Fethallah Benmansour

Automated reconstruction of tree structures using path classifiers and Mixed Integer Programming

Although tracing linear structures in 2D images and 3D image stacks has received much attention over the years, full automation remains elusive. In this paper, we formulate the delineation problem as one of solving a Quadratic Mixed Integer Program (Q-MIP) in a graph of potential paths, which can be done optimally up to a very small tolerance. We further propose a novel approach to weighting these paths, which results in a Q-MIP solution that accurately matches the ground truth. We demonstrate that our approach outperforms a state-of-the-art technique based on the k-Minimum Spanning Tree formulation on a 2D dataset of aerial images and a 3D dataset of confocal microscopy stacks.

Reconstructing Loopy Curvilinear Structures Using Integer Programming

We propose a novel approach to automated delineation of linear structures that form complex and potentially loopy networks. This is in contrast to earlier approaches that usually assume a tree topology for the networks. At the heart of our method is an Integer Programming formulation that allows us to find the global optimum of an objective function designed to allow cycles but penalize spurious junctions and early terminations. We demonstrate that it outperforms state-of-the-art techniques on a wide range of datasets.

Detecting Irregular Curvilinear Structures in Gray Scale and Color Imagery Using Multi-directional Oriented Flux

We propose a new approach to detecting irregular curvilinear structures in noisy image stacks. In contrast to earlier approaches that rely on circular models of the cross-sections, ours allows for the arbitrarily-shaped ones that are prevalent in biological imagery. This is achieved by maximizing the image gradient flux along multiple directions and radii, instead of only two with a unique radius as is usually done. This yields a more complex optimization problem for which we propose a computationally efficient solution. We demonstrate the effectiveness of our approach on a wide range of challenging gray scale and color datasets and show that it outperforms existing techniques, especially on very irregular structures.

Reconstructing Curvilinear Networks Using Path Classifiers and Integer Programming

We propose a novel approach to automated delineation of curvilinear structures that form complex and potentially loopy networks. By representing the image data as a graph of potential paths, we first show how to weight these paths using discriminatively-trained classifiers that are both robust and generic enough to be applied to very different imaging modalities. We then present an Integer Programming approach to finding the optimal subset of paths, subject to structural and topological constraints that eliminate implausible solutions. Unlike earlier approaches that assume a tree topology for the networks, ours explicitly models the fact that the networks may contain loops, and can reconstruct both cyclic and acyclic ones. We demonstrate the effectiveness of our approach on a variety of challenging datasets including aerial images of road networks and micrographs of neural arbors, and show that it outperforms state-of-the-art techniques.

Inhibition of EGF Uptake by Nephrotoxic Antisense Drugs In Vitro and Implications for Preclinical Safety Profiling

Antisense oligonucleotide (AON) therapeutics offer new avenues to pursue clinically relevant targets inaccessible with other technologies. Advances in improving AON affinity and stability by incorporation of high affinity nucleotides, such as locked nucleic acids (LNA), have sometimes been stifled by safety liabilities related to their accumulation in the kidney tubule. In an attempt to predict and understand the mechanisms of LNA-AON-induced renal tubular toxicity, we established human cell models that recapitulate in vivo behavior of pre-clinically and clinically unfavorable LNA-AON drug candidates. We identified elevation of extracellular epidermal growth factor (EGF) as a robust and sensitive in vitro biomarker of LNA-AON-induced cytotoxicity in human kidney tubule epithelial cells. We report the time-dependent negative regulation of EGF uptake and EGF receptor (EGFR) signaling by toxic but not innocuous LNA-AONs and revealed the importance of EGFR signaling in LNA-AON-mediated decrease in cellular activity. The robust EGF-based in vitro safety profiling of LNA-AON drug candidates presented here, together with a better understanding of the underlying molecular mechanisms, constitutes a significant step toward developing safer antisense therapeutics.

On the Relevance of Sparsity for Image Classification

In this paper we empirically analyze the importance of sparsifying representations for classification purposes. We focus on those obtained by convolving images with linear filters, which can be either hand designed or learned, and perform extensive experiments on two important Computer Vision problems, image categorization and pixel classification. To this end, we adopt a simple modular architecture that encompasses many recently proposed models. The key outcome of our investigations is that enforcing sparsity constraints on features extracted in a convolutional architecture does not improve classification performance, whereas it does so when redundancy is artificially introduced. This is very relevant for practical purposes, since it implies that the expensive run-time optimization required to sparsify the representation is not always justified, and therefore that computational costs can be drastically reduced.

Automated quantification of morphodynamics for high-throughput live cell time-lapse datasets

We present a fully automatic method to track and quantify the morphodynamics of differentiating neurons in fluorescence time-lapse datasets. Previous high-throughput studies have been limited to static analysis or simple behavior. Our approach opens the door to rich dynamic analysis of complex cellular behavior in high-throughput time-lapse data. It is capable of robustly detecting, tracking, and segmenting all the components of the neuron including the nucleus, soma, neurites, and filopodia. It was designed to be efficient enough to handle the massive amount of data from a high-throughput screen. Each image is processed in approximately two seconds on a notebook computer. To validate the approach, we applied our method to over 500 neuronal differentiation videos from a small-scale RNAi screen. Our fully automated analysis of over 7,000 neurons quantifies and confirms with strong statistical significance static and dynamic behaviors that had been previously observed by biologists, but never measured.

Mechanistic Investigations of Diarrhea Toxicity Induced by anti-HER2/3 Combination Therapy

Combination of targeted therapies is expected to provide superior efficacy in the treatment of cancer either by enhanced antitumor activity or by preventing or delaying the development of resistance. Common challenges in developing combination therapies include the potential of additive and aggravated toxicities associated with pharmacologically related adverse effects. We have recently reported that combination of anti-HER2 and anti-HER3 antibodies, pertuzumab and lumretuzumab, along with paclitaxel chemotherapy in metastatic breast cancer, resulted in a high incidence of diarrhea that ultimately limited further clinical development of this combination. Here, we further dissected the diarrhea profile of the various patient dose cohorts and carried out in vitro investigations in human colon cell lines and explants to decipher the contribution and the mechanism of anti-HER2/3 therapeutic antibodies to intestinal epithelium malfunction. Our clinical investigations in patients revealed that while dose reduction of lumretuzumab, omission of pertuzumab loading dose, and introduction of a prophylactic antidiarrheal treatment reduced most severe adverse events, patients still suffered from persistent diarrhea during the treatment. Our in vitro investigations showed that pertuzumab and lumretuzumab combination treatment resulted in upregulation of chloride channel activity without indication of intestinal barrier disruption. Overall, our findings provide a mechanistic rationale to explore alternative of conventional antigut motility using medication targeting chloride channel activity to mitigate diarrhea of HER combination therapies. Mol Cancer Ther; 17(7); 1464–74. ©2018 AACR.

Automated quantification of morphodynamics for high-throughput live cell time-lapse dataset

We present a fully automatic method to track and quantify the morphodynamics of differentiating neurons in uorescence time-lapse microscopy datasets. While previous efforts have successfully quantified the dynamics of organelles such as the cell body, nucleus, or chromosomes of cultured cells, neurons have proved to be uniquely challenging due to their highly deformable neurites which expand, branch, and collapse. Our approach is capable of robustly detecting, tracking, and segmenting all the components of each neuron present in the sequence including the nucleus, soma, neurites, and filopodia. To meet the demands required for high-throughput processing, our framework is designed tobe extremely effcient, capable of processing a single image in approximately two seconds on a conventional notebook computer. For validation of our approach, we analyzed neuronal differentiation datasets in which a set of genes was perturbed using RNA interference. Our analysis confirms previous quantitative findings measured from static images, as well as previous qualitative observations of morphodynamic phenotypes that could not be measured on a large scale. Finally, we present new observations about the behavior of neurons made possible by our quantitative analysis, which are not immediately obvious to a human observer.