Search

Question Answering (QA) system is the task where arbitrary question IS posed in the form of natural language statements and a brief and concise text returned as an answer. Contrary to search engines where a long list of relevant documents returned as a result of a query, QA system aims at providing the direct answer or passage containing the answer. We propose a general purpose question answering system which can answer wh-interrogated questions. This system is using Wikipedia data as its knowledge source. We have implemented major components of a QA system which include challenging tasks of Named Entity Tagging, Question Classification, Information Retrieval and Answer Extraction. Implementation of state-of-the-art Entity Tagging mechanism has helped identify entities where systems like OpenEphyra or DBpedia spotlight have failed. The information retrieval task includes development of a framework to extract tabular information known as Infobox from Wikipedia pages which has ensured availability of latest updated information. Answer Extraction module has implemented an attributes mapping mechanism which is helpful to extract answer from data. The system is comparable in results with other available general purpose QA systems.

The aim of this study was to record the outcome of healing in laparotomy wounds, managed by delayed versus primary skin closure in terms of hospital stay, major and minor wound infection. Design and Duration: Quasi experimental study from September, 2006 to March, 2007. Setting: Surgical floor of Allied Hospital, (Punjab Medical College) Faisalabad. Methodology: Detailed data of each patient including presentation, operative findings, procedure performed, post operative outcome was entered on a specially designed proforma. The main outcome measures found significant were major and minor wound infection, time of presentation and advancing age and hospital stay. Results: Sixty patients underwent exploratory laparotomy through vertical abdominal incision during Sep.2006 to Mar.2007. Skin wound of the first thirty patients (group A) were left open and closed on 4th day while that of next thirty patients (group B) closed primarily. Out of sixty patients ten patients developed major wound infection leading to wound dehiscence (16.66 %). Four belonged to group A(13.33%) and six belonged to group B(20 %) (p

The purpose of the study is to highlight the varied presentation of tuberculosis (TB) simulating a brain tumour. Headache and seizures are becoming frequent presenting complaints without any history of tuberculosis. The study comprises 1200 patients of both sexes with ages ranging from ten to sixty years. CT scan and MRI brain control with and without contrast medium were the investigations performed in these cases. In some patients Electroencephalography (EEG), cerebral angiography (DSA) and spectroscopy were also performed. The final diagnosis of tuberculosis was made on the basis of craniotomy, stereotactic and burr hole biopsies with histopathology in most of the cases. Forty per cent of the patients were followed up for eight months. They were put on anti-tuberculosis treatment with symptomatic and anti-epileptic drugs. The incidence was 544 and 757 per 100,000 in Africa and Indo Pakistan respectively. The male to female ratio was 1:1. Tuberculosis, especially with CNS involvement, is not only common in immunosuppressed patients in our setting, but TB has been and remains an important public health problem. TB may involve the CNS either as meningitis or as parenchymal granulomas or abscesses. Patients with brain TB usually present with fever, multiple cranial nerve involvement and occasional behavioural changes. CSF findings remain non specific in most cases. The most common sites are the cerebral hemisphere and basal ganglion in adults and the cerebellum in children. Tuberculosis has unique findings on brain CT and MRI. Cortical and subcortical locations are typical whereas the brain stem is a less common site. Tuberculosis lesions are usually solitary but multiple in 10% to 35% of cases. In spite of all these facts some cases of brain TB still need aggressive neurointervention to reach the final diagnosis of brain TB. Tuberculosis in the CNS may manifest in many different ways. So one should always include tuberculosis in the differential diagnosis in the etiology of delayed onset epilepsy and acute on chronic headache. In case of a discrepancy between clinical manifestations and CT/MRI findings, one can always anticipate tuberculous lesion in the brain.

We share our experience of 2000 spinal tuberculosis (TB) cases, including both typical and atypical presentations. The aim of the study is to estimate the incidence and types of spinal TB referred to our department for diagnosis. 2000 patients were selected by convenience sampling from January 2006 to September 2010. Study design was descriptive and among 2000 mostly unknown cases without evidence of symptoms of systemic TB (1080 males and 920 females). MRI without and with IV contrast, CECT with MPR, and in some cases not fit for MRI, CT myelography, were performed. Out of 2000 cases of tuberculous spine, 1080 (54%) were male and 920 (46%) were female. Their age ranged from 8–60 years. About 90% of patients were below the age of 40 years. Peak age among the males and females was 20–29 years and 14–35 years, respectively. The most common site of involvement was dorsal spine (45%) followed by lumbo-sacral spine (33%), cervical spine (10%) and at multiple levels (12%). Biopsies were done in 240 (12%) cases. Spinal TB should always be suspected when radiographs demonstrate a destructive spinal process. Awareness and prompt management of TB spine will help in reducing the continuing morbidity of this disease.

To compare the neurological outcome of microsurgical clipping versus coiling in patients with anterior circulation aneurysm.Comparative study.Department of Neurosurgery, Lahore General Hospital, Lahore, from January 2010 to December 2013.Patients aged 14 - 60 years, with ruptured cerebral aneurysm of anterior circulation and World Federation of Neurosurgical Society (WFNS) grades 1, 2 and 3 were included. Patients more than 60 years, medically unfit patient and posterior circulation aneurysms and WFNS grades 4 and 5 were excluded. Aneurysm sac obliteration was done in randomized manner with microsurgical clipping or coiling. Postoperatively, the patients were assessed and followed-up upto one year for outcome parameters on the bases of WFNS grade and Modified Ranking Scale (mRS) as favourable (mRS ≤2 ) and unfavourable (mRS2).Among 140 subjects selected for study, 70 were included in group A, i.e. coiling and other 70 were in group B, i.e. clipping. The median age of patients in group A was 52.5 ± 10 years and in group B was 51.00 ± 10 years. Overall, 56 (40%) males, 28 (60%) males in each group; and 84 (60%) females, 42 (60%) in each group were included. The male to female ratio in this study was 1:1.5. In group A, i.e. coiling, 27 (38.6%) patients had no disability (grades 1 and 2), 25 (35.7%) were slightly disabled (grade 3) and 18 (25.7%) had moderate disability (grade 4); whereas in group B, i.e. clipping group 23 (32.9%) patients had no disability (grades 1 and 2), 23 (32.9%) were slightly disabled (grade 3) and 24 (34.3%) had moderate disability (grade 4). At one year follow-up, in group A, favourable outcome was achieved in 56 (80%) of patients compared to 48 (68.6%) in group B; whilst, 14 (20%) patients in group Aand 22 (33.1%) in group B showed unfavourable outcome. Although mortality rate was higher in clipping (n=3, 4.3%) as compared to coiling (n=1, 1.4%), but was not statistically significant (p = 0.310).Endovascular coiling of anterior circulation aneurysms is safe and as effective and successful as aneurysm clipping and is less invasive also.