The development of a machine learning algorithm for early detection of viral hepatitis B infection in Nigerian patients.

Access to Hepatitis B Virus (HBV) testing for people in low-resource settings has long been challenging due to the gold standard, enzyme immunoassay, being prohibitively expensive, and requiring specialised skills and facilities that are not readily available, particularly in remote and isolated laboratories. Routine pathology data in tandem with cutting-edge machine learning shows promising diagnostic potential. In this study, recursive partitioning ("trees") and Support Vector Machines (SVMs) were applied to interrogate patient dataset (n = 916) that comprised results for Hepatitis B Surface Antigen (HBsAg) and routine clinical chemistry and haematology blood tests. These algorithms were used to develop a predictive diagnostic model of HBV infection. Our SVM-based diagnostic model of infection (accuracy = 85.4%, sensitivity = 91%, specificity = 72.6%, precision = 88.2%, F1-score = 0.89, Area Under the Receiver Operating Curve, AUC = 0.90) proved to be highly accurate for discriminating HBsAg positive from negative patients, and thus rivals with immunoassay. Therefore, we propose a predictive model based on routine blood tests as a novel diagnostic for early detection of HBV infection. Early prediction of HBV infection via routine pathology markers and pattern recognition algorithms will offer decision-support to clinicians and enhance early diagnosis, which is critical for optimal clinical management and improved patient outcomes.

Urban-Rural Differences in the Epidemiology of Asthma and Allergies in Nigeria: A Population-Based Study.

PURPOSE: Urbanization is associated with the risk of developing allergic conditions. Few studies have evaluated the urban-rural disparity of allergic diseases in sub-Saharan Africa. OBJECTIVE: To compare the epidemiology of adult asthma and allergies in urban and rural Nigeria. SUBJECTS AND METHODS: A population-based cross-sectional study was performed among 910 subjects in Kwara State, North Central Nigeria, comprising 635 urban and 275 rural adults who were randomly selected. We used standardized questionnaires for data collection. RESULTS: The age-adjusted prevalence of adults reporting a previous "asthma attack" or "currently taking asthma medication" within the preceding 12 months (ECRHS asthma definition) was 3.4% urban, 0.5% rural, current allergic rhinoconjunctivitis (26.2% urban, 22.2% rural), and current skin allergy (13.9% urban, 10.5% rural). The age-adjusted prevalence of "physician-diagnosed allergic conditions": asthma (3.3% urban, 1.5% rural), allergic rhinoconjunctivitis (4.9% urban, 3.2% rural), and skin allergy (4.8% urban, 4.6% rural) were higher in urban areas than in rural areas. Urban areas recorded a higher age-adjusted 12 months prevalence of wheezing, night waking by breathlessness, night waking by chest tightness, asthma attack (p=0.042), and current use of asthma medication (p=0.031) than the rural areas. In the urban areas, 81% of those with asthma significantly had current allergic rhinoconjunctivitis, and 40.5% had current skin allergy, whereas in the rural areas, all subjects with asthma had current allergic rhinoconjunctivitis and 12.5% had current skin allergy (p=0.482). The most common trigger for asthma attack/respiratory symptoms among the urban household was exposure to environmental smoke (17.2%), and among the rural household, it was dust exposure (18.2%). Living in urban areas significantly increased the odds of having asthma [aOR: 5.6 (95% CI:1.6-19.6)] and allergic rhinoconjunctivitis [aOR: 1.7 (95% CI: 1.2-2.4)]. CONCLUSION: This study shows that urban residents frequently reported more allergic and respiratory symptoms and were at risk of having asthma and allergic rhinitis compared to rural residents. The findings would assist the physicians in understanding the urban-rural differences in the occurrence of allergic conditions, symptom triggers, and comorbidity, which are relevant in patient's clinical evaluation, treatment, and disease prevention.

Oesophageal intramural pseudodiverticulosis: a rare endoscopic finding.

