How Pictor Labs is turning tissue into data to revolutionize diagnosis

Histopathology, the practice of examining tissue under a microscope to diagnose disease, forms the foundation of many critical medical decisions. It is through this method that clinicians confirm cancer, assess disease progression, and guide treatment plans. However, the core technology behind histopathology has remained largely unchanged for over a century. When a biopsy is sent to the lab for testing, it is sliced into paper-thin sections that are dipped into a sequence of chemical dyes. These dyes "stain" the tissue samples to reveal hidden cellular structures, enabling doctors to determine factors such as cancer type or disease progression.

Despite its importance, the staining process can be extremely slow and resource-intensive. Labs often batch samples together, meaning it can take days to get results to doctors, and even longer for those results to move through reporting systems, consultations, and scheduling before patients can be appropriately treated. Furthermore, conventional staining workflows consume and alter tissue sections, leaving less material available for subsequent crucial steps like molecular diagnostics, genetic sequencing, and other precision medicine processes.

Yair Rivenson, cofounder and CEO of digital pathology startup Pictor Labs, witnessed this inefficiency firsthand while working as a researcher at UCLA in 2019. Alongside cofounder Aydogan Ozcan, Rivenson and Ozcan worked with tissue samples as part of their deep learning microscopy research. The reliance on limited physical tissue samples in an era of digital and AI advancements struck Rivenson as peculiar. "For me," he says, "the question was immediately, why can't we teach a computer to do that? Why do we need the chemistry part?"

Pictor Labs was founded to address this question. The company aims to simulate the staining process, turning cell stains into software. "The information is there," explains Rivenson. "We can basically image the tissue, edit it as it is in its native form, and compute that contrast so that the pathologist can study the same exact tissue in the exact same way that they're currently doing."

By leveraging advanced imaging technologies and machine learning algorithms, Pictor Labs has developed a system that can replicate the staining process without the need for chemical dyes. This innovation not only speeds up the diagnostic process but also preserves the integrity of the tissue samples, ensuring that more material is available for further analysis.

The startup's technology works by capturing high-resolution images of the tissue samples and using these images to train machine learning models. These models learn to identify the cellular structures and patterns that are typically revealed through chemical staining. Once trained, the models can apply similar contrast enhancements to the raw images, making it possible for pathologists to analyze the tissue as they traditionally would with stained samples.

This approach not only eliminates the time-consuming and resource-intensive staining process but also reduces variability in the diagnostic process. Chemical staining can sometimes produce inconsistent results due to variations in dye application or tissue preparation. By relying on software-based staining, Pictor Labs aims to provide more consistent and accurate diagnostic outcomes.

Moreover, the preservation of tissue integrity is a significant advantage of Pictor Labs' technology. Traditional staining methods can degrade the tissue samples, making it difficult to perform additional tests or analyses. With Pictor Labs' approach, pathologists can analyze the tissue samples without altering their structure, allowing for a wider range of diagnostic options and more comprehensive patient care.

The potential impact of Pictor Labs' innovation extends beyond the efficiency of the diagnostic process. By enabling pathologists to analyze tissue samples in a more accurate and consistent manner, the company's technology could lead to earlier and more accurate diagnoses, improved treatment plans, and ultimately better patient outcomes.

As Pictor Labs continues to develop its technology and collaborate with healthcare providers, the company is poised to revolutionize the field of histopathology. By turning tissue into data and leveraging the power of AI, Pictor Labs is set to transform the way diseases are diagnosed and treated, ushering in a new era of precision medicine.