Aldo Zaimi
Aldo Zaimi alzaia.io
Research

Research

Below are my current research interests and a selected list of papers.

Publications

Selected papers

A curated subset. For the full list, see Google Scholar.

View all on Scholar →

CableInspect-AD: An Expert-Annotated Anomaly Detection Dataset

Neurips · 2024

Machine learning models are increasingly being deployed in real-world contexts. However, systematic studies on their transferability to specific and critical applications are underrepresented in the research literature. An important example is visual anomaly detection (VAD) for robotic power line inspection. While existing VAD methods perform well in controlled environments, real-world scenarios present diverse and unexpected anomalies that current datasets fail to capture. To address this gap, we introduce CableInspect-AD, a high-quality, publicly available dataset created and annotated by domain experts from Hydro-Québec, a Canadian public utility. This dataset includes high-resolution images with challenging real-world anomalies, covering defects with varying severity levels. To address the challenges of collecting diverse anomalous and nominal examples for setting a detection threshold, we propose an enhancement to the celebrated PatchCore algorithm. This enhancement enables its use in scenarios with limited labeled data. We also present a comprehensive evaluation protocol based on cross-validation to assess models' performances. We evaluate our Enhanced-PatchCore for few-shot and many-shot detection, and Vision-Language Models for zero-shot detection. While promising, these models struggle to detect all anomalies, highlighting the dataset's value as a challenging benchmark for the broader research community.

AxonDeepSeg: automatic axon and myelin segmentation from microscopy data using convolutional neural networks

Nature Scientific Reports · 2018

Segmentation of axon and myelin from microscopy images of the nervous system provides useful quantitative information about the tissue microstructure, such as axon density and myelin thickness. This could be used for instance to document cell morphometry across species, or to validate novel non-invasive quantitative magnetic resonance imaging techniques. Most currently-available segmentation algorithms are based on standard image processing and usually require multiple processing steps and/or parameter tuning by the user to adapt to different modalities. Moreover, only a few methods are publicly available. We introduce AxonDeepSeg, an open-source software that performs axon and myelin segmentation of microscopic images using deep learning. AxonDeepSeg features: (i) a convolutional neural network architecture; (ii) an easy training procedure to generate new models based on manually-labelled data and (iii) two ready-to-use models trained from scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Results show high pixel-wise accuracy across various species: 85% on rat SEM, 81% on human SEM, 95% on mice TEM and 84% on macaque TEM. Segmentation of a full rat spinal cord slice is computed and morphological metrics are extracted and compared against the literature.

Advanced atom-level representations for protein flexibility prediction utilizing graph neural networks

Preprint · 2024

Protein dynamics play a crucial role in many biological processes and drug interactions. However, measuring, and simulating protein dynamics is challenging and time-consuming. While machine learning holds promise in deciphering the determinants of protein dynamics from structural information, most existing methods for protein representation learning operate at the residue level, ignoring the finer details of atomic interactions. In this work, we propose for the first time to use graph neural networks (GNNs) to learn protein representations at the atomic level and predict B-factors from protein 3D structures. The B-factor reflects the atomic displacement of atoms in proteins, and can serve as a surrogate for protein flexibility. We compared different GNN architectures to assess their performance. The Meta-GNN model achieves a correlation coefficient of 0.71 on a large and diverse test set of over 4k proteins (17M atoms) from the Protein Data Bank (PDB), outperforming previous methods by a large margin. Our work demonstrates the potential of representations learned by GNNs for protein flexibility prediction and other related tasks.

Construction of a rat spinal cord atlas of axon morphometry

NeuroImage · 2019

Atlases of the central nervous system are essential for understanding the pathophysiology of neurological diseases, which remains one of the greatest challenges in neuroscience research today. These atlases provide insight into the underlying white matter microstructure and have been created from a variety of animal models, including rats. Although existing atlases of the rat spinal cord provide some details of axon microstructure, there is currently no histological dataset that quantifies axon morphometry exhaustively in the entire spinal cord. In this study, we created the first comprehensive rat spinal cord atlas of the white matter microstructure with quantifiable axon and myelin morphometrics. Using full-slice scanning electron microscopy images and state-of-the-art segmentation algorithms, we generated an atlas of microstructural metrics such as axon diameter, axonal density and g-ratio. After registering the Watson spinal cord white matter atlas to our template, we computed statistics across metrics, spinal levels and tracts. We notably found that g-ratio is relatively constant, whereas axon diameter showed the greatest variation.

Specific Adsorption via Peptide Tags: Oriented Grafting and Release of Growth Factors for Tissue Engineering

Biomacromolecules · 2015

Numerous strategies have been proposed to decorate biomaterials with growth factors (GFs) for tissue engineering applications; their practicability as clinical tools, however, remains uncertain. We previously presented two complementary amphipathic peptides, namely, E5 and K5, which could be utilized as tags to direct GF capture onto organic materials via E5/K5 coiled-coil interactions. We here investigated their potential as mediators of GF physical adsorption. Enzyme-linked immunosorbent assays highlighted that both electrostatic and hydrophobic interactions could contribute to the adsorption process, without interfering with the peptides propensity for coiled-coil interactions. E5-tagged vascular endothelial growth factor, in particular, was efficiently adsorbed to poly(allylamine)-functionalized polystyrene, was maintained in a bioactive state and was steadily liberated over several days with little initial burst. This simple immobilization procedure was successfully applied to poly(ethylene terephthalate) films. Altogether, our data demonstrated that coil-tag-directed adsorption is a tunable, versatile and straightforward strategy to decorate biomaterials with GFs.