arXiv subjects

Yosef E. Maruvka

Publications and source records attributed to Yosef E. Maruvka.

8 recordsLinked to original sources

arXiv2025

PSA-MIL: A Probabilistic Spatial Attention-Based Multiple Instance Learning for Whole Slide Image Classification

Whole Slide Images (WSIs) are high-resolution digital scans widely used in medical diagnostics. WSI classification is typically approached using Multiple Instance Learning (MIL), where the slide is partitioned into tiles treated as interconnected instances. While attention-based MIL methods aim to identify the most informative tiles, they often fail to fully exploit the spatial relationships among them, potentially overlooking intricate tissue structures crucial for accurate diagnosis. To address this limitation, we propose Probabilistic Spatial Attention MIL (PSA-MIL), a novel attention-based MIL framework that integrates spatial context into the attention mechanism through learnable distance-decayed priors, formulated within a probabilistic interpretation of self-attention as a posterior distribution. This formulation enables a dynamic inference of spatial relationships during training, eliminating the need for predefined assumptions often imposed by previous approaches. Additionally, we suggest a spatial pruning strategy for the posterior, effectively reducing self-attention's quadratic complexity. To further enhance spatial modeling, we introduce a diversity loss that encourages variation among attention heads, ensuring each captures distinct spatial representations. Together, PSA-MIL enables a more data-driven and adaptive integration of spatial context, moving beyond predefined constraints. We achieve state-of-the-art performance across both contextual and non-contextual baselines, while significantly reducing computational costs.

cs.CV↗

arXiv2024

CIMIL-CRC: a clinically-informed multiple instance learning framework for patient-level colorectal cancer molecular subtypes classification from H\&E stained images

Treatment approaches for colorectal cancer (CRC) are highly dependent on the molecular subtype, as immunotherapy has shown efficacy in cases with microsatellite instability (MSI) but is ineffective for the microsatellite stable (MSS) subtype. There is promising potential in utilizing deep neural networks (DNNs) to automate the differentiation of CRC subtypes by analyzing Hematoxylin and Eosin (H\&E) stained whole-slide images (WSIs). Due to the extensive size of WSIs, Multiple Instance Learning (MIL) techniques are typically explored. However, existing MIL methods focus on identifying the most representative image patches for classification, which may result in the loss of critical information. Additionally, these methods often overlook clinically relevant information, like the tendency for MSI class tumors to predominantly occur on the proximal (right side) colon. We introduce `CIMIL-CRC', a DNN framework that: 1) solves the MSI/MSS MIL problem by efficiently combining a pre-trained feature extraction model with principal component analysis (PCA) to aggregate information from all patches, and 2) integrates clinical priors, particularly the tumor location within the colon, into the model to enhance patient-level classification accuracy. We assessed our CIMIL-CRC method using the average area under the curve (AUC) from a 5-fold cross-validation experimental setup for model development on the TCGA-CRC-DX cohort, contrasting it with a baseline patch-level classification, MIL-only approach, and Clinically-informed patch-level classification approach. Our CIMIL-CRC outperformed all methods (AUROC: $0.92\pm0.002$ (95\% CI 0.91-0.92), vs. $0.79\pm0.02$ (95\% CI 0.76-0.82), $0.86\pm0.01$ (95\% CI 0.85-0.88), and $0.87\pm0.01$ (95\% CI 0.86-0.88), respectively). The improvement was statistically significant.

cs.CV↗

arXiv2024

Multi-Cohort Framework with Cohort-Aware Attention and Adversarial Mutual-Information Minimization for Whole Slide Image Classification

Whole Slide Images (WSIs) are critical for various clinical applications, including histopathological analysis. However, current deep learning approaches in this field predominantly focus on individual tumor types, limiting model generalization and scalability. This relatively narrow focus ultimately stems from the inherent heterogeneity in histopathology and the diverse morphological and molecular characteristics of different tumors. To this end, we propose a novel approach for multi-cohort WSI analysis, designed to leverage the diversity of different tumor types. We introduce a Cohort-Aware Attention module, enabling the capture of both shared and tumor-specific pathological patterns, enhancing cross-tumor generalization. Furthermore, we construct an adversarial cohort regularization mechanism to minimize cohort-specific biases through mutual information minimization. Additionally, we develop a hierarchical sample balancing strategy to mitigate cohort imbalances and promote unbiased learning. Together, these form a cohesive framework for unbiased multi-cohort WSI analysis. Extensive experiments on a uniquely constructed multi-cancer dataset demonstrate significant improvements in generalization, providing a scalable solution for WSI classification across diverse cancer types. Our code for the experiments is publicly available at .

eess.IV↗

arXiv2024

Exploring the Interplay Between Colorectal Cancer Subtypes Genomic Variants and Cellular Morphology: A Deep-Learning Approach

