Pathologists examine histochemically stained tissue biopsy sections to make medical diagnoses. Hematoxylin and eosin (H&E) is the most widely used histochemical stain in pathology, accounting for the vast majority of human tissue biopsy stains worldwide. In many clinical cases, however, additional “special stains” are required to bring contrast and color to various tissue components and allow pathologists to obtain a clearer diagnostic image. These specific stains frequently necessitate much more tissue preparation time, as well as tedious effort and monitoring by specialist histotechnologists, all of which raise the expenses and length of time to diagnose.
By computationally translating existing photos of H&E stained tissue into special stains, UCLA researchers devised a deep learning-based approach that can be used to eliminate the need for these specific stains to be prepared by human histotechnologists. This AI-based technique was demonstrated by creating a full panel of special stains for kidney tissue, including Periodic acid–Schiff (PAS), Jones silver stain, and Masson’s Trichrome, all of which were computationally transformed from existing images of H&E stained tissue biopsies using specialized deep neural networks. The researchers used this panel of special stains to conduct a clinical trial to demonstrate the efficacy of this stain-to-stain transformation process on various clinical samples representing a wide range of kidney disorders.
This study, conducted by a multi-institution team of board-certified renal pathologists, discovered that employing neural networks produced special stains and H&E photos improved diagnoses by a statistically significant margin when compared to using solely H&E images. A separate study found that the quality of the virtually re-stained photos is statistically equal to that of histochemically stained images by humans.
The stain-to-stain change is quick for a needle core tissue biopsy segment, taking less than one minute. This speed increases the quality of preliminary diagnostics that require specific stains while also saving time and money. In addition, because the virtual re-staining approach is used on existing stains, it is simple to implement.
A quick and precise diagnosis allows for prompt treatment, which can lead to better clinical outcomes. It does not necessitate any changes to pathology’s current tissue processing workflow. These benefits are especially valuable when it comes to recognising medical issues like transplant rejection.