Oesophageal intramural pseudodiverticulosis is an uncommon condition, characterized by multiple small rounded cavities seen in the oesophageal wall during oesophagogastroduodenoscopy. It is often associated with gastro-oesophageal reflux disease, achalasia, oesophageal candidiasis and diabetes mellitus. We report a 40 year old Nigerian man who presented with recurrent dysphagia and endoscopic findings typical of oesophageal intramural pseudodiverticulosis. The patient was managed medically with resolution of the dysphagia. This report highlights the occurrence of this rare and benign cause of dysphagia in Nigeria. FUNDING: None declared.

Prevalence of minimal hepatic encephalopathy among patients with chronic liver disease in Ilorin, Nigeria.

BACKGROUND: Minimal Hepatic encephalopathy is the mildest form of Hepatic Encephalopathy which presents with significant cognitive impairment and affectation of activities of daily living. The literature is scanty on the prevalence of minimal hepatic encephalopathy in Nigerians with chronic liver disease. AIM: This study aimed at determining the prevalence of minimal hepatic encephalopathy among patients with chronic liver disease using neuro-psychometric tests. METHODS: The study was a hospital-based cross-sectional study carried out at the University of Ilorin Teaching Hospital, Ilorin from February 2015 to February 2016. Chronic liver disease was diagnosed with the presence of peripheral stigmata of liver disease, liver biochemistry, prothrombin time, and sonographic findings in keeping with liver disease. Minimal hepatic encephalopathy was diagnosed using number connection tests-A and B for patients who were educated while Line tracing test and constructional dyspraxia were used for patients without any formal education. Data obtained were analysed using Statistical Package for the Social Sciences (SPSS) version 20 computer software package. RESULTS: Sixty-four patients with chronic liver disease were recruited. The mean age (SD) of the patients was 47.1±14.6 yrs, and the 30-39 and 40-49 yrs age groups each had the highest frequency of 21(32.8%). There were 54 (84.4%) males and 10 (15.6%) females. The prevalence of Minimal Hepatic Encephalopathy was 43.8%. CONCLUSION: The prevalence of Minimal Hepatic Encephalopathy in this study was similar to previous studies. Significant number of patients with minimal HE were in Child-Pugh class B and C. FUNDING: None.

A symptom-based model to predict colorectal cancer in low-resource countries: Results from a prospective study of patients at high risk for colorectal cancer.

BACKGROUND: Colorectal cancer (CRC) rates in low-resource countries, which typically lack CRC screening programs, are rising. This study determined whether a risk model for patients with rectal bleeding could identify patients with curable CRC. METHODS: This prospective, cross-sectional study evaluated a model constructed from data from 1 hospital and validated at 2 other hospitals. The primary endpoint was the ability of the model to predict CRC, as diagnosed by colonoscopy, from clinical characteristics. The secondary endpoint was to determine the percentage of patients who had CRC. RESULTS: Consecutive patients who were 45 years old or older and had self-reported rectal bleeding for more than 1 week were evaluated. From January 2014 to July 2016, 362 patients answered a questionnaire and underwent colonoscopy. In the validation cohort, 56% of patients with rectal bleeding, weight loss, and changes in bowel habits had CRC, whereas 2% of patients with bleeding alone did. Overall, 18.2% of the patients had CRC, and 8.6% had adenomas. The proportion of CRC patients with potentially curable stage II or III disease was 74%, whereas the historical rate was 36%. The combination of rectal bleeding with both symptoms significantly predicted CRC in the validation set (odds ratio, 12.8; 95% confidence interval, 4.6-35.4; P < .001). CONCLUSIONS: In low-resource settings, patients with rectal bleeding, weight loss, and changes in bowel habits should be classified as high risk for CRC. Patients with a high risk score should be prioritized for colonoscopy to increase the number of patients diagnosed with potentially curable CRC. Cancer 2018;124:2766-2773. © 2018 American Cancer Society.