This study explores the efficacy of diffusion probabilistic models for generating synthetic histopathological images, specifically canine Perivascular Wall Tumours (cPWT), to supplement limited datasets for deep learning applications in digital pathology. This research evaluates an open-source medical domain-focused diffusion model called Medfusion, where the model was trained on a small (1,000 patches) and a large dataset (17,000 patches) of cPWT images to compare performance on the different sized datasets. A Receiver Operating Characteristic (ROC) study was implemented to investigate the ability of six veterinary medical professionals and pathologists to discern between generated and real cPWT patch images. The participants engaged in two separate rounds, where each round corresponded to models that had been trained on the two different sized datasets. The ROC study revealed mean average Area Under the Curve (AUC) values close to 0.5 for both rounds. The results from this study suggests that diffusion models can create histopathological patch images that are convincingly realistic where our participants often struggled to reliably differentiate between generated and real images. This underscores the potential of these models as a valuable tool for augmenting digital pathology datasets.

Evaluating diffusion model generated synthetic histopathology image data against authentic digital pathology images

Gola, Cecilia;Ressel, Lorenzo;
2024-01-01

Abstract

This study explores the efficacy of diffusion probabilistic models for generating synthetic histopathological images, specifically canine Perivascular Wall Tumours (cPWT), to supplement limited datasets for deep learning applications in digital pathology. This research evaluates an open-source medical domain-focused diffusion model called Medfusion, where the model was trained on a small (1,000 patches) and a large dataset (17,000 patches) of cPWT images to compare performance on the different sized datasets. A Receiver Operating Characteristic (ROC) study was implemented to investigate the ability of six veterinary medical professionals and pathologists to discern between generated and real cPWT patch images. The participants engaged in two separate rounds, where each round corresponded to models that had been trained on the two different sized datasets. The ROC study revealed mean average Area Under the Curve (AUC) values close to 0.5 for both rounds. The results from this study suggests that diffusion models can create histopathological patch images that are convincingly realistic where our participants often struggled to reliably differentiate between generated and real images. This underscores the potential of these models as a valuable tool for augmenting digital pathology datasets.
2024
Medical Imaging 2024: Digital and Computational Pathology
Stati uniti d'america - San Diego
Febbraio 2024
Progress in Biomedical Optics and Imaging - Proceedings of SPIE
SPIE-INT SOC OPTICAL ENGINEERING
12933
1
1
Artificial Intelligence; Deep Learning; Digital Pathology; Generative Models; Diffusion Models; Image Synthesis
Rai, Taranpreet; Gola, Cecilia; Hernández, Marta; Fingerhood, Sai; Marrero, Javier; Diaz Santana, Pablo; Giglia, Giuseppe; Morisi, Ambra; Bacci, Barba...espandi
File in questo prodotto:
File Dimensione Formato  
129331A.pdf

Accesso aperto

Dimensione 3.66 MB
Formato Adobe PDF
3.66 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2318/2029643
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact