Artificial Intelligence and Cellular Segmentation in Tissue Microscopy Images

Durkee, Madeleine S.; Abraham, Rebecca; Clark, Marcus R.; Giger, Maryellen L.

doi:10.1016/j.ajpath.2021.05.022

Cited by 46 publications

(32 citation statements)

References 50 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods have been developed and validated in LN image data sets TII and identification of prognostic features has been historically impeded by the complexities of analyzing immunofluorescence data from chronically inflamed kidneys, including tissue autofluorescence due to scarring, antibody cross-reactivity, and patient heterogeneity. Artificial intelligence algorithms have led to significant progress in automated detection and analysis of cells in confocal images (19)(20)(21)(22). These approaches have been used suc-…”

Section: Introductionmentioning

confidence: 99%

Specific in situ inflammatory states associate with progression to renal failure in lupus nephritis

Abraham¹,

Durkee²,

Ai³

et al. 2022

Journal of Clinical Investigation

Self Cite

View full text Add to dashboard Cite

BACKGROUND. In human lupus nephritis (LN), tubulointerstitial inflammation (TII) on biopsy predicts progression to endstage renal disease (ESRD). However, only about half of patients with moderate-to-severe TII develop ESRD. We hypothesized that this heterogeneity in outcome reflects different underlying inflammatory states. Therefore, we interrogated renal biopsies from LN longitudinal and cross-sectional cohorts.METHODS. Data were acquired using conventional and highly multiplexed confocal microscopy. To accurately segment cells across whole biopsies, and to understand their spatial relationships, we developed computational pipelines by training and implementing several deep-learning models and other computer vision techniques.RESULTS. High B cell densities were associated with protection from ESRD. In contrast, high densities of CD8 + , γδ, and other CD4 -CD8 -T cells were associated with both acute renal failure and progression to ESRD. B cells were often organized into large periglomerular neighborhoods with Tfh cells, while CD4 -T cells formed small neighborhoods in the tubulointerstitium, with frequency that predicted progression to ESRD.CONCLUSION. These data reveal that specific in situ inflammatory states are associated with refractory and progressive renal disease.

show abstract

Section: Introductionmentioning

confidence: 99%

Specific in situ inflammatory states associate with progression to renal failure in lupus nephritis

Abraham¹,

Durkee²,

Ai³

et al. 2022

Journal of Clinical Investigation

Self Cite

View full text Add to dashboard Cite

show abstract

“…3 The evolution of AI in pathology is illuminated by earlier but analogous trends in radiology. 4 Analogies with the use of AI in reduction of interobserver variability are considered, 5 as are current strategies for reducing the need for massive annotation in machine learning through either the use of existing supervised frameworks or by exploiting hybrid models using unsupervised learning, generative models, and/or synthetic data. 6 Because of the rapid emergence of generative algorithms, a separate Review is included on generative deep learning in the pathology environment.…”

Section: Q11mentioning

confidence: 99%

Artificial Intelligence in Pathology

Cohen

Levenson

Pantanowitz

2021

The American Journal of Pathology

View full text Add to dashboard Cite

“…The ability to accurately locate, classify, and segment cells within microscopy images is fundamental for bioimaging applications in both basic and translational biomedical research, as it enables the study of cell spatial relationships and morphologies as they pertain to physiology, drug response, and disease ( 1 , 2 ). Automating these otherwise time-consuming and expert-dependent tasks has been an active area of research and development ( 3 , 4 ), particularly since the advent of convolutional neural networks (CNNs), which have attained substantially higher accuracies than previous “hand-crafted,” feature-engineered methods ( 5 ). Fluorescence tissue imaging presents a number of challenges that can reduce the accuracy of CNNs in cell segmentation and classification, including auto-fluorescence, dense clustering, and variation in cell size, shape, and density among different cell types ( 6 ).…”

Section: Introductionmentioning

confidence: 99%

“…Fluorescence tissue imaging presents a number of challenges that can reduce the accuracy of CNNs in cell segmentation and classification, including auto-fluorescence, dense clustering, and variation in cell size, shape, and density among different cell types ( 6 ). Much of the effort in overcoming these challenges has been focused on improving CNN architectures or creating large datasets for pretraining CNNs so that they attain high accuracies with limited training examples ( 4 , 6–8 ). However, improvements may also be made by leveraging advances in microscope technology for use in combination with CNNs or other deep-learning architectures.…”

Section: Introductionmentioning

confidence: 99%

Fluorescence lifetime image microscopy prediction with convolutional neural networks for cell detection and classification in tissues

Smolen

Wooley

2022

PNAS Nexus

View full text Add to dashboard Cite

Convolutional neural networks (CNNs) and other deep-learning models have proven to be transformative tools for the automated analysis of microscopy images, particularly in the domain of cellular and tissue imaging. These computer-vision models have primarily been applied with traditional microscopy image modalities (e.g. brightfield and fluorescence), likely due to the availability of large datasets in these regimes. However, more advanced microscopy imaging techniques could, potentially, allow for improved model performance in various computational histopathology tasks. In this work, we demonstrate that CNNs can achieve high accuracy in cell detection and classification without large amounts of data when applied to histology images acquired by FLIM. This accuracy is higher than what would be achieved with regular single or dual-channel fluorescence images under the same settings, particularly for CNNs pretrained on publicly available fluorescent cell or general photographic datasets. Additionally, generated FLIM images could be predicted from just the fluorescence image data by using a dense U-Net CNN model trained on a subset of ground-truth FLIM images. These U-Net CNN generated FLIM images demonstrated high similarity to ground truth and improved accuracy in cell detection and classification over fluorescence alone when used as input to a variety of commonly used CNNs. This improved accuracy was maintained even when the FLIM images were generated by a U-Net CNN trained on only a few example FLIM images.

show abstract

Artificial Intelligence and Cellular Segmentation in Tissue Microscopy Images

Cited by 46 publications

References 50 publications

Specific in situ inflammatory states associate with progression to renal failure in lupus nephritis

Specific in situ inflammatory states associate with progression to renal failure in lupus nephritis

Artificial Intelligence in Pathology

Fluorescence lifetime image microscopy prediction with convolutional neural networks for cell detection and classification in tissues

Contact Info

Product

Resources

About