Publications by authors named "Imtiaz Hossain"

6 Publications

  • Page 1 of 1

HistoNet: A Deep Learning-Based Model of Normal Histology.

Toxicol Pathol 2021 Mar 3:192623321993425. Epub 2021 Mar 3.

Novartis Institutes for BioMedical Research, Basel, Switzerland.

We introduce HistoNet, a deep neural network trained on normal tissue. On 1690 slides with rat tissue samples from 6 preclinical toxicology studies, tissue regions were outlined and annotated by pathologists into 46 different tissue classes. From these annotated regions, we sampled small 224 × 224 pixels images (patches) at 6 different levels of magnification. Using 4 studies as training set and 2 studies as test set, we trained VGG-16, ResNet-50, and Inception-v3 networks separately at each magnification level. Among these model architectures, Inception-v3 and ResNet-50 outperformed VGG-16. Inception-v3 identified the tissue from query images, with an accuracy up to 83.4%. Most misclassifications occurred between histologically similar tissues. Investigation of the features learned by the model (embedding layer) using Uniform Manifold Approximation and Projection revealed not only coherent clusters associated with the individual tissues but also subclusters corresponding to histologically meaningful structures that had not been annotated or trained for. This suggests that the histological representation learned by HistoNet could be useful as the basis of other machine learning algorithms and data mining. Finally, we found that models trained on rat tissues can be used on non-human primate and minipig tissues with minimal retraining.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/0192623321993425DOI Listing
March 2021

Biomarker-Based Classification and Localization of Renal Lesions Using Learned Representations of Histology-A Machine Learning Approach to Histopathology.

Toxicol Pathol 2021 Feb 24:192623320987202. Epub 2021 Feb 24.

Novartis Institutes for Biomedical Research (NIBR), Basel, Switzerland.

Several deep learning approaches have been proposed to address the challenges in computational pathology by learning structural details in an unbiased way. Transfer learning allows starting from a learned representation of a pretrained model to be directly used or fine-tuned for a new domain. However, in histopathology, the problem domain is tissue-specific and putting together a labelled data set is challenging. On the other hand, whole slide-level annotations, such as biomarker levels, are much easier to obtain. We compare two pretrained models, one histology-specific and one from ImageNet on various computational pathology tasks. We show that a domain-specific model (HistoNet) contains richer information for biomarker classification, localization of biomarker-relevant morphology within a slide, and the prediction of expert-graded features. We use a weakly supervised approach to discriminate slides based on biomarker level and simultaneously predict which regions contribute to that prediction. We employ multitask learning to show that learned representations correlate with morphological features graded by expert pathologists. All of these results are demonstrated in the context of renal toxicity in a mechanistic study of compound toxicity in rat models. Our results emphasize the importance of histology-specific models and their knowledge representations for solving a wide range of computational pathology tasks.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/0192623320987202DOI Listing
February 2021

Morphological Deconvolution of Beta-Lactam Polyspecificity in E. coli.

ACS Chem Biol 2019 06 11;14(6):1217-1226. Epub 2019 Jun 11.

Infectious Diseases , Novartis Institutes for BioMedical Research , Emeryville , California , United States.

Beta-lactams comprise one of the earliest classes of antibiotic therapies. These molecules covalently inhibit enzymes from the family of penicillin-binding proteins (PBPs), which are essential in construction of the bacterial cell wall. As a result, beta-lactams cause striking changes to cellular morphology, the nature of which varies by the range of PBPs simultaneously engaged in the cell. The traditional method of exploring beta-lactam polyspecificity is a gel-based binding assay which is low-throughput and typically is run  ex situ in cell extracts. Here, we describe a medium-throughput, image-based assay combined with machine learning methods to automatically profile the activity of beta-lactams in E. coli cells. By testing for morphological change across a panel of strains with perturbations to individual PBP enzymes, our approach automatically and quantifiably relates different beta-lactam antibiotics according to their preferences for individual PBPs in cells. We show the potential of our approach for guiding the design of novel inhibitors toward different PBP-binding profiles by predicting the mechanisms of two recently reported PBP inhibitors.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1021/acschembio.9b00141DOI Listing
June 2019

A multi-scale convolutional neural network for phenotyping high-content cellular images.

Bioinformatics 2017 Jul;33(13):2010-2019

Novartis Institutes for BioMedical Research Inc., Basel, Switzerland.

Motivation: Identifying phenotypes based on high-content cellular images is challenging. Conventional image analysis pipelines for phenotype identification comprise multiple independent steps, with each step requiring method customization and adjustment of multiple parameters.

Results: Here, we present an approach based on a multi-scale convolutional neural network (M-CNN) that classifies, in a single cohesive step, cellular images into phenotypes by using directly and solely the images' pixel intensity values. The only parameters in the approach are the weights of the neural network, which are automatically optimized based on training images. The approach requires no a priori knowledge or manual customization, and is applicable to single- or multi-channel images displaying single or multiple cells. We evaluated the classification performance of the approach on eight diverse benchmark datasets. The approach yielded overall a higher classification accuracy compared with state-of-the-art results, including those of other deep CNN architectures. In addition to using the network to simply obtain a yes-or-no prediction for a given phenotype, we use the probability outputs calculated by the network to quantitatively describe the phenotypes. This study shows that these probability values correlate with chemical treatment concentrations. This finding validates further our approach and enables chemical treatment potency estimation via CNNs.

Availability And Implementation: The network specifications and solver definitions are provided in Supplementary Software 1.

Contact: william_jose.godinez_navarro@novartis.com or xian-1.zhang@novartis.com.

Supplementary Information: Supplementary data are available at Bioinformatics online.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1093/bioinformatics/btx069DOI Listing
July 2017

Jenkins-CI, an Open-Source Continuous Integration System, as a Scientific Data and Image-Processing Platform.

SLAS Discov 2017 03 13;22(3):238-249. Epub 2016 Dec 13.

3 Developmental and Molecular Pathways, NIBR, Postfach, Basel, Switzerland.

High-throughput screening generates large volumes of heterogeneous data that require a diverse set of computational tools for management, processing, and analysis. Building integrated, scalable, and robust computational workflows for such applications is challenging but highly valuable. Scientific data integration and pipelining facilitate standardized data processing, collaboration, and reuse of best practices. We describe how Jenkins-CI, an "off-the-shelf," open-source, continuous integration system, is used to build pipelines for processing images and associated data from high-content screening (HCS). Jenkins-CI provides numerous plugins for standard compute tasks, and its design allows the quick integration of external scientific applications. Using Jenkins-CI, we integrated CellProfiler, an open-source image-processing platform, with various HCS utilities and a high-performance Linux cluster. The platform is web-accessible, facilitates access and sharing of high-performance compute resources, and automates previously cumbersome data and image-processing tasks. Imaging pipelines developed using the desktop CellProfiler client can be managed and shared through a centralized Jenkins-CI repository. Pipelines and managed data are annotated to facilitate collaboration and reuse. Limitations with Jenkins-CI (primarily around the user interface) were addressed through the selection of helper plugins from the Jenkins-CI community.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1177/1087057116679993DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5322829PMC
March 2017

The structure of the cornified claw sheath in the domesticated cat (Felis catus): implications for the claw-shedding mechanism and the evolution of cornified digital end organs.

J Anat 2009 Apr;214(4):620-43

Department of Biological Sciences, Louisiana State University, Baton Rouge, 70803-1715, USA.

The morphology of cornified structures is notoriously difficult to analyse because of the extreme range of hardness of their component tissues. Hence, a correlative approach using light microscopy, scanning electron microscopy, three-dimensional reconstructions based on x-ray computed tomography data, and graphic modeling was applied to study the morphology of the cornified claw sheath of the domesticated cat as a model for cornified digital end organs. The highly complex architecture of the cornified claw sheath is generated by the living epidermis that is supported by the dermis and distal phalanx. The latter is characterized by an ossified unguicular hood, which overhangs the bony articular base and unguicular process of the distal phalanx and creates an unguicular recess. The dermis covers the complex surface of the bony distal phalanx but also creates special structures, such as a dorsal dermal papilla that points distally and a curved ledge on the medial and lateral sides of the unguicular process. The hard-cornified external coronary horn and proximal cone horn form the root of the cornified claw sheath within the unguicular recess, which is deeper on the dorsal side than on the medial and lateral sides. As a consequence, their rate of horn production is greater dorsally, which contributes to the overall palmo-apical curvature of the cornified claw sheath. The external coronary and proximal cone horn is worn down through normal use as it is pushed apically. The hard-cornified apical cone horn is generated by the living epidermis enveloping the base and free part of the dorsal dermal papilla. It forms nested horn cones that eventually form the core of the hardened tip of the cornified claw. The sides of the cornified claw sheath are formed by the newly described hard-cornified blade horn, which originates from the living epidermis located on the slanted face of the curved ledge. As the blade horn is moved apically, it entrains and integrates the hard-cornified parietal horn on its internal side. It is covered by the external coronary and proximal cone horn on its external side. The soft-cornified terminal horn extends distally from the parietal horn and covers the dermal claw bed at the tip of the uniguicular process, thereby filling the space created by the converging apical cone and blade horn. The soft-cornified sole horn fills the space between the cutting edges of blade horn on the palmar side of the cornified claw sheath. The superficial soft-cornified perioplic horn is produced on the internal side of the unguicular pleat, which surrounds the root of the cornified claw sheath. The shedding of apical horn caps is made possible by the appearance of microcracks in the superficial layers of the external coronary and proximal cone horn in the course of deformations of the cornified claw sheath, which is subjected to tensile forces during climbing or prey catching. These microcracks propagate tangentially through the coronary horn and do not injure the underlying living epidermal and dermal tissues. This built-in shedding mechanism maintains sharp claw tips and ensures the freeing of the claws from the substrate.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1111/j.1469-7580.2009.01068.xDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2736126PMC
April 2009