Molecular subtypes of colorectal cancer (CRC) significantly influence treatment decisions. While convolutional neural networks (CNNs) have recently been introduced for automated CRC subtype identification using H&E stained histopathological images, the correlation between CRC subtype genomic variants and their corresponding cellular morphology expressed by their imaging phenotypes is yet to be fully explored. The goal of this study was to determine such correlations by incorporating genomic variants in CNN models for CRC subtype classification from H&E images. We utilized the publicly available TCGA-CRC-DX dataset, which comprises whole slide images from 360 CRC-diagnosed patients (260 for training and 100 for testing). This dataset also provides information on CRC subtype classifications and genomic variations. We trained CNN models for CRC subtype classification that account for potential correlation between genomic variations within CRC subtypes and their corresponding cellular morphology patterns. We assessed the interplay between CRC subtypes' genomic variations and cellular morphology patterns by evaluating the CRC subtype classification accuracy of the different models in a stratified 5-fold cross-validation experimental setup using the area under the ROC curve (AUROC) and average precision (AP) as the performance metrics. Combining the CNN models account for variations in CIMP and SNP further improved classification accuracy (AUROC: 0.847$\pm$0.01 vs. 0.787$\pm$0.03, p$=$0.01, AP: 0.68$\pm$0.02 vs. 0.64$\pm$0.05).

cs.CV↗

arXiv2022

Patient-level Microsatellite Stability Assessment from Whole Slide Images By Combining Momentum Contrast Learning and Group Patch Embeddings

Assessing microsatellite stability status of a patient's colorectal cancer is crucial in personalizing treatment regime. Recently, convolutional-neural-networks (CNN) combined with transfer-learning approaches were proposed to circumvent traditional laboratory testing for determining microsatellite status from hematoxylin and eosin stained biopsy whole slide images (WSI). However, the high resolution of WSI practically prevent direct classification of the entire WSI. Current approaches bypass the WSI high resolution by first classifying small patches extracted from the WSI, and then aggregating patch-level classification logits to deduce the patient-level status. Such approaches limit the capacity to capture important information which resides at the high resolution WSI data. We introduce an effective approach to leverage WSI high resolution information by momentum contrastive learning of patch embeddings along with training a patient-level classifier on groups of those embeddings. Our approach achieves up to 7.4\% better accuracy compared to the straightforward patch-level classification and patient level aggregation approach with a higher stability (AUC, $0.91 \pm 0.01$ vs. $0.85 \pm 0.04$, p-value$<0.01$). Our code can be found at https://github.com/TechnionComputationalMRILab/colorectal_cancer_ai.

cs.CV↗

arXiv2010

The Birth-Death-Mutation process: a new paradigm for fat tailed distributions

Fat tailed statistics and power-laws are ubiquitous in many complex systems. Usually the appearance of of a few anomalously successful individuals (bio-species, investors, websites) is interpreted as reflecting some inherent "quality" (fitness, talent, giftedness) as in Darwin's theory of natural selection. Here we adopt the opposite, "neutral", outlook, suggesting that the main factor explaining success is merely luck. The statistics emerging from the neutral birth-death-mutation (BDM) process is shown to fit marvelously many empirical distributions. While previous neutral theories have focused on the power-law tail, our theory economically and accurately explains the entire distribution. We thus suggest the BDM distribution as a standard neutral model: effects of fitness and selection are to be identified by substantial deviations from it.

cond-mat.stat-mech↗

arXiv2009

Universal features of surname distribution in a subsample of a growing population

We examine the problem of family size statistics (the number of individuals carrying the same surname, or the same DNA sequence) in a given size subsample of an exponentially growing population. We approach the problem from two directions. In the first, we construct the family size distribution for the subsample from the stable distribution for the full population. This latter distribution is calculated for an arbitrary growth process in the limit of slow growth, and is seen to depend only on the average and variance of the number of children per individual, as well as the mutation rate. The distribution for the subsample is shifted left with respect to the original distribution, tending to eliminate the part of the original distribution reflecting the small families, and thus increasing the mean family size. From the subsample distribution, various bulk quantities such as the average family size and the percentage of singleton families are calculated. In the second approach, we study the past time development of these bulk quantities, deriving the statistics of the genealogical tree of the subsample. This approach reproduces that of the first when the current statistics of the subsample is considered. The surname distribution from th e 2000 U.S. Census is examined in light of these findings, and found to misrepresent the population growth rate by a factor of 1000.

q-bio.PE↗

arXiv2005

Nonlocal competition and logistic growth: patterns, defects and fronts

Logistic growth of diffusing reactants on spatial domains with long range competition is studied. The bifurcations cascade involved in the transition from the homogenous state to a spatially modulated stable solution is presented, and a distinction is made between a modulated phase, dominated by single or few wavenumbers, and the spiky phase, where localized colonies are separated by depleted region. The characteristic defects in the periodic structure are presented for each phase, together with the invasion dynamics in case of local initiation. It is shown that the basic length scale that controls the bifurcation is the width of the Fisher front, and that the total population grows as this width decreases. A mix of analytic results and extensive numerical simulations yields a comprehensive examination of the possible phases for logistic growth in the presence of nonlocal competition.

nlin.PS↗