Diagram of microscopic tissue slides with magnified views highlighting cellular structures on blue background.
Abstract pattern with red and yellow organic shapes on a blue background

Integrating Insights

Digital pathology sits at the heart of AI in healthcare. Bridging the gap between digital imaging data and its clinical interpretation - or ‘explainability’ - ensures AI tools remain accountable and trusted in medical practice. 

Illustration of histological tissue sections on glass slides
Abstract pink shape with colorful patterns, curves, and outlines.
Abstract medical illustration with anatomical structures and lymph nodes highlighted in various colors.

Enhancing Outcomes

  • Line drawing of an open coffee machine with visible internal components.

    Laboratory professionals are navigating the transition from traditional microscopy to digital whole slide image (WSI) analyis.

  • Illustration of a microscope examining a purple heart shape on a slide set on a stage against a light blue background.

    Digitization of patient samples is the foundation of explainability in AI.

  • Diagram of tissue layers in a cross-section, showing structure and details.

    The initial step in ensuring transparency and interpretability in AI-driven diagnostics begins with patient images and their data.

  • Convolutional Neural Network; CNN; Gleason Grade; Heat map; prostate cancer; prostate cancer staging;

    Driving these diagnostics are image analysis (IA) algorithms derived from deep learning (DL) networks.

    Sophisticated algorithms are crafted from an array of medical images and associated metadata - sourced from the laboratory.

  • heatmap

    A shared language bridges medical expertise and technological innovation.

    Development must involve collaborative inputs from both pathologists and engineers.

  • Microscopic image of tissue showing immunohistochemistry staining with brown circular patterns on a blue background

    Bridging multiple domains

    Computer vision, operations expertise, and data science are just some. Beyond traditional training, interdisciplinary collaboration is required to effectively manage and interpret vast datasets, develop AI, and ensure the responsible implementation of technology in clinical practice.

  • heat map, WSI, whole slide image

    Multimodal data is infused in technology-enabled clinical and research workflows throughout the globe.

    Clinicians and computer scientists are sharing roles to visualize and materialize solutions that advance patient care.

  • Digital pathology data is central to AI development, but whole slide images (WSI) pose significant storage and operational challenges. Data compression, cloud storage, and retrieval systems are just some of many important elements essential to this operation.

  • Illustration of a balance scale with a stack of bricks on one side and a purple whale on the other, both appearing to weigh equally.

    Average weight of a humpback whale = approximate volume of traditional Hollerith “punch” cards needed to store 1gb of WSI data

  • Illustration of a balance scale with cells on one side and a space station on the other, suggesting a comparison or relationship between cell research and space exploration.

    2x the weight of the International Space Station (ISS) = the average volume of glass slides handled yearly at major academic medical centers, large regional hospitals, and specialized reference laboratories serving multiple healthcare facilities

  • Illustration of a balance scale with a brain on one side and a laptop displaying film icons on the other.

    Data requirements for storing cumulative (new and archival) WSI data at a high-volume, fully digitized academic medical center (in PB) = approximately 20x the total storage capacity of typical major streaming service providers (Netflix, Amazon Prime Video, Hulu, etc.)

  • Flowchart of a medical image processing system showing archival and clinical scanning. It depicts the transition of DICOM and SVS formats through on-premise and virtual private cloud, with metadata integration and storage options like DICOMstore, image viewers, and metadata analysis tools. Involves processes like 40X layering, base layering, pyramid layering, and LIS data utilization.

    Successful laboratory digitization requires close collaboration between pathologists and IT specialists. By aligning technological infrastructure with clinical workflows, we ensure efficient, sustainable systems that enhance diagnostic capabilities.

Diagram illustrating the process of digitizing histology slides. Shows a person operating a scanner, a workflow of inserting a glass slide into a WSI scanner, optics and detection, and software processing. Includes visual elements of slide imaging and data analysis on a computer, with micrograph examples and layered images.

Pathologists are stewards of laboratory data, establishing diagnostic ground truth for AI model validation. With 70% of medical decisions driven by laboratory diagnostics, pathologists' expertise forms the foundation of explainable AI. While current AI solutions focus on imaging, the future lies in integrating broader laboratory data.

Microscopic image of a breast core biopsy slide showing tissue samples stained pink and purple, accompanied by metadata on the right side.
microscopic slide showing tissue sample with magnification data
Microscopic image of colon tissue slide with annotations and menu options.
Close-up of a Philips medical device screen showing a slide rack interface. A notification displays an error message "Slide skipped: No tissue found." The screen includes options for rack priority and slide info.
Microscopic slides with tissue samples labeled Block A1, A2, A3, and A4. A notification indicates a rescan is requested for Block A3.