Automated Classification of Acute Rejection from Endomyocardial Biopsies

Giuste, Felipe; Venkatesan, Mythreye; Zhao, Changxuan; Li, Tong; Zhu, Yuanda; Deshpande, Shriprasad; Wang, May D.

doi:10.1145/3388440.3412430

Cited by 6 publications

(2 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our classifier also outperforms other studies which date back to the year 2017, when Tong et al constructed a shallow neural network based on handcrafted features derived from 43 WSIs (Children's Healthcare of Atlanta cohort). This dataset has been used several times afterwards improving the performance of the cross validated model while adopting newer methodology but remains limited due to the very small dataset size [14,[29][30][31].…”

Section: Discussionmentioning

confidence: 99%

Prediction of heart transplant rejection from routine pathology slides with self-supervised Deep Learning

Seraphin

Luedde

Roderburg

et al. 2022

Preprint

View full text Add to dashboard Cite

One of the most important complications of heart transplantation is organ rejection, which is diagnosed on endomyocardial biopsies by pathologists. Computer-based systems could assist in the diagnostic process and potentially improve reproducibility. Here, we evaluated the feasibility of using deep learning in predicting the degree of cellular rejection from pathology slides as defined by the International Society for Heart and Lung Transplantation (ISHLT) grading system. We collected 1079 slides from 325 patients from three transplant centers in Germany. We trained an attention-based deep neural network to predict rejection in the primary cohort and evaluated its performance using cross validation and by deploying it to three cohorts. For binary prediction (rejection yes/no) the mean Area Under the Receiver Operating Curve (AUROC) was 0.849 in the cross-validated experiment and 0.734, 0.729 and 0.716 in external validation cohorts. For a prediction of the ISHLT grade (0R, 1R, 2/3R), AUROCs were 0.835, 0.633 and 0.905 in the cross-validated experiment and 0.764, 0.597, 0.913, and 0.631, 0.633, 0.682, and 0.722, 0.601, 0.805 in the validation cohorts, respectively. The predictions of the AI model were interpretable by human experts and highlighted plausible morphological patterns. We conclude that artificial intelligence can detect patterns of cellular transplant rejection in routine pathology, even when trained on small patient cohorts.

show abstract

Section: Discussionmentioning

confidence: 99%

Prediction of heart transplant rejection from routine pathology slides with self-supervised Deep Learning

Seraphin

Luedde

Roderburg

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Incorporate clinical insights [105] Small sample size Data imputation [32], [59], [114] Interactive user interface [42], [90], [94], [105] Bad data quality Artifact correction [64], [115] Clinical feedback [90], [106] Imbalanced classes Data augmentation [62], [65] Visualization of feature importance [33], [71], [80] Complex disease phenotype Multi-modality data [57], [116] Clustering analysis [67], [87] Data heterogeneity Data normalization [59] Decision tree [4], [32], [33] Lack of expert annotation Weakly supervised learning [51], [71], [87] Use multiple feature importance approaches [13], [41], [74] Unkown sources of signal Key feature extraction [37], [58], [69] Cross validation when comparing models [46], [105], [107], [110] Explanations unclear Pre-processing changes [67], [68] Use appropriate and robust performance metrics (AUROC, MCC) [110], [111] Data leakage Patient-level split [45]…”

Section: Suggestions Problems Solutionsmentioning

confidence: 99%

Explainable Artificial Intelligence Methods in Combating Pandemics: A Systematic Review

Giuste

Shi

Zhu

et al. 2023

IEEE Rev. Biomed. Eng.

Self Cite

View full text Add to dashboard Cite

Despite the myriad peer-reviewed papers demonstrating novel Artificial Intelligence (AI)-based solutions to COVID-19 challenges during the pandemic, few have made a significant clinical impact, especially in diagnosis and disease precision staging. One major cause for such low impact is the lack of model transparency, significantly limiting the AI adoption in real clinical practice. To solve this problem, AI models need to be explained to users. Thus, we have conducted a comprehensive study of Explainable Artificial Intelligence (XAI) using PRISMA technology. Our findings suggest that XAI can improve model performance, instill trust in the users, and assist users in decisionmaking. In this systematic review, we introduce common XAI techniques and their utility with specific examples of their application. We discuss the evaluation of XAI results because it is an important step for maximizing the value of AI-based clinical decision support systems. Additionally, we present the traditional, modern, and advanced XAI models to demonstrate the evolution of novel techniques. Finally, we provide a best practice guideline that developers can refer to during the model experimentation. We also offer potential solutions with specific examples for common challenges in AI model experimentation. This comprehensive review, hopefully, can promote AI adoption in biomedicine and healthcare.

show abstract