Both community acquired pneumonia and diabetes mellitus are common in Bangladesh. Though hospitalization of diabetic patients with CAP is increasing, data regarding their clinical presentation, microbial characteristics, antimicrobial susceptibility and outcome are lacking. This study was aimed at finding any difference in clinical presentation, bacterial causes, antimicrobial susceptibility pattern of isolated bacteria and outcome in diabetic and non-diabetic hospitalized patients with CAP. In this study total 47 diabetic and 43 non-diabetic adult hospitalized patients with CAP were enrolled. Clinical presentation of CAP differed in diabetics and non-diabetics. Frequency of atypical presentation and CURB-65 score were significantly higher in diabetics. Pleural effusion with multilobar infiltration was also common feature for CAP in diabetic patients. Klebsiella pneumoniae was the most frequent causative pathogen for CAP in diabetic patients, whereas Streptococcus pneumoniae was the most frequent causative agent for non-diabetic patients. Bacteria isolated from sputum sample of diabetic patients with CAP were resistant to almost all recommended antibiotics used for CAP but 100% of isolates were sensitive to Carbapenems. Pulmonary complications were relatively more in diabetics than in non-diabetics. Hospitalized diabetics with CAP required referral to intensive care unit more than that of non-diabetics. So, diabetic patients with CAP need extra attention.

Objective: The objective of this study was to find out the profile of childhood cancers in National Institute of Cancer Research and Hospital, (NICRH), Dhaka during 2005 to 2009. Methodology: It was a retrospective study using hospital based cancer registry records from January 2005 to December 2009. All the children below 15 years with confirmed diagnosis of cancer by means of histological or cytological examinations were included in this study. Results: There were 28409 new confirmed cases attended out patient department of NICRH during these 5 years. Among which 1250 were below 15 years of age. An average 250 cases attended per year. Overall pediatric tumours were 4.4% of total cancers. The frequency of cancer was found to be higher among boys (62%) than girls (38%) with a ratio of 1.6:1. Majority of the children were from rural areas (67%) compared to (33%) from urban areas. The results showed that Lymphoma (24.2%), Retinoblastoma (17.4%) and Leukaemia (14.3%) were the commonly found childhood cancers among the children attended at NICRH during data collection period. Other less commonly found tumor were bone tumour (7.2%), kidney tumor (6.8%) Central Nervous System Tumour (3.7%),Testicular Tumour (3.7%), and Hepatocellular cancer (1.3%). Conclusion: Lymphoma, acute lymphoblastic leukemia and bone tumor commonly found in children above 5 years in contradiction to retinoblastoma, leukaemia and lymphoma which were prevalent in children less than 5 years of age. Key Words: Cancer registry; paediatric malignancies; cancer profile. DOI: 10.3329/jdmc.v19i1.6249 J Dhaka Med Coll. 2010; 19(1) : 33-38.

Modern software systems are increasingly including machine learning (ML) as an integral component. However, we do not yet understand the difficulties faced by software developers when learning about ML libraries and using them within their systems. To that end, this work reports on a detailed (manual) examination of 3,243 highly-rated Q&A posts related to ten ML libraries, namely Tensorflow, Keras, scikit-learn, Weka, Caffe, Theano, MLlib, Torch, Mahout, and H2O, on Stack Overflow, a popular online technical Q&A forum. We classify these questions into seven typical stages of an ML pipeline to understand the correlation between the library and the stage. Then we study the questions and perform statistical analysis to explore the answer to four research objectives (finding the most difficult stage, understanding the nature of problems, nature of libraries and studying whether the difficulties stayed consistent over time). Our findings reveal the urgent need for software engineering (SE) research in this area. Both static and dynamic analyses are mostly absent and badly needed to help developers find errors earlier. While there has been some early research on debugging, much more work is needed. API misuses are prevalent and API design improvements are sorely needed. Last and somewhat surprisingly, a tug of war between providing higher levels of abstractions and the need to understand the behavior of the trained model is prevalent.

Introduction:Malignancy is one of the leading causes of morbidity and mortality worldwide. According to GLOBOCAN 2012, an estimated 14.1 million new cancer cases and 8.2 million cancer-related deaths occurred in 2012. It is estimated that childhood malignancies are 0.5–4.6% of total malignancies. However, from the point of view of potential year lost due to childhood malignancies, it is more important than adult.Materials and Methods:To find out the probable components for the delay in diagnosis and treatment of childhood malignancies in Bangladesh, cross-sectional observational study was done at the National Institute of Cancer Research and Hospital, Dhaka, Bangladesh, from January 2014 to June 2014.Results:A total of 171 patients were included in the study. They were divided into four age groups. The mean age was 8.422 years with standard deviation ± 5.381 years and their age ranged from 2 months to 18 years. In aggregate, about 70% of the cases had to wait for more than 90 days for the treatment. About 15% had to wait for 31–60 days. Negligible percentage of patients got treatment before 30 days. Among the three components of delay, patients delay was influenced by age of the child, economic status of the family, parental education, and awareness of the parents about malignancy.Conclusion:More than one-third of the pediatric patients had to wait three months or more for treatment to start for various reasons. By raising awareness among the stake holders this problem can be minimized. Further studies are recommended to explore the other factors which might cause delayed referral.

Variation in the electromyogram pattern recognition (EMG-PR) performance with the muscle contraction force is a key limitation of the available prosthetic hand. To alleviate this problem, we propose a scheme to realize electromyogram signal normalization across channels before feature extraction. The proposed signal normalization scheme is validated over a dataset of nine transradial amputees that includes three force levels with six hand gestures. Moreover, we employ three classifiers, namely, linear discriminant analysis (LDA), support vector machine (SVM) and k-nearest neighbour (KNN), to evaluate the EMG-PR performance. In addition to the signal normalization scheme, we perform nonlinear transformation of the features by using the logarithm function. Both schemes facilitate merging of the muscle activation patterns of different force levels. The experimental results indicate that the force invariant EMG-PR performance (F1 score of at least 3.24% to 4.34%) of the proposed schemes is significantly enhanced compared to that obtained in recent studies. Therefore, we recommend using these features along with the proposed signal normalization scheme and nonlinear transformation of the features to improve the force invariant EMG-PR performance. The proposed feature extraction method achieves the highest F1 score of 91.28%, 91.39% and 90.56% when using the LDA, SVM and KNN classifiers, respectively.INDEX TERMS EMG Pattern recognition, Force invariant features, Muscle activation pattern, Signal normalization.