Classification of scanning electron microscope images of pharmaceutical excipients using deep convolutional neural networks with transfer learning

Convolutional Neural Networks (CNNs) are image analysis techniques that have been applied to image classification in various fields. In this study, we applied a CNN to classify scanning electron microscopy (SEM) images of pharmaceutical raw material powders to determine if a CNN can evaluate particle morphology. We tested 10 pharmaceutical excipients with widely different particle morphologies. SEM images for each excipient were acquired and divided into training, validation, and test sets. Classification models were constructed by applying transfer learning to pretrained CNN models such as VGG16 and ResNet50. The results of a 5-fold cross-validation showed that the classification accuracy of the CNN model was sufficiently high using either pretrained model and that the type of excipient could be classified with high accuracy. The results suggest that the CNN model can detect differences in particle morphology, such as particle size, shape, and surface condition. By applying Grad-CAM to the constructed CNN model, we succeeded in finding particularly important regions in the particle image of the excipients. CNNs have been found to have the potential to be applied to the identification and characterization of raw material powders for pharmaceutical development.

Predicting PBT and CMR properties of substances of very high concern (SVHCs) using QSAR models, and application for K-REACH

•

BIOWIN is effective for predicting persistence and bioaccumulation.

•

Toxtree is effective for predicting carcinogenicity and mutagenicity.

•

WoE approach enhances the sensitivity.

•

It is recommended to set a conservative criteria of log Kow more than 4.5 in K-REACH.

Quantitative structure-activity relationship (QSAR) models have been applied to predict a variety of toxicity endpoints. Their performance needs to be validated, in a variety of cases, to increase their applicability to chemical regulation. Using the data set of substances of very high concern (SVHCs), the performance of QSAR models were evaluated to predict the persistence and bioaccumulation of PBT, and the carcinogenicity and mutagenicity of CMR. BIOWIN and Toxtree showed higher sensitivity than other QSAR models – the former for persistence and bioaccumulation, the latter for carcinogenicity. In terms of mutagenicity, the sensitivities of QSAR models were underestimated, Toxtree was more accurate and specific than lazy structure–activity relationships (LAZARs) and Computer Assisted Evaluation of industrial chemical Substances According to Regulations (CAESAR). Using the weight of evidence (WoE) approach, which integrates results of individual QSAR models, enhanced the sensitivity of each toxicity endpoint. On the basis of obtained results, in particular the prediction of persistence and bioaccumulation by KOWWIN, a conservative criterion is recommended of log Kow greater than 4.5 in K-REACH, without an upper limit. This study suggests that reliable production of toxicity data by QSAR models is facilitated by a better understanding of the performance of these models.

Identifying natural compounds as multi-target-directed ligands against Alzheimer's disease: an in silico approach

Alzheimer's disease (AD) is a multi-factorial disease, which can be simply outlined as an irreversible and progressive neurodegenerative disorder with an unclear root cause. It is a major cause of dementia in old aged people. In the present study, utilizing the structural and biological activity information of ligands for five important and mostly studied vital targets (i.e. cyclin-dependant kinase 5, β-secretase, monoamine oxidase B, glycogen synthase kinase 3β, acetylcholinesterase) that are believed to be effective against AD, we have developed five classification models using linear discriminant analysis (LDA) technique. Considering the importance of data curation, we have given more attention towards the chemical and biological data curation, which is a difficult task especially in case of big data-sets. Thus, to ease the curation process we have designed Konstanz Information Miner (KNIME) workflows, which are made available at http://teqip.jdvu.ac.in/QSAR_Tools/ . The developed models were appropriately validated based on the predictions for experiment derived data from test sets, as well as true external set compounds including known multi-target compounds. The domain of applicability for each classification model was checked based on a confidence estimation approach. Further, these validated models were employed for screening of natural compounds collected from the InterBioScreen natural database ( https://www.ibscreen.com/natural-compounds ). Further, the natural compounds that were categorized as 'actives' in at least two classification models out of five developed models were considered as multi-target leads, and these compounds were further screened using the drug-like filter, molecular docking technique and then thoroughly analyzed using molecular dynamics studies. Finally, the most potential multi-target natural compounds against AD are suggested.

A review on automatic analysis techniques for color fundus photographs

In this paper, we give a review on automatic image processing tools to recognize diseases causing specific distortions in the human retina. After a brief summary of the biology of the retina, we give an overview of the types of lesions that may appear as biomarkers of both eye and non-eye diseases. We present several state-of-the-art procedures to extract the anatomic components and lesions in color fundus photographs and decision support methods to help clinical diagnosis. We list publicly available databases and appropriate measurement techniques to compare quantitatively the performance of these approaches. Furthermore, we discuss on how the performance of image processing-based systems can be improved by fusing the output of individual detector algorithms. Retinal image analysis using mobile phones is also addressed as an expected future trend in this field.

Value of shared preclinical safety studies - The eTOX database

•

First analysis of the eTOX database for 1214 drugs or drug candidates.

•

Shared data mainly from short term <20 days preclinical studies in rat via oral route.

•

Identified the most frequent treatment related findings.

•

Evaluated predictivity of clinical chemistry biomarkers.

•

Present a first use case of the database during early drug development.

A first analysis of a database of shared preclinical safety data for 1214 small molecule drugs and drug candidates extracted from 3970 reports donated by thirteen pharmaceutical companies for the eTOX project (www.etoxproject.eu) is presented. Species, duration of exposure and administration route data were analysed to assess if large enough subsets of homogenous data are available for building in silico predictive models. Prevalence of treatment related effects for the different types of findings recorded were analysed. The eTOX ontology was used to determine the most common treatment-related clinical chemistry and histopathology findings reported in the database. The data were then mined to evaluate sensitivity of established in vivo biomarkers for liver toxicity risk assessment. The value of the database to inform other drug development projects during early drug development is illustrated by a case study.