Advancements in Computational Pathology

The field of computational pathology is rapidly evolving, with a focus on developing innovative methods for image analysis and disease diagnosis. Recent studies have explored the potential of vision transformers and convolutional neural networks for lymphoma diagnosis, with promising results. The use of pathology foundation models has also shown significant improvements in image segmentation and classification tasks. Additionally, researchers have investigated the application of deep learning techniques for tumor detection, cell detection, and gene expression prediction. Notable papers in this area include:

  • A study that compared the performance of vision transformers and convolutional neural networks for lymphoma diagnosis, demonstrating comparable accuracy between the two models.
  • A survey that presented a hierarchical taxonomy for organizing pathology foundation models, providing a framework for analyzing and evaluating these models.
  • A paper that proposed a novel approach for histopathology image segmentation using latent diffusion models, achieving significant improvements over traditional methods.
  • A study that developed a parameter-efficient knowledge transfer framework for predicting gene expression from digitalized histopathology images, outperforming baseline models and alternative fine-tuning strategies.

Sources

Artificial intelligence application in lymphoma diagnosis: from Convolutional Neural Network to Vision Transformer

A Survey of Pathology Foundation Model: Progress and Future Directions

Evaluation framework for Image Segmentation Algorithms

Training state-of-the-art pathology foundation models with orders of magnitude less data

An ensemble deep learning approach to detect tumors on Mohs micrographic surgery slides

Towards Varroa destructor mite detection using a narrow spectra illumination

Hybrid CNN with Chebyshev Polynomial Expansion for Medical Image Analysis

PathSegDiff: Pathology Segmentation using Diffusion model representations

A Comparison of Deep Learning Methods for Cell Detection in Digital Cytology

Teaching pathology foundation models to accurately predict gene expression with parameter efficient knowledge transfer

Built with on top of