Aim of this study is to use machine learning approaches for predicting blood glucose based on basic non-invasive health checkup test results, dietary information, and socio-demographic characteristics and to develop a web application to predict blood glucose easily. We evaluated the performance of five widely used machine learning models. Data have been collected from 271 employees of Grameen Bank complex, in Dhaka, Bangladesh. This study used continuous blood glucose data to train the model and predicted new blood glucose values using the trained data. Finally, we developed a blood glucose prediction web application. The Boosted Decision Tree Regression model showed the best performance among other models based on the Root Mean Squared Error (RMSE) 2.30, this RMSE is better than any reported in the literature. This study developed a blood glucose prediction model and web application which is easier, more convenient, and more efficient for people. People can also easily check their blood glucose values using our app, especially in remote areas of developing countries that lack adequate skilled doctors and nurses. By predicting blood glucose, this study can help to save medical costs and time and to reduce health management costs. Our system can be helpful in achieving SDGs, Universal Health Coverage and thus reducing overall morbidity and mortality.

No abstract

Acne scarring occurs in 95% of people with acne vulgaris due to collagen loss or gains when the body is healing the damages of the skin caused by acne inflammation. Accurate classification of acne scars is a vital factor in providing a timely, effective treatment protocol. Dermatologists mainly recognize the type of acne scars manually based on visual inspections, which are time-and energy-consuming and subject to intra-and inter-reader variability. In this paper, a novel automated acne scar classification system is proposed based on a deep Convolutional Neural Network (CNN) model. First, a dataset of 250 images from five different classes is collected and labeled by a well-experienced dermatologist. The pre-processed input images are fed into our proposed model, namely ScarNet, for deep feature map extraction. The optimizer, loss function, activation functions, filter and kernel sizes, regularization methods, and the batch size of the proposed architecture are tuned so that the classification performance is maximized while minimizing the computational cost. Experimental results demonstrate the feasibility of the proposed method with accuracy, specificity, and kappa score of 92.53%, 95.38%, and 76.7%, respectively.

No abstract

No abstract