AI in cancer detection

Pixel-Precise Lesion Localization in WSIs via Weakly Supervised Streaming Convolution with ReLSE and Adaptive Self-Training

The paper improves lesion localization in whole-slide images using weakly supervised streaming convolution with ReLSE and adaptive self-training, achieving higher recall than strong supervision.

HIPI: Spatially Resolved Multiplexed Protein Expression Inferred from H&E WSIs

HIPI predicts cell marker expression from H&E images to infer spatial protein expression in tumors.

DEEP LEARNING APPROACH FOR RENAL CELL CARCINOMA DETECTION,SUBTYPING, AND GRADING

This paper introduces an end-to-end deep learning pipeline for renal cell carcinoma detection, subtyping, and grading using The Cancer Genome Atlas data.

CLEAR-IT: Contrastive Learning to Capture the Immune Composition of Tumor Microenvironments

CLEAR-IT improves cell phenotyping in tumor microenvironments by using contrastive learning, outperforming existing methods across imaging modalities and requiring fewer labels.

Towards Digital Quantification of Ploidy from Pan-Cancer Digital Pathology Slides using Deep Learning

PloiViT is a transformer-based model that quantifies tumor ploidy from pathology slides, outperforming traditional methods and achieving low prediction error across multiple cancer types.

High-Performance Classification of Breast CancerHistopathological Images Using Fine-TunedVision Transformers on the BreakHis Dataset

The paper fine-tunes a Vision Transformer on the BreakHis dataset to achieve state-of-the-art performance in breast cancer histopathological image classification.

Spatial Immunophenotyping from Whole-Slide Multiplexed Tissue Imaging Using Convolutional Neural Networks

CellGate automates spatial immunophenotyping of multiplexed tissue images with high accuracy, enabling scalable analysis of tumor-immune microenvironments.

AnNoBrainer, an Automated Annotation of Mouse Brain Images using Deep Learning

AnNoBrainer automates mouse brain image annotation using deep learning, improving accuracy, reproducibility, and efficiency compared to manual methods.

Exploit Spatially Resolved Transcriptomic Data to Infer Cellular Features from Pathology Imaging Data}

This paper introduces STpath, a transfer-learning model that uses spatially resolved transcriptomic data to annotate pathology images and classify tumor microenvironments, showing promising performance with limited training data.

RAF2Net: Automated grading of Renal cell Carcinoma utilizing Attention-enhanced deep learning models through Feature Fusion

RAF2Net is a deep learning model for automated RCC grading using attention-enhanced feature fusion, achieving over 92% accuracy on the KMC dataset.

Deep Learning-based Modeling for Preclinical Drug Safety Assessment

This paper introduces TRACE, a deep learning model for automated liver histopathology assessment in preclinical drug safety, achieving higher concordance than pathologists in lesion scoring and dose-response analysis.

Development of a Deep Learning model Tailored for HER2 Detection in Breast Cancer to aid pathologists in interpreting HER2-Low cases

This paper develops a deep learning model to detect HER2 status in breast cancer, showing higher concordance with ground truth than pathologists for HER2-low cases.

Self-supervised learning for characterising histomorphological diversity and spatial RNA expression prediction across 23 human tissue types

This paper uses self-supervised learning to analyze histology images and predict spatial RNA expression across 23 human tissues, revealing population variability and its impact on gene expression.

Synthetic Genitourinary Image Synthesis via Generative Adversarial Networks: Enhancing AI Diagnostic Precision

This paper generates synthetic genitourinary images using GANs to improve diagnostic model performance, showing high quality and pathologist approval.

MurSS: Multi-resolution Selective Segmentation Model for Breast Cancer

MurSS improves breast cancer segmentation by combining multi-resolution context and content information, achieving higher accuracy than existing models.

An Introductory Implementation of Breast Cancer Detection from Mammograms and Pixel Intensity with Efficient-Net Other Neural Nets

The paper uses deep learning models to classify mammograms as normal or abnormal using the MIAS dataset, achieving accuracy in breast cancer detection.

Graph attention-based fusion of pathology images and gene expression for prediction of cancer survival

This paper develops a graph attention-based framework that fuses pathology images and gene expression data to predict cancer survival, showing improved performance over existing methods.

Deep Learning Glioma Grading with the Tumor Microenvironment Analysis Protocol for A Comprehensive Learning, Discovering, and Quantifying Microenvironmental Features

This paper develops a deep learning model for glioma grading using a tumor microenvironment analysis protocol, achieving 0.85 accuracy on a small, imbalanced dataset with single-stained biopsies.

Deep learning-based Spatial Feature Extraction for Prognostic Prediction of Hepatocellular Carcinoma from Pathological Images

This study develops a deep learning method to extract spatial features from H&E images of HCC, identifying features associated with patient survival and improving prognostic stratification when combined with clinical features.

SN-FPN: Self-attention Nested Feature Pyramid Network for Digital Pathology Image Segmentation

The paper introduces a deep learning model for digital pathology image segmentation that outperforms existing methods in tumor and stromal region segmentation.

Enhancing Whole Slide Image Classification with Discriminative and Contrastive Learning

This study improves whole slide image classification by using discriminative and contrastive learning to create WSI-level positive and negative samples, enhancing classification performance and feature robustness.

Democratizing computational pathology: optimized Whole Slide Image representations for The Cancer Genome Atlas

This paper introduces Giga-SSL, a self-supervised method for learning Whole Slide Image representations from The Cancer Genome Atlas data, improving classification performance and enabling efficient analysis.

Multiplexed tumor profiling with generative AI accelerates histopathology workflows and improves clinical predictions

This paper introduces VirtualMultiplexer, a generative AI tool that translates H&E images to IHC images, improving clinical predictions and accelerating histopathology workflows.

A Multi-Granularity Approach to Similarity Search in Multiplexed Immunofluorescence Images

The paper introduces MIISS, a framework for similarity search in multiplexed immunofluorescence images, showing domain-specific models outperform others.

3D multiplexed tissue imaging reconstruction and optimized region-of-interest (ROI) selection through deep learning model of channels embedding

The paper proposes deep learning methods to reconstruct 3D CyCIF images and select optimal ROIs from histology and MTI data for tumor representation.

Whole slide image representation in bone marrow cytology

This paper develops a deep learning method to generate compact whole slide image representations for bone marrow cytology, achieving improved diagnostic prediction and image retrieval.

Digital profiling of cancer transcriptomes from histology images with grouped vision attention

This paper introduces SEQUOIA, a transformer model that predicts cancer transcriptomes from histology images, achieving high accuracy in predicting gene expression and recurrence risk in breast cancer.

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.

Synthetic whole-slide image tile generation with gene expression profiles infused deep generative models

This paper introduces RNA-GAN, a deep generative model that creates synthetic whole-slide image tiles using gene expression profiles, outperforming traditional GANs and showing imputation capabilities.

Deep-learning Based Pathological Assessment of Frozen Procurement Kidney Biopsies Predicts Graft Loss and Guides Organ Utilization: A Large-scale Retrospective Study

This paper develops a deep-learning model to assess kidney biopsy images, predicting graft loss and guiding organ utilization by improving upon traditional scoring methods.