Abstract
Microscopic review of tissue sections is of foundational importance in pathology, yet the traditional chemistry-based histology laboratory methods are labour intensive, tissue destructive, poorly scalable to the evolving needs of precision medicine and cause delays in patient diagnosis and treatment. Recent AI-based techniques offer promise in upending histology workflow; one such method developed by PictorLabs can generate near-instantaneous diagnostic images via a machine learning algorithm. Here, we demonstrate the utility of virtual staining in a blinded, wash-out controlled study of 16 cases of lymph node excisional biopsies, including a spectrum of diagnoses from reactive to lymphoma and compare the diagnostic performance of virtual and chemical H&Es across a range of stain quality, image quality, morphometric assessment and diagnostic interpretation parameters as well as proposed follow-up immunostains. Our results show non-inferior performance of virtual H&E stains across all parameters, including an improved stain quality pass rate (92% vs 79% for virtual vs chemical stains, respectively) and an equivalent rate of binary diagnostic concordance (90% vs 92%). More detailed adjudicated reviews of differential diagnoses and proposed IHC panels showed no major discordances. Virtual H&Es appear fit for purpose and non-inferior to chemical H&Es in diagnostic assessment of clinical lymph node samples, in a limited pilot study.
Overview
- The study aims to evaluate the diagnostic performance of virtual H&E stains compared to chemical H&Es in lymph node excisional biopsies. The study uses a machine learning algorithm developed by PictorLabs to generate near-instantaneous diagnostic images. The study includes 16 cases of lymph node excisional biopsies with a spectrum of diagnoses from reactive to lymphoma. The study compares the diagnostic performance of virtual and chemical H&Es across a range of stain quality, image quality, morphometric assessment, and diagnostic interpretation parameters. The study's primary objective is to determine if virtual H&E stains are non-inferior to chemical H&Es in diagnostic assessment of clinical lymph node samples. The study is a blinded, wash-out controlled study.
Comparative Analysis & Findings
- The study shows non-inferior performance of virtual H&E stains across all parameters, including an improved stain quality pass rate (92% vs 79% for virtual vs chemical stains, respectively) and an equivalent rate of binary diagnostic concordance (90% vs 92%). More detailed adjudicated reviews of differential diagnoses and proposed IHC panels showed no major discordances. The study suggests that virtual H&Es are fit for purpose and non-inferior to chemical H&Es in diagnostic assessment of clinical lymph node samples, in a limited pilot study.
Implications and Future Directions
- The study's findings suggest that virtual H&E stains are a promising alternative to traditional chemistry-based histology laboratory methods. The study's non-inferior performance of virtual H&E stains compared to chemical H&Es in diagnostic assessment of clinical lymph node samples highlights the potential of AI-based techniques to upend histology workflow. The study's limitations include the small sample size and the need for further validation in larger studies. Future research directions could include validation of virtual H&E stains in larger studies, exploration of virtual H&E stains in other tissue types, and investigation of the potential of virtual H&E stains in precision medicine.