Engineering the future of 3D pathology

Abstract

In recent years, technological advances in tissue preparation, high-throughput volumetric microscopy, and computational infrastructure have enabled rapid developments in nondestructive 3D pathology, in which high-resolution histologic datasets are obtained from thick tissue specimens, such as whole biopsies, without the need for physical sectioning onto glass slides. While 3D pathology generates massive datasets that are attractive for automated computational analysis, there is also a desire to use 3D pathology to improve the visual assessment of tissue histology. In this perspective, we discuss and provide examples of potential advantages of 3D pathology for the visual assessment of clinical specimens and the challenges of dealing with large 3D datasets (of individual or multiple specimens) that pathologists have not been trained to interpret. We discuss the need for artificial intelligence triaging algorithms and explainable analysis methods to assist pathologists or other domain experts in the interpretation of these novel, often complex, large datasets.

Keywords: digital pathology; light-sheet microscopy; nondestructive 3D pathology; prognosis; prostate cancer.

Figure 1

Examples of 3D pathology datasets with diagnostically significant spatial variations. These clinical specimens (FFPE) were deparaffinized, stained with a fluorescent analog of H&E, optically cleared to make them transparent, and imaged with OTLS microscopy. The volumetric fluorescence datasets were then false colored to mimic the appearance of standard H&E staining. (A) Volume rendering of a prostate core‐needle biopsy specimen. (B) A cropped view of one sagittal section from the prostate biopsy. (C) A zoomed‐in view of a volumetric block of data where a series of computationally generated 2D cross sections is shown at different depths. Here, a cancer gland appears poorly formed (Gleason pattern 4) at one depth but is revealed to be a tangential section of a fully formed gland (Gleason pattern 3) upon inspection of adjacent depths (blue arrows). (D) Volume rendering of a 3‐mm diameter punch biopsy of colorectal cancer. (E) A sagittal section is shown. (F) A zoomed‐in view of a volumetric block of data where a series of 2D cross sections is shown at different depths. Here, a few tumor cells appear to be clustered as an isolated ‘tumor bud’ (<4 cells) but are revealed to be connected to a larger mass of tumor cells upon inspection of adjacent depths (green arrows). 3D renderings created with the assistance of Imaris software.