Search

Summary Background Screening for Barrett's oesophagus relies on endoscopy, which is invasive and few who undergo the procedure are found to have the condition. We aimed to use machine learning techniques to develop and externally validate a simple risk prediction panel to screen individuals for Barrett's oesophagus. Methods In this prospective study, machine learning risk prediction in Barrett's oesophagus (MARK-BE), we used data from two case-control studies, BEST2 and BOOST, to compile training and validation datasets. From the BEST2 study, we analysed questionnaires from 1299 patients, of whom 880 (67·7%) had Barrett's oesophagus, including 40 with invasive oesophageal adenocarcinoma, and 419 (32·3%) were controls. We randomly split (6:4) the cohort using a computer algorithm into a training dataset of 776 patients and a testing dataset of 523 patients. We compiled an external validation cohort from the BOOST study, which included 398 patients, comprising 198 patients with Barrett's oesophagus (23 with oesophageal adenocarcinoma) and 200 controls. We identified independently important diagnostic features of Barrett's oesophagus using the machine learning techniques information gain and correlation-based feature selection. We assessed multiple classification tools to create a multivariable risk prediction model. Internal validation of the model using the BEST2 testing dataset was followed by external validation using the BOOST external validation dataset. From these data we created a prediction panel to identify at-risk individuals. Findings The BEST2 study included 40 diagnostic features. Of these, 19 added information gain but after correlation-based feature selection only eight showed independent diagnostic value including age, sex, cigarette smoking, waist circumference, frequency of stomach pain, duration of heartburn and acidic taste, and taking antireflux medication, of which all were associated with increased risk of Barrett's oesophagus, except frequency of stomach pain, with was inversely associated in a case-control population. Logistic regression offered the highest prediction quality with an area under the receiver-operator curve (AUC) of 0·87 (95% CI 0·84–0·90; sensitivity set at 90%; specificity of 68%). In the testing dataset, AUC was 0·86 (0·83–0·89; sensitivity set at 90%; specificity of 65%). In the external validation dataset, the AUC was 0·81 (0·74–0·84; sensitivity set at 90%; specificity of 58%). Interpretation Our diagnostic model offers valid predictions of diagnosis of Barrett's oesophagus in patients with symptomatic gastro-oesophageal reflux disease, assisting in identifying who should go forward to invasive confirmatory testing. Our predictive panel suggests that overweight men who have been taking antireflux medication for a long time might merit particular consideration for further testing. Our risk prediction panel is quick and simple to administer but will need further calibration and validation in a prospective study in primary care. Funding Charles Wolfson Charitable Trust and Guts UK.

Most research in robotic swarms assumes a benign operational environment for the swarm. This paper follows on from previous work on attacks on a robotic swarms and presents an analysis of how Intrusion Detection Systems (IDS) may be applied to protect robotic swarms. This paper shows that due to the characteristics of a swarm, anomaly-based intrusion detection techniques cannot be applied. Signature based techniques however can be used. The idea of an IDS Swarm is introduced and it is demonstrated that a signature based IDS can used to protect a swarm in a hostile environment.

Much research has been undertaken in the area of swarm robotics and applications the vast majority of which assumes a benign operational environment. We assume a hostile environment and review the threats to robotic swarms. We then define an adversary’s capabilities and classify generic attack types and attack vectors that may be undertaken by this adversary. We conclude that while the threats to a swarm are no different to those for traditional systems, there are differences with the attack types and attack vectors. This is primarily due to the ability to manipulate the emergent behaviour of the swarm by attacking individual swarm elements. We identify characteristics of the swarm that allow such attacks to take place.

Work has been undertaken in the research of swarm robotics and potential applications. This has included several specific models and simulations, with the majority of the work assuming a benign operational environment for the swarm. This paper proposes a generic model to represent a swarm and then adapts the generic model to consider malicious elements within the swarm.

Barrett's esophagus is an increasingly common disease that is strongly associated with reflux of stomach acid and usually a hiatus hernia, and it strongly predisposes to esophageal adenocarcinoma (EAC), a tumor with a very poor prognosis. We report the first genome-wide association study on Barrett's esophagus, comprising 1,852 UK cases and 5,172 UK controls in the discovery stage and 5,986 cases and 12,825 controls in the replication stage. Variants at two loci were associated with disease risk: chromosome 6p21, rs9257809 (P combined = 4.09 × 10-9; odds ratio (OR) = 1.21, 95% confidence interval (CI) =1.13-1.28), within the major histocompatibility complex locus, and chromosome 16q24, rs9936833 (P combined = 2.74 × 10-10; OR = 1.14, 95% CI = 1.10-1.19), for which the closest protein-coding gene is FOXF1, which is implicated in esophageal development and structure. We found evidence that many common variants of small effect contribute to genetic susceptibility to Barrett's esophagus and that SNP alleles predisposing to obesity also increase risk for Barrett's esophagus. © 2012 Nature America, Inc. All rights reserved.

Road traffic collisions (RTCs) are one of the most common preventable causes of death and disability worldwide. We investigated changes in numbers of motor vehicles, case fatality rate, and crash injury rate for the most present recorded year (2002) 5 and 10 years before that in the United Kingdom (UK) and Sri Lanka (SL). We also investigated environmental and individual factors impacting patients at South Birmingham Trauma Unit, UK and Colombo General Hospital, SL. We conducted a descriptive cross-sectional study (both quantitative and qualitative). Over the 10-year period, numbers of motor vehicles have risen in both countries; the crash injury remained stable in both countries. Case fatality rate (far higher) in SL has decreased, as in the UK. Three hundred and twenty-five patients took part in the survey in SL, with 83 in the UK. In the categories investigated, including patient demographics, RTC environment, visual impairment, pedestrian and driver factors, the majority of results were significantly different between the two countries. Target factors such as inadequate street lighting, visual impairment, speeding, and not wearing seatbelts at time of accident were identified, and recommendations developed as a result.

Arterial injury is most commonly seen with penetrating injuries. An arterial disruption in a closed fracture is rare. It has been reported in only 0.3% of long bone fractures. The authors report a case of femoral artery injury in a polytrauma patient associated with a closed femoral fracture. The arterial injury was not apparent until the patient had an above-knee guillotine amputation done to revise his ipsilateral below-knee traumatic amputation, which released the soft -tissue tamponade effect on the leak. Unless the possibility of this injury is kept in mind, above-knee amputation or any other open procedure on a patient with a femoral fracture can cause exsanguination of the patient.