Classification of skin cancer from dermoscopic images using deep neural network architectures.

A powerful medical decision support system for classifying skin lesions from dermoscopic images is an important tool to prognosis of skin cancer. In the recent years, Deep Convolutional Neural Network (DCNN) have made a significant advancement in detecting skin cancer types from dermoscopic images, in-spite of its fine grained variability in its appearance. The main objective of this research work is to develop a DCNN based model to automatically classify skin cancer types into melanoma and non-melanoma with high accuracy. The datasets used in this work were obtained from the popular challenges ISIC-2019 and ISIC-2020, which have different image resolutions and class imbalance problems. To address these two problems and to achieve high performance in classification we have used EfficientNet architecture based on transfer learning techniques, which learns more complex and fine grained patterns from lesion images by automatically scaling depth, width and resolution of the network. We have augmented our dataset to overcome the class imbalance problem and also used metadata information to improve the classification results. Further to improve the efficiency of the EfficientNet we have used ranger optimizer which considerably reduces the hyper parameter tuning, which is required to achieve state-of-the-art results. We have conducted several experiments using different transferring models and our results proved that EfficientNet variants outperformed in the skin lesion classification tasks when compared with other architectures. The performance of the proposed system was evaluated using Area under the ROC curve (AUC - ROC) and obtained the score of 0.9681 by optimal fine tuning of EfficientNet-B6 with ranger optimizer.

A computational model for enhancing recombinant Penicillin G Acylase production from Escherichia coli DH5α.

An attempt was made to develop a computational model based on artificial neural network and ant colony optimization to estimate the composition of medium components for maximizing the productivity of Penicillin G Acylase (PGA) enzyme from Escherichia coli DH5α strain harboring the plasmid pPROPAC. As a first step, an artificial neural network (ANN) model was developed to predict the PGA activity by considering the concentrations of seven important components of the medium. Design of experiments employing central composite design technique was used to obtain the training samples. In the second step, ant colony optimization technique for continuous domain was employed to maximize the PGA activity by finding the optimal inputs for the developed ANN model. Further, the effect of a combination of ant colony optimization for continuous domain with a preferential local search strategy was studied to analyze the performance. For a comparative study, the training samples were fed into the response surface methodology optimization software to maximize the PGA production. The obtained PGA activity (56.94 U/mL) by the proposed approach was found to be higher than that of the obtained value (45.60 U/mL) with the response surface methodology. The optimum solution obtained computationally was experimentally verified. The observed PGA activity (55.60 U/mL) exhibited a close agreement with the model predictions.