and family history. CONCLUSION: PMS is a common problem in young Saudi students in Al Ahsa. Severe PMS was associated with more impairment of daily activities and psychological distress symptoms. Older student age, rural residence, earlier age of menarche, regular cycles and positive family history are possible risk factors for PMS.

Minor psychiatric morbidity in young saudi mothers using Mini International Neuropsychiatric Interview (MINI).

OBJECTIVE: To determine the prevalence of postpartum psychiatric disorders in young mothers using a brief structured psychiatric interview, the Mini International Neuropsychiatric Interview (MINI). STUDY DESIGN: A cross-sectional study. PLACE AND DURATION OF STUDY: The Primary Health Care Centres (PHCCs) in Al-Ahsa region, Saudi Arabia, during the period 2008 to 2009. METHODOLOGY: Young primigravid mothers (in their teen age) were interviewed using MINI. Psychiatric morbidity was analyzed in relation to the different sociodemographic, psychiatric and obstetric characteristics. The distribution and frequency of the MINI subscales were presented in number and percentage. To quantify the risk, Univariate analysis was employed with reporting of crude Odds Ratio (OR) and 95% Confidence Intervals (CI); p-value of < 0.05 was considered significant. RESULTS: The prevalence of psychiatric disorders was 22.6% with preponderance of anxiety disorders due to increased prevalence of generalized anxiety disorder and social phobia. Postpartum anxiety disorders were significantly associated with urban residency, poor husband support, past history of psychiatric illness, anemia, caesarean mode of delivery and female baby gender. CONCLUSION: These results highlighted importance of addressing screening for psychiatric morbidity particularly anxiety disorders in the implementation of perinatal care for the pregnant Saudi adolescents.