Ali Foroughi pour

ExpertTwins: Auditable Long Horizon Scientific Panels through Grounded Expertise

ExpertTwins is a framework for auditable scientific deliberation using evidence-based roles, showing high quotation grounding and differentiated reasoning across models and tasks.

Prediction of Outcome from Spatial Protein Profiling of Triple-Negative Breast Cancers

The paper introduces SparTile, a computational method for analyzing multiplex proteomics images to identify clinically relevant spatial microenvironments in TNBC, finding specific markers and distances associated with poor survival.

SAMPLER: Empirical distribution representations for rapid analysis of whole slide tissue images

SAMPLER provides a statistical method to generate efficient whole slide image representations, achieving comparable or better accuracy than attention models while being significantly faster.

A pan-cancer PDX histology image repository with genomic and pathological annotations for deep learning analysis

The paper presents a pan-cancer PDX histology image repository with genomic and pathological annotations for deep learning analysis of tumor regions and xenograft-transplant lymphoproliferative disorder prediction.

Deep Learning Features Encode Interpretable Morphologies within Histological Images

This paper introduces 'mones' as interpretable CNN features from H&E images that correlate with biological phenotypes and can accurately classify tumors and normals.

Deep learning-based cross-classifications reveal conserved spatial behaviors within tumor histological images

This paper uses CNNs to classify and compare histopathological images across cancer types, revealing conserved spatial patterns and mutation-related features.