Background: Intra-tumor heterogeneity for PD-L1 expression in non-small cell lung cancer (NSCLC) might lead to inaccurate stratification of patients to immunotherapy. The purpose of this research was to quantitate the effect of different factors on the risk of inaccurate diagnosis of PD-L1 expression. Methods: MATLAB software was used to model tumor with a different fraction, distribution and clustering of PD-L1 protein expression and their effect on false positive and negative diagnosis in subsets of the modeled tumor (representing biopsies). Additionally, we evaluated the agreement between PD-L1 status in random segments and whole slides of PD-L1 stained clinical NSCLC cases. Results: Our computer-based model showed a significant increase in error rate when the fraction of PD-L1 positive cells was closer to the cutoff value (error rate of 33.33 %, 0.45 % and 0.74 % for PD-L1 positivity in 40−60%, ≤20 % and ≥80 % of tumor cells, respectively, P < 0.0001). In addition, biopsy size showed negative correlation with error rate (P < 0.0001) and larger clusters of PD-L1 positive cells were associated with higher error rate (P < 0.0001). Analysis of the clinical samples supported those of the computer-based model with higher error rate in cases with positive PD-L1 expression closer to the cutoff value. Based on our computerized model and clinical analysis, we developed a model to predict error rate based on biopsy size and the fraction of PD-L1 positive cells in the biopsy. Conclusion: Analysis of small biopsies for PD-L1 expression might be associated with significant error rate. The model presented can be used to identify cases with increased risk for error in whom interpretation of the test results should be made with caution.
Histopathologic diagnosis of Hirschsprung's disease (HSCR) is time consuming and requires expertise. The use of artificial intelligence (AI) in digital pathology is actively researched and may improve the diagnosis of HSCR. The purpose of this research was to develop an algorithm capable of identifying ganglion cells in digital pathology slides and implement it as an assisting tool for the pathologist in the diagnosis of HSCR. Ninety five digital pathology slides were used for the construction and training of the algorithm. Fifty cases suspected for HSCR (727 slides) were used as a validation cohort. Image sets suspected to contain ganglion cells were chosen by the algorithm and then reviewed and scored by five pathologists, one HSCR expert and 4 non-experts. The algorithm was able to identify ganglion cells with 96% sensitivity and 99% specificity (in normal colon) as well as to correctly identify a case previously misdiagnosed as non-HSCR. The expert was able to achieve perfectly accurate diagnoses based solely on the images suggested by the algorithm, with over 95% time saved. Non-experts would require expert consultation in 20–58% of the cases to achieve similar results. The use of AI in the diagnosis of HSCR can greatly reduce the time and effort required for diagnosis and improve accